diff options
| author | rekby <rekby@ydb.tech> | 2023-12-14 16:56:50 +0300 |
|---|---|---|
| committer | rekby <rekby@ydb.tech> | 2023-12-14 18:09:44 +0300 |
| commit | b2b2bb5997507072ca64548efe64447dd6395426 (patch) | |
| tree | bbfbf77d11f1972c93ae4101fe561fd440d6ad6a | |
| parent | 8b8678a6a4f57c62e348cdad8afd3849011a5f11 (diff) | |
| download | ydb-b2b2bb5997507072ca64548efe64447dd6395426.tar.gz | |
KIKIMR-19900 switch arcadia to python ydb sdk from contrib
Этот PR создан скриптом - для переключения зависимостей на python ydb sdk с версии внутри ydb на код, приезжающий через контриб.
Код в обеих версиях одинаковый, так что поломок/изменения функционала на ожидается.
На всякий случай посмотрите свои проекты и если будут возражения пишите сюда в issues или в тикет KIKIMR-19900.
Если всё ок - шипните, для определённости.
При отсутствии блокеров PR будет перегенерирован и влит с force-мёрджем в четверг, 14 декабря.
297 files changed, 61369 insertions, 204 deletions
diff --git a/contrib/python/aiohttp/.dist-info/METADATA b/contrib/python/aiohttp/.dist-info/METADATA new file mode 100644 index 0000000000..c2f6befe9e --- /dev/null +++ b/contrib/python/aiohttp/.dist-info/METADATA @@ -0,0 +1,255 @@ +Metadata-Version: 2.1 +Name: aiohttp +Version: 3.8.1 +Summary: Async http client/server framework (asyncio) +Home-page: https://github.com/aio-libs/aiohttp +Maintainer: aiohttp team <team@aiohttp.org> +Maintainer-email: team@aiohttp.org +License: Apache 2 +Project-URL: Chat: Gitter, https://gitter.im/aio-libs/Lobby +Project-URL: CI: GitHub Actions, https://github.com/aio-libs/aiohttp/actions?query=workflow%3ACI +Project-URL: Coverage: codecov, https://codecov.io/github/aio-libs/aiohttp +Project-URL: Docs: Changelog, https://docs.aiohttp.org/en/stable/changes.html +Project-URL: Docs: RTD, https://docs.aiohttp.org +Project-URL: GitHub: issues, https://github.com/aio-libs/aiohttp/issues +Project-URL: GitHub: repo, https://github.com/aio-libs/aiohttp +Platform: UNKNOWN +Classifier: Development Status :: 5 - Production/Stable +Classifier: Framework :: AsyncIO +Classifier: Intended Audience :: Developers +Classifier: License :: OSI Approved :: Apache Software License +Classifier: Operating System :: POSIX +Classifier: Operating System :: MacOS :: MacOS X +Classifier: Operating System :: Microsoft :: Windows +Classifier: Programming Language :: Python +Classifier: Programming Language :: Python :: 3 +Classifier: Programming Language :: Python :: 3.6 +Classifier: Programming Language :: Python :: 3.7 +Classifier: Programming Language :: Python :: 3.8 +Classifier: Programming Language :: Python :: 3.9 +Classifier: Programming Language :: Python :: 3.10 +Classifier: Topic :: Internet :: WWW/HTTP +Requires-Python: >=3.6 +Description-Content-Type: text/x-rst +License-File: LICENSE.txt +Requires-Dist: attrs (>=17.3.0) +Requires-Dist: charset-normalizer (<3.0,>=2.0) +Requires-Dist: multidict (<7.0,>=4.5) +Requires-Dist: async-timeout (<5.0,>=4.0.0a3) +Requires-Dist: yarl (<2.0,>=1.0) +Requires-Dist: frozenlist (>=1.1.1) +Requires-Dist: aiosignal (>=1.1.2) +Requires-Dist: idna-ssl (>=1.0) ; python_version < "3.7" +Requires-Dist: asynctest (==0.13.0) ; python_version < "3.8" +Requires-Dist: typing-extensions (>=3.7.4) ; python_version < "3.8" +Provides-Extra: speedups +Requires-Dist: aiodns ; extra == 'speedups' +Requires-Dist: Brotli ; extra == 'speedups' +Requires-Dist: cchardet ; extra == 'speedups' + +================================== +Async http client/server framework +================================== + +.. image:: https://raw.githubusercontent.com/aio-libs/aiohttp/master/docs/aiohttp-plain.svg + :height: 64px + :width: 64px + :alt: aiohttp logo + +| + +.. image:: https://github.com/aio-libs/aiohttp/workflows/CI/badge.svg + :target: https://github.com/aio-libs/aiohttp/actions?query=workflow%3ACI + :alt: GitHub Actions status for master branch + +.. image:: https://codecov.io/gh/aio-libs/aiohttp/branch/master/graph/badge.svg + :target: https://codecov.io/gh/aio-libs/aiohttp + :alt: codecov.io status for master branch + +.. image:: https://badge.fury.io/py/aiohttp.svg + :target: https://pypi.org/project/aiohttp + :alt: Latest PyPI package version + +.. image:: https://readthedocs.org/projects/aiohttp/badge/?version=latest + :target: https://docs.aiohttp.org/ + :alt: Latest Read The Docs + +.. image:: https://img.shields.io/discourse/status?server=https%3A%2F%2Faio-libs.discourse.group + :target: https://aio-libs.discourse.group + :alt: Discourse status + +.. image:: https://badges.gitter.im/Join%20Chat.svg + :target: https://gitter.im/aio-libs/Lobby + :alt: Chat on Gitter + + +Key Features +============ + +- Supports both client and server side of HTTP protocol. +- Supports both client and server Web-Sockets out-of-the-box and avoids + Callback Hell. +- Provides Web-server with middlewares and plugable routing. + + +Getting started +=============== + +Client +------ + +To get something from the web: + +.. code-block:: python + + import aiohttp + import asyncio + + async def main(): + + async with aiohttp.ClientSession() as session: + async with session.get('http://python.org') as response: + + print("Status:", response.status) + print("Content-type:", response.headers['content-type']) + + html = await response.text() + print("Body:", html[:15], "...") + + loop = asyncio.get_event_loop() + loop.run_until_complete(main()) + +This prints: + +.. code-block:: + + Status: 200 + Content-type: text/html; charset=utf-8 + Body: <!doctype html> ... + +Coming from `requests <https://requests.readthedocs.io/>`_ ? Read `why we need so many lines <https://aiohttp.readthedocs.io/en/latest/http_request_lifecycle.html>`_. + +Server +------ + +An example using a simple server: + +.. code-block:: python + + # examples/server_simple.py + from aiohttp import web + + async def handle(request): + name = request.match_info.get('name', "Anonymous") + text = "Hello, " + name + return web.Response(text=text) + + async def wshandle(request): + ws = web.WebSocketResponse() + await ws.prepare(request) + + async for msg in ws: + if msg.type == web.WSMsgType.text: + await ws.send_str("Hello, {}".format(msg.data)) + elif msg.type == web.WSMsgType.binary: + await ws.send_bytes(msg.data) + elif msg.type == web.WSMsgType.close: + break + + return ws + + + app = web.Application() + app.add_routes([web.get('/', handle), + web.get('/echo', wshandle), + web.get('/{name}', handle)]) + + if __name__ == '__main__': + web.run_app(app) + + +Documentation +============= + +https://aiohttp.readthedocs.io/ + + +Demos +===== + +https://github.com/aio-libs/aiohttp-demos + + +External links +============== + +* `Third party libraries + <http://aiohttp.readthedocs.io/en/latest/third_party.html>`_ +* `Built with aiohttp + <http://aiohttp.readthedocs.io/en/latest/built_with.html>`_ +* `Powered by aiohttp + <http://aiohttp.readthedocs.io/en/latest/powered_by.html>`_ + +Feel free to make a Pull Request for adding your link to these pages! + + +Communication channels +====================== + +*aio-libs discourse group*: https://aio-libs.discourse.group + +*gitter chat* https://gitter.im/aio-libs/Lobby + +We support `Stack Overflow +<https://stackoverflow.com/questions/tagged/aiohttp>`_. +Please add *aiohttp* tag to your question there. + +Requirements +============ + +- Python >= 3.6 +- async-timeout_ +- attrs_ +- charset-normalizer_ +- multidict_ +- yarl_ + +Optionally you may install the cChardet_ and aiodns_ libraries (highly +recommended for sake of speed). + +.. _charset-normalizer: https://pypi.org/project/charset-normalizer +.. _aiodns: https://pypi.python.org/pypi/aiodns +.. _attrs: https://github.com/python-attrs/attrs +.. _multidict: https://pypi.python.org/pypi/multidict +.. _yarl: https://pypi.python.org/pypi/yarl +.. _async-timeout: https://pypi.python.org/pypi/async_timeout +.. _cChardet: https://pypi.python.org/pypi/cchardet + +License +======= + +``aiohttp`` is offered under the Apache 2 license. + + +Keepsafe +======== + +The aiohttp community would like to thank Keepsafe +(https://www.getkeepsafe.com) for its support in the early days of +the project. + + +Source code +=========== + +The latest developer version is available in a GitHub repository: +https://github.com/aio-libs/aiohttp + +Benchmarks +========== + +If you are interested in efficiency, the AsyncIO community maintains a +list of benchmarks on the official wiki: +https://github.com/python/asyncio/wiki/Benchmarks + + diff --git a/contrib/python/aiohttp/.dist-info/top_level.txt b/contrib/python/aiohttp/.dist-info/top_level.txt new file mode 100644 index 0000000000..ee4ba4f3d7 --- /dev/null +++ b/contrib/python/aiohttp/.dist-info/top_level.txt @@ -0,0 +1 @@ +aiohttp diff --git a/contrib/python/aiohttp/LICENSE.txt b/contrib/python/aiohttp/LICENSE.txt new file mode 100644 index 0000000000..054102f2db --- /dev/null +++ b/contrib/python/aiohttp/LICENSE.txt @@ -0,0 +1,13 @@ + Copyright 2013-2020 aio-libs collaboration. + + Licensed under the Apache License, Version 2.0 (the "License"); + you may not use this file except in compliance with the License. + You may obtain a copy of the License at + + http://www.apache.org/licenses/LICENSE-2.0 + + Unless required by applicable law or agreed to in writing, software + distributed under the License is distributed on an "AS IS" BASIS, + WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + See the License for the specific language governing permissions and + limitations under the License. diff --git a/contrib/python/aiohttp/README.rst b/contrib/python/aiohttp/README.rst new file mode 100644 index 0000000000..d057acbe2f --- /dev/null +++ b/contrib/python/aiohttp/README.rst @@ -0,0 +1,204 @@ +================================== +Async http client/server framework +================================== + +.. image:: https://raw.githubusercontent.com/aio-libs/aiohttp/master/docs/aiohttp-plain.svg + :height: 64px + :width: 64px + :alt: aiohttp logo + +| + +.. image:: https://github.com/aio-libs/aiohttp/workflows/CI/badge.svg + :target: https://github.com/aio-libs/aiohttp/actions?query=workflow%3ACI + :alt: GitHub Actions status for master branch + +.. image:: https://codecov.io/gh/aio-libs/aiohttp/branch/master/graph/badge.svg + :target: https://codecov.io/gh/aio-libs/aiohttp + :alt: codecov.io status for master branch + +.. image:: https://badge.fury.io/py/aiohttp.svg + :target: https://pypi.org/project/aiohttp + :alt: Latest PyPI package version + +.. image:: https://readthedocs.org/projects/aiohttp/badge/?version=latest + :target: https://docs.aiohttp.org/ + :alt: Latest Read The Docs + +.. image:: https://img.shields.io/discourse/status?server=https%3A%2F%2Faio-libs.discourse.group + :target: https://aio-libs.discourse.group + :alt: Discourse status + +.. image:: https://badges.gitter.im/Join%20Chat.svg + :target: https://gitter.im/aio-libs/Lobby + :alt: Chat on Gitter + + +Key Features +============ + +- Supports both client and server side of HTTP protocol. +- Supports both client and server Web-Sockets out-of-the-box and avoids + Callback Hell. +- Provides Web-server with middlewares and plugable routing. + + +Getting started +=============== + +Client +------ + +To get something from the web: + +.. code-block:: python + + import aiohttp + import asyncio + + async def main(): + + async with aiohttp.ClientSession() as session: + async with session.get('http://python.org') as response: + + print("Status:", response.status) + print("Content-type:", response.headers['content-type']) + + html = await response.text() + print("Body:", html[:15], "...") + + loop = asyncio.get_event_loop() + loop.run_until_complete(main()) + +This prints: + +.. code-block:: + + Status: 200 + Content-type: text/html; charset=utf-8 + Body: <!doctype html> ... + +Coming from `requests <https://requests.readthedocs.io/>`_ ? Read `why we need so many lines <https://aiohttp.readthedocs.io/en/latest/http_request_lifecycle.html>`_. + +Server +------ + +An example using a simple server: + +.. code-block:: python + + # examples/server_simple.py + from aiohttp import web + + async def handle(request): + name = request.match_info.get('name', "Anonymous") + text = "Hello, " + name + return web.Response(text=text) + + async def wshandle(request): + ws = web.WebSocketResponse() + await ws.prepare(request) + + async for msg in ws: + if msg.type == web.WSMsgType.text: + await ws.send_str("Hello, {}".format(msg.data)) + elif msg.type == web.WSMsgType.binary: + await ws.send_bytes(msg.data) + elif msg.type == web.WSMsgType.close: + break + + return ws + + + app = web.Application() + app.add_routes([web.get('/', handle), + web.get('/echo', wshandle), + web.get('/{name}', handle)]) + + if __name__ == '__main__': + web.run_app(app) + + +Documentation +============= + +https://aiohttp.readthedocs.io/ + + +Demos +===== + +https://github.com/aio-libs/aiohttp-demos + + +External links +============== + +* `Third party libraries + <http://aiohttp.readthedocs.io/en/latest/third_party.html>`_ +* `Built with aiohttp + <http://aiohttp.readthedocs.io/en/latest/built_with.html>`_ +* `Powered by aiohttp + <http://aiohttp.readthedocs.io/en/latest/powered_by.html>`_ + +Feel free to make a Pull Request for adding your link to these pages! + + +Communication channels +====================== + +*aio-libs discourse group*: https://aio-libs.discourse.group + +*gitter chat* https://gitter.im/aio-libs/Lobby + +We support `Stack Overflow +<https://stackoverflow.com/questions/tagged/aiohttp>`_. +Please add *aiohttp* tag to your question there. + +Requirements +============ + +- Python >= 3.6 +- async-timeout_ +- attrs_ +- charset-normalizer_ +- multidict_ +- yarl_ + +Optionally you may install the cChardet_ and aiodns_ libraries (highly +recommended for sake of speed). + +.. _charset-normalizer: https://pypi.org/project/charset-normalizer +.. _aiodns: https://pypi.python.org/pypi/aiodns +.. _attrs: https://github.com/python-attrs/attrs +.. _multidict: https://pypi.python.org/pypi/multidict +.. _yarl: https://pypi.python.org/pypi/yarl +.. _async-timeout: https://pypi.python.org/pypi/async_timeout +.. _cChardet: https://pypi.python.org/pypi/cchardet + +License +======= + +``aiohttp`` is offered under the Apache 2 license. + + +Keepsafe +======== + +The aiohttp community would like to thank Keepsafe +(https://www.getkeepsafe.com) for its support in the early days of +the project. + + +Source code +=========== + +The latest developer version is available in a GitHub repository: +https://github.com/aio-libs/aiohttp + +Benchmarks +========== + +If you are interested in efficiency, the AsyncIO community maintains a +list of benchmarks on the official wiki: +https://github.com/python/asyncio/wiki/Benchmarks diff --git a/contrib/python/aiohttp/aiohttp/__init__.py b/contrib/python/aiohttp/aiohttp/__init__.py new file mode 100644 index 0000000000..4bbcef2935 --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/__init__.py @@ -0,0 +1,216 @@ +__version__ = "3.8.1" + +from typing import Tuple + +from . import hdrs as hdrs +from .client import ( + BaseConnector as BaseConnector, + ClientConnectionError as ClientConnectionError, + ClientConnectorCertificateError as ClientConnectorCertificateError, + ClientConnectorError as ClientConnectorError, + ClientConnectorSSLError as ClientConnectorSSLError, + ClientError as ClientError, + ClientHttpProxyError as ClientHttpProxyError, + ClientOSError as ClientOSError, + ClientPayloadError as ClientPayloadError, + ClientProxyConnectionError as ClientProxyConnectionError, + ClientRequest as ClientRequest, + ClientResponse as ClientResponse, + ClientResponseError as ClientResponseError, + ClientSession as ClientSession, + ClientSSLError as ClientSSLError, + ClientTimeout as ClientTimeout, + ClientWebSocketResponse as ClientWebSocketResponse, + ContentTypeError as ContentTypeError, + Fingerprint as Fingerprint, + InvalidURL as InvalidURL, + NamedPipeConnector as NamedPipeConnector, + RequestInfo as RequestInfo, + ServerConnectionError as ServerConnectionError, + ServerDisconnectedError as ServerDisconnectedError, + ServerFingerprintMismatch as ServerFingerprintMismatch, + ServerTimeoutError as ServerTimeoutError, + TCPConnector as TCPConnector, + TooManyRedirects as TooManyRedirects, + UnixConnector as UnixConnector, + WSServerHandshakeError as WSServerHandshakeError, + request as request, +) +from .cookiejar import CookieJar as CookieJar, DummyCookieJar as DummyCookieJar +from .formdata import FormData as FormData +from .helpers import BasicAuth, ChainMapProxy, ETag +from .http import ( + HttpVersion as HttpVersion, + HttpVersion10 as HttpVersion10, + HttpVersion11 as HttpVersion11, + WebSocketError as WebSocketError, + WSCloseCode as WSCloseCode, + WSMessage as WSMessage, + WSMsgType as WSMsgType, +) +from .multipart import ( + BadContentDispositionHeader as BadContentDispositionHeader, + BadContentDispositionParam as BadContentDispositionParam, + BodyPartReader as BodyPartReader, + MultipartReader as MultipartReader, + MultipartWriter as MultipartWriter, + content_disposition_filename as content_disposition_filename, + parse_content_disposition as parse_content_disposition, +) +from .payload import ( + PAYLOAD_REGISTRY as PAYLOAD_REGISTRY, + AsyncIterablePayload as AsyncIterablePayload, + BufferedReaderPayload as BufferedReaderPayload, + BytesIOPayload as BytesIOPayload, + BytesPayload as BytesPayload, + IOBasePayload as IOBasePayload, + JsonPayload as JsonPayload, + Payload as Payload, + StringIOPayload as StringIOPayload, + StringPayload as StringPayload, + TextIOPayload as TextIOPayload, + get_payload as get_payload, + payload_type as payload_type, +) +from .payload_streamer import streamer as streamer +from .resolver import ( + AsyncResolver as AsyncResolver, + DefaultResolver as DefaultResolver, + ThreadedResolver as ThreadedResolver, +) +from .streams import ( + EMPTY_PAYLOAD as EMPTY_PAYLOAD, + DataQueue as DataQueue, + EofStream as EofStream, + FlowControlDataQueue as FlowControlDataQueue, + StreamReader as StreamReader, +) +from .tracing import ( + TraceConfig as TraceConfig, + TraceConnectionCreateEndParams as TraceConnectionCreateEndParams, + TraceConnectionCreateStartParams as TraceConnectionCreateStartParams, + TraceConnectionQueuedEndParams as TraceConnectionQueuedEndParams, + TraceConnectionQueuedStartParams as TraceConnectionQueuedStartParams, + TraceConnectionReuseconnParams as TraceConnectionReuseconnParams, + TraceDnsCacheHitParams as TraceDnsCacheHitParams, + TraceDnsCacheMissParams as TraceDnsCacheMissParams, + TraceDnsResolveHostEndParams as TraceDnsResolveHostEndParams, + TraceDnsResolveHostStartParams as TraceDnsResolveHostStartParams, + TraceRequestChunkSentParams as TraceRequestChunkSentParams, + TraceRequestEndParams as TraceRequestEndParams, + TraceRequestExceptionParams as TraceRequestExceptionParams, + TraceRequestRedirectParams as TraceRequestRedirectParams, + TraceRequestStartParams as TraceRequestStartParams, + TraceResponseChunkReceivedParams as TraceResponseChunkReceivedParams, +) + +__all__: Tuple[str, ...] = ( + "hdrs", + # client + "BaseConnector", + "ClientConnectionError", + "ClientConnectorCertificateError", + "ClientConnectorError", + "ClientConnectorSSLError", + "ClientError", + "ClientHttpProxyError", + "ClientOSError", + "ClientPayloadError", + "ClientProxyConnectionError", + "ClientResponse", + "ClientRequest", + "ClientResponseError", + "ClientSSLError", + "ClientSession", + "ClientTimeout", + "ClientWebSocketResponse", + "ContentTypeError", + "Fingerprint", + "InvalidURL", + "RequestInfo", + "ServerConnectionError", + "ServerDisconnectedError", + "ServerFingerprintMismatch", + "ServerTimeoutError", + "TCPConnector", + "TooManyRedirects", + "UnixConnector", + "NamedPipeConnector", + "WSServerHandshakeError", + "request", + # cookiejar + "CookieJar", + "DummyCookieJar", + # formdata + "FormData", + # helpers + "BasicAuth", + "ChainMapProxy", + "ETag", + # http + "HttpVersion", + "HttpVersion10", + "HttpVersion11", + "WSMsgType", + "WSCloseCode", + "WSMessage", + "WebSocketError", + # multipart + "BadContentDispositionHeader", + "BadContentDispositionParam", + "BodyPartReader", + "MultipartReader", + "MultipartWriter", + "content_disposition_filename", + "parse_content_disposition", + # payload + "AsyncIterablePayload", + "BufferedReaderPayload", + "BytesIOPayload", + "BytesPayload", + "IOBasePayload", + "JsonPayload", + "PAYLOAD_REGISTRY", + "Payload", + "StringIOPayload", + "StringPayload", + "TextIOPayload", + "get_payload", + "payload_type", + # payload_streamer + "streamer", + # resolver + "AsyncResolver", + "DefaultResolver", + "ThreadedResolver", + # streams + "DataQueue", + "EMPTY_PAYLOAD", + "EofStream", + "FlowControlDataQueue", + "StreamReader", + # tracing + "TraceConfig", + "TraceConnectionCreateEndParams", + "TraceConnectionCreateStartParams", + "TraceConnectionQueuedEndParams", + "TraceConnectionQueuedStartParams", + "TraceConnectionReuseconnParams", + "TraceDnsCacheHitParams", + "TraceDnsCacheMissParams", + "TraceDnsResolveHostEndParams", + "TraceDnsResolveHostStartParams", + "TraceRequestChunkSentParams", + "TraceRequestEndParams", + "TraceRequestExceptionParams", + "TraceRequestRedirectParams", + "TraceRequestStartParams", + "TraceResponseChunkReceivedParams", +) + +try: + from .worker import GunicornUVLoopWebWorker, GunicornWebWorker + + __all__ += ("GunicornWebWorker", "GunicornUVLoopWebWorker") +except ImportError: # pragma: no cover + pass diff --git a/contrib/python/aiohttp/aiohttp/_cparser.pxd b/contrib/python/aiohttp/aiohttp/_cparser.pxd new file mode 100644 index 0000000000..49055d6a56 --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/_cparser.pxd @@ -0,0 +1,190 @@ +from libc.stdint cimport ( + int8_t, + int16_t, + int32_t, + int64_t, + uint8_t, + uint16_t, + uint32_t, + uint64_t, +) + + +cdef extern from "llhttp.h": + + struct llhttp__internal_s: + int32_t _index + void* _span_pos0 + void* _span_cb0 + int32_t error + const char* reason + const char* error_pos + void* data + void* _current + uint64_t content_length + uint8_t type + uint8_t method + uint8_t http_major + uint8_t http_minor + uint8_t header_state + uint8_t lenient_flags + uint8_t upgrade + uint8_t finish + uint16_t flags + uint16_t status_code + void* settings + + ctypedef llhttp__internal_s llhttp__internal_t + ctypedef llhttp__internal_t llhttp_t + + ctypedef int (*llhttp_data_cb)(llhttp_t*, const char *at, size_t length) except -1 + ctypedef int (*llhttp_cb)(llhttp_t*) except -1 + + struct llhttp_settings_s: + llhttp_cb on_message_begin + llhttp_data_cb on_url + llhttp_data_cb on_status + llhttp_data_cb on_header_field + llhttp_data_cb on_header_value + llhttp_cb on_headers_complete + llhttp_data_cb on_body + llhttp_cb on_message_complete + llhttp_cb on_chunk_header + llhttp_cb on_chunk_complete + + llhttp_cb on_url_complete + llhttp_cb on_status_complete + llhttp_cb on_header_field_complete + llhttp_cb on_header_value_complete + + ctypedef llhttp_settings_s llhttp_settings_t + + enum llhttp_errno: + HPE_OK, + HPE_INTERNAL, + HPE_STRICT, + HPE_LF_EXPECTED, + HPE_UNEXPECTED_CONTENT_LENGTH, + HPE_CLOSED_CONNECTION, + HPE_INVALID_METHOD, + HPE_INVALID_URL, + HPE_INVALID_CONSTANT, + HPE_INVALID_VERSION, + HPE_INVALID_HEADER_TOKEN, + HPE_INVALID_CONTENT_LENGTH, + HPE_INVALID_CHUNK_SIZE, + HPE_INVALID_STATUS, + HPE_INVALID_EOF_STATE, + HPE_INVALID_TRANSFER_ENCODING, + HPE_CB_MESSAGE_BEGIN, + HPE_CB_HEADERS_COMPLETE, + HPE_CB_MESSAGE_COMPLETE, + HPE_CB_CHUNK_HEADER, + HPE_CB_CHUNK_COMPLETE, + HPE_PAUSED, + HPE_PAUSED_UPGRADE, + HPE_USER + + ctypedef llhttp_errno llhttp_errno_t + + enum llhttp_flags: + F_CONNECTION_KEEP_ALIVE, + F_CONNECTION_CLOSE, + F_CONNECTION_UPGRADE, + F_CHUNKED, + F_UPGRADE, + F_CONTENT_LENGTH, + F_SKIPBODY, + F_TRAILING, + F_TRANSFER_ENCODING + + enum llhttp_lenient_flags: + LENIENT_HEADERS, + LENIENT_CHUNKED_LENGTH + + enum llhttp_type: + HTTP_REQUEST, + HTTP_RESPONSE, + HTTP_BOTH + + enum llhttp_finish_t: + HTTP_FINISH_SAFE, + HTTP_FINISH_SAFE_WITH_CB, + HTTP_FINISH_UNSAFE + + enum llhttp_method: + HTTP_DELETE, + HTTP_GET, + HTTP_HEAD, + HTTP_POST, + HTTP_PUT, + HTTP_CONNECT, + HTTP_OPTIONS, + HTTP_TRACE, + HTTP_COPY, + HTTP_LOCK, + HTTP_MKCOL, + HTTP_MOVE, + HTTP_PROPFIND, + HTTP_PROPPATCH, + HTTP_SEARCH, + HTTP_UNLOCK, + HTTP_BIND, + HTTP_REBIND, + HTTP_UNBIND, + HTTP_ACL, + HTTP_REPORT, + HTTP_MKACTIVITY, + HTTP_CHECKOUT, + HTTP_MERGE, + HTTP_MSEARCH, + HTTP_NOTIFY, + HTTP_SUBSCRIBE, + HTTP_UNSUBSCRIBE, + HTTP_PATCH, + HTTP_PURGE, + HTTP_MKCALENDAR, + HTTP_LINK, + HTTP_UNLINK, + HTTP_SOURCE, + HTTP_PRI, + HTTP_DESCRIBE, + HTTP_ANNOUNCE, + HTTP_SETUP, + HTTP_PLAY, + HTTP_PAUSE, + HTTP_TEARDOWN, + HTTP_GET_PARAMETER, + HTTP_SET_PARAMETER, + HTTP_REDIRECT, + HTTP_RECORD, + HTTP_FLUSH + + ctypedef llhttp_method llhttp_method_t; + + void llhttp_settings_init(llhttp_settings_t* settings) + void llhttp_init(llhttp_t* parser, llhttp_type type, + const llhttp_settings_t* settings) + + llhttp_errno_t llhttp_execute(llhttp_t* parser, const char* data, size_t len) + llhttp_errno_t llhttp_finish(llhttp_t* parser) + + int llhttp_message_needs_eof(const llhttp_t* parser) + + int llhttp_should_keep_alive(const llhttp_t* parser) + + void llhttp_pause(llhttp_t* parser) + void llhttp_resume(llhttp_t* parser) + + void llhttp_resume_after_upgrade(llhttp_t* parser) + + llhttp_errno_t llhttp_get_errno(const llhttp_t* parser) + const char* llhttp_get_error_reason(const llhttp_t* parser) + void llhttp_set_error_reason(llhttp_t* parser, const char* reason) + const char* llhttp_get_error_pos(const llhttp_t* parser) + const char* llhttp_errno_name(llhttp_errno_t err) + + const char* llhttp_method_name(llhttp_method_t method) + + void llhttp_set_lenient_headers(llhttp_t* parser, int enabled) + void llhttp_set_lenient_chunked_length(llhttp_t* parser, int enabled) diff --git a/contrib/python/aiohttp/aiohttp/_find_header.c b/contrib/python/aiohttp/aiohttp/_find_header.c new file mode 100644 index 0000000000..012cba33ac --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/_find_header.c @@ -0,0 +1,9870 @@ +/* The file is autogenerated from aiohttp/hdrs.py +Run ./tools/gen.py to update it after the origin changing. */ + +#include "_find_header.h" + +#define NEXT_CHAR() \ +{ \ + count++; \ + if (count == size) { \ + /* end of search */ \ + return -1; \ + } \ + pchar++; \ + ch = *pchar; \ + last = (count == size -1); \ +} while(0); + +int +find_header(const char *str, int size) +{ + char *pchar = str; + int last; + char ch; + int count = -1; + pchar--; + + + NEXT_CHAR(); + switch (ch) { + case 'A': + if (last) { + return -1; + } + goto A; + case 'a': + if (last) { + return -1; + } + goto A; + case 'C': + if (last) { + return -1; + } + goto C; + case 'c': + if (last) { + return -1; + } + goto C; + case 'D': + if (last) { + return -1; + } + goto D; + case 'd': + if (last) { + return -1; + } + goto D; + case 'E': + if (last) { + return -1; + } + goto E; + case 'e': + if (last) { + return -1; + } + goto E; + case 'F': + if (last) { + return -1; + } + goto F; + case 'f': + if (last) { + return -1; + } + goto F; + case 'H': + if (last) { + return -1; + } + goto H; + case 'h': + if (last) { + return -1; + } + goto H; + case 'I': + if (last) { + return -1; + } + goto I; + case 'i': + if (last) { + return -1; + } + goto I; + case 'K': + if (last) { + return -1; + } + goto K; + case 'k': + if (last) { + return -1; + } + goto K; + case 'L': + if (last) { + return -1; + } + goto L; + case 'l': + if (last) { + return -1; + } + goto L; + case 'M': + if (last) { + return -1; + } + goto M; + case 'm': + if (last) { + return -1; + } + goto M; + case 'O': + if (last) { + return -1; + } + goto O; + case 'o': + if (last) { + return -1; + } + goto O; + case 'P': + if (last) { + return -1; + } + goto P; + case 'p': + if (last) { + return -1; + } + goto P; + case 'R': + if (last) { + return -1; + } + goto R; + case 'r': + if (last) { + return -1; + } + goto R; + case 'S': + if (last) { + return -1; + } + goto S; + case 's': + if (last) { + return -1; + } + goto S; + case 'T': + if (last) { + return -1; + } + goto T; + case 't': + if (last) { + return -1; + } + goto T; + case 'U': + if (last) { + return -1; + } + goto U; + case 'u': + if (last) { + return -1; + } + goto U; + case 'V': + if (last) { + return -1; + } + goto V; + case 'v': + if (last) { + return -1; + } + goto V; + case 'W': + if (last) { + return -1; + } + goto W; + case 'w': + if (last) { + return -1; + } + goto W; + case 'X': + if (last) { + return -1; + } + goto X; + case 'x': + if (last) { + return -1; + } + goto X; + default: + return -1; + } + +A: + NEXT_CHAR(); + switch (ch) { + case 'C': + if (last) { + return -1; + } + goto AC; + case 'c': + if (last) { + return -1; + } + goto AC; + case 'G': + if (last) { + return -1; + } + goto AG; + case 'g': + if (last) { + return -1; + } + goto AG; + case 'L': + if (last) { + return -1; + } + goto AL; + case 'l': + if (last) { + return -1; + } + goto AL; + case 'U': + if (last) { + return -1; + } + goto AU; + case 'u': + if (last) { + return -1; + } + goto AU; + default: + return -1; + } + +AC: + NEXT_CHAR(); + switch (ch) { + case 'C': + if (last) { + return -1; + } + goto ACC; + case 'c': + if (last) { + return -1; + } + goto ACC; + default: + return -1; + } + +ACC: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto ACCE; + case 'e': + if (last) { + return -1; + } + goto ACCE; + default: + return -1; + } + +ACCE: + NEXT_CHAR(); + switch (ch) { + case 'P': + if (last) { + return -1; + } + goto ACCEP; + case 'p': + if (last) { + return -1; + } + goto ACCEP; + case 'S': + if (last) { + return -1; + } + goto ACCES; + case 's': + if (last) { + return -1; + } + goto ACCES; + default: + return -1; + } + +ACCEP: + NEXT_CHAR(); + switch (ch) { + case 'T': + if (last) { + return 0; + } + goto ACCEPT; + case 't': + if (last) { + return 0; + } + goto ACCEPT; + default: + return -1; + } + +ACCEPT: + NEXT_CHAR(); + switch (ch) { + case '-': + if (last) { + return -1; + } + goto ACCEPT_; + default: + return -1; + } + +ACCEPT_: + NEXT_CHAR(); + switch (ch) { + case 'C': + if (last) { + return -1; + } + goto ACCEPT_C; + case 'c': + if (last) { + return -1; + } + goto ACCEPT_C; + case 'E': + if (last) { + return -1; + } + goto ACCEPT_E; + case 'e': + if (last) { + return -1; + } + goto ACCEPT_E; + case 'L': + if (last) { + return -1; + } + goto ACCEPT_L; + case 'l': + if (last) { + return -1; + } + goto ACCEPT_L; + case 'R': + if (last) { + return -1; + } + goto ACCEPT_R; + case 'r': + if (last) { + return -1; + } + goto ACCEPT_R; + default: + return -1; + } + +ACCEPT_C: + NEXT_CHAR(); + switch (ch) { + case 'H': + if (last) { + return -1; + } + goto ACCEPT_CH; + case 'h': + if (last) { + return -1; + } + goto ACCEPT_CH; + default: + return -1; + } + +ACCEPT_CH: + NEXT_CHAR(); + switch (ch) { + case 'A': + if (last) { + return -1; + } + goto ACCEPT_CHA; + case 'a': + if (last) { + return -1; + } + goto ACCEPT_CHA; + default: + return -1; + } + +ACCEPT_CHA: + NEXT_CHAR(); + switch (ch) { + case 'R': + if (last) { + return -1; + } + goto ACCEPT_CHAR; + case 'r': + if (last) { + return -1; + } + goto ACCEPT_CHAR; + default: + return -1; + } + +ACCEPT_CHAR: + NEXT_CHAR(); + switch (ch) { + case 'S': + if (last) { + return -1; + } + goto ACCEPT_CHARS; + case 's': + if (last) { + return -1; + } + goto ACCEPT_CHARS; + default: + return -1; + } + +ACCEPT_CHARS: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto ACCEPT_CHARSE; + case 'e': + if (last) { + return -1; + } + goto ACCEPT_CHARSE; + default: + return -1; + } + +ACCEPT_CHARSE: + NEXT_CHAR(); + switch (ch) { + case 'T': + if (last) { + return 1; + } + goto ACCEPT_CHARSET; + case 't': + if (last) { + return 1; + } + goto ACCEPT_CHARSET; + default: + return -1; + } + +ACCEPT_E: + NEXT_CHAR(); + switch (ch) { + case 'N': + if (last) { + return -1; + } + goto ACCEPT_EN; + case 'n': + if (last) { + return -1; + } + goto ACCEPT_EN; + default: + return -1; + } + +ACCEPT_EN: + NEXT_CHAR(); + switch (ch) { + case 'C': + if (last) { + return -1; + } + goto ACCEPT_ENC; + case 'c': + if (last) { + return -1; + } + goto ACCEPT_ENC; + default: + return -1; + } + +ACCEPT_ENC: + NEXT_CHAR(); + switch (ch) { + case 'O': + if (last) { + return -1; + } + goto ACCEPT_ENCO; + case 'o': + if (last) { + return -1; + } + goto ACCEPT_ENCO; + default: + return -1; + } + +ACCEPT_ENCO: + NEXT_CHAR(); + switch (ch) { + case 'D': + if (last) { + return -1; + } + goto ACCEPT_ENCOD; + case 'd': + if (last) { + return -1; + } + goto ACCEPT_ENCOD; + default: + return -1; + } + +ACCEPT_ENCOD: + NEXT_CHAR(); + switch (ch) { + case 'I': + if (last) { + return -1; + } + goto ACCEPT_ENCODI; + case 'i': + if (last) { + return -1; + } + goto ACCEPT_ENCODI; + default: + return -1; + } + +ACCEPT_ENCODI: + NEXT_CHAR(); + switch (ch) { + case 'N': + if (last) { + return -1; + } + goto ACCEPT_ENCODIN; + case 'n': + if (last) { + return -1; + } + goto ACCEPT_ENCODIN; + default: + return -1; + } + +ACCEPT_ENCODIN: + NEXT_CHAR(); + switch (ch) { + case 'G': + if (last) { + return 2; + } + goto ACCEPT_ENCODING; + case 'g': + if (last) { + return 2; + } + goto ACCEPT_ENCODING; + default: + return -1; + } + +ACCEPT_L: + NEXT_CHAR(); + switch (ch) { + case 'A': + if (last) { + return -1; + } + goto ACCEPT_LA; + case 'a': + if (last) { + return -1; + } + goto ACCEPT_LA; + default: + return -1; + } + +ACCEPT_LA: + NEXT_CHAR(); + switch (ch) { + case 'N': + if (last) { + return -1; + } + goto ACCEPT_LAN; + case 'n': + if (last) { + return -1; + } + goto ACCEPT_LAN; + default: + return -1; + } + +ACCEPT_LAN: + NEXT_CHAR(); + switch (ch) { + case 'G': + if (last) { + return -1; + } + goto ACCEPT_LANG; + case 'g': + if (last) { + return -1; + } + goto ACCEPT_LANG; + default: + return -1; + } + +ACCEPT_LANG: + NEXT_CHAR(); + switch (ch) { + case 'U': + if (last) { + return -1; + } + goto ACCEPT_LANGU; + case 'u': + if (last) { + return -1; + } + goto ACCEPT_LANGU; + default: + return -1; + } + +ACCEPT_LANGU: + NEXT_CHAR(); + switch (ch) { + case 'A': + if (last) { + return -1; + } + goto ACCEPT_LANGUA; + case 'a': + if (last) { + return -1; + } + goto ACCEPT_LANGUA; + default: + return -1; + } + +ACCEPT_LANGUA: + NEXT_CHAR(); + switch (ch) { + case 'G': + if (last) { + return -1; + } + goto ACCEPT_LANGUAG; + case 'g': + if (last) { + return -1; + } + goto ACCEPT_LANGUAG; + default: + return -1; + } + +ACCEPT_LANGUAG: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return 3; + } + goto ACCEPT_LANGUAGE; + case 'e': + if (last) { + return 3; + } + goto ACCEPT_LANGUAGE; + default: + return -1; + } + +ACCEPT_R: + NEXT_CHAR(); + switch (ch) { + case 'A': + if (last) { + return -1; + } + goto ACCEPT_RA; + case 'a': + if (last) { + return -1; + } + goto ACCEPT_RA; + default: + return -1; + } + +ACCEPT_RA: + NEXT_CHAR(); + switch (ch) { + case 'N': + if (last) { + return -1; + } + goto ACCEPT_RAN; + case 'n': + if (last) { + return -1; + } + goto ACCEPT_RAN; + default: + return -1; + } + +ACCEPT_RAN: + NEXT_CHAR(); + switch (ch) { + case 'G': + if (last) { + return -1; + } + goto ACCEPT_RANG; + case 'g': + if (last) { + return -1; + } + goto ACCEPT_RANG; + default: + return -1; + } + +ACCEPT_RANG: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto ACCEPT_RANGE; + case 'e': + if (last) { + return -1; + } + goto ACCEPT_RANGE; + default: + return -1; + } + +ACCEPT_RANGE: + NEXT_CHAR(); + switch (ch) { + case 'S': + if (last) { + return 4; + } + goto ACCEPT_RANGES; + case 's': + if (last) { + return 4; + } + goto ACCEPT_RANGES; + default: + return -1; + } + +ACCES: + NEXT_CHAR(); + switch (ch) { + case 'S': + if (last) { + return -1; + } + goto ACCESS; + case 's': + if (last) { + return -1; + } + goto ACCESS; + default: + return -1; + } + +ACCESS: + NEXT_CHAR(); + switch (ch) { + case '-': + if (last) { + return -1; + } + goto ACCESS_; + default: + return -1; + } + +ACCESS_: + NEXT_CHAR(); + switch (ch) { + case 'C': + if (last) { + return -1; + } + goto ACCESS_C; + case 'c': + if (last) { + return -1; + } + goto ACCESS_C; + default: + return -1; + } + +ACCESS_C: + NEXT_CHAR(); + switch (ch) { + case 'O': + if (last) { + return -1; + } + goto ACCESS_CO; + case 'o': + if (last) { + return -1; + } + goto ACCESS_CO; + default: + return -1; + } + +ACCESS_CO: + NEXT_CHAR(); + switch (ch) { + case 'N': + if (last) { + return -1; + } + goto ACCESS_CON; + case 'n': + if (last) { + return -1; + } + goto ACCESS_CON; + default: + return -1; + } + +ACCESS_CON: + NEXT_CHAR(); + switch (ch) { + case 'T': + if (last) { + return -1; + } + goto ACCESS_CONT; + case 't': + if (last) { + return -1; + } + goto ACCESS_CONT; + default: + return -1; + } + +ACCESS_CONT: + NEXT_CHAR(); + switch (ch) { + case 'R': + if (last) { + return -1; + } + goto ACCESS_CONTR; + case 'r': + if (last) { + return -1; + } + goto ACCESS_CONTR; + default: + return -1; + } + +ACCESS_CONTR: + NEXT_CHAR(); + switch (ch) { + case 'O': + if (last) { + return -1; + } + goto ACCESS_CONTRO; + case 'o': + if (last) { + return -1; + } + goto ACCESS_CONTRO; + default: + return -1; + } + +ACCESS_CONTRO: + NEXT_CHAR(); + switch (ch) { + case 'L': + if (last) { + return -1; + } + goto ACCESS_CONTROL; + case 'l': + if (last) { + return -1; + } + goto ACCESS_CONTROL; + default: + return -1; + } + +ACCESS_CONTROL: + NEXT_CHAR(); + switch (ch) { + case '-': + if (last) { + return -1; + } + goto ACCESS_CONTROL_; + default: + return -1; + } + +ACCESS_CONTROL_: + NEXT_CHAR(); + switch (ch) { + case 'A': + if (last) { + return -1; + } + goto ACCESS_CONTROL_A; + case 'a': + if (last) { + return -1; + } + goto ACCESS_CONTROL_A; + case 'E': + if (last) { + return -1; + } + goto ACCESS_CONTROL_E; + case 'e': + if (last) { + return -1; + } + goto ACCESS_CONTROL_E; + case 'M': + if (last) { + return -1; + } + goto ACCESS_CONTROL_M; + case 'm': + if (last) { + return -1; + } + goto ACCESS_CONTROL_M; + case 'R': + if (last) { + return -1; + } + goto ACCESS_CONTROL_R; + case 'r': + if (last) { + return -1; + } + goto ACCESS_CONTROL_R; + default: + return -1; + } + +ACCESS_CONTROL_A: + NEXT_CHAR(); + switch (ch) { + case 'L': + if (last) { + return -1; + } + goto ACCESS_CONTROL_AL; + case 'l': + if (last) { + return -1; + } + goto ACCESS_CONTROL_AL; + default: + return -1; + } + +ACCESS_CONTROL_AL: + NEXT_CHAR(); + switch (ch) { + case 'L': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALL; + case 'l': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALL; + default: + return -1; + } + +ACCESS_CONTROL_ALL: + NEXT_CHAR(); + switch (ch) { + case 'O': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLO; + case 'o': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLO; + default: + return -1; + } + +ACCESS_CONTROL_ALLO: + NEXT_CHAR(); + switch (ch) { + case 'W': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW; + case 'w': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW; + default: + return -1; + } + +ACCESS_CONTROL_ALLOW: + NEXT_CHAR(); + switch (ch) { + case '-': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_; + default: + return -1; + } + +ACCESS_CONTROL_ALLOW_: + NEXT_CHAR(); + switch (ch) { + case 'C': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_C; + case 'c': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_C; + case 'H': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_H; + case 'h': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_H; + case 'M': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_M; + case 'm': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_M; + case 'O': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_O; + case 'o': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_O; + default: + return -1; + } + +ACCESS_CONTROL_ALLOW_C: + NEXT_CHAR(); + switch (ch) { + case 'R': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_CR; + case 'r': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_CR; + default: + return -1; + } + +ACCESS_CONTROL_ALLOW_CR: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_CRE; + case 'e': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_CRE; + default: + return -1; + } + +ACCESS_CONTROL_ALLOW_CRE: + NEXT_CHAR(); + switch (ch) { + case 'D': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_CRED; + case 'd': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_CRED; + default: + return -1; + } + +ACCESS_CONTROL_ALLOW_CRED: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_CREDE; + case 'e': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_CREDE; + default: + return -1; + } + +ACCESS_CONTROL_ALLOW_CREDE: + NEXT_CHAR(); + switch (ch) { + case 'N': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_CREDEN; + case 'n': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_CREDEN; + default: + return -1; + } + +ACCESS_CONTROL_ALLOW_CREDEN: + NEXT_CHAR(); + switch (ch) { + case 'T': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_CREDENT; + case 't': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_CREDENT; + default: + return -1; + } + +ACCESS_CONTROL_ALLOW_CREDENT: + NEXT_CHAR(); + switch (ch) { + case 'I': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_CREDENTI; + case 'i': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_CREDENTI; + default: + return -1; + } + +ACCESS_CONTROL_ALLOW_CREDENTI: + NEXT_CHAR(); + switch (ch) { + case 'A': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_CREDENTIA; + case 'a': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_CREDENTIA; + default: + return -1; + } + +ACCESS_CONTROL_ALLOW_CREDENTIA: + NEXT_CHAR(); + switch (ch) { + case 'L': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_CREDENTIAL; + case 'l': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_CREDENTIAL; + default: + return -1; + } + +ACCESS_CONTROL_ALLOW_CREDENTIAL: + NEXT_CHAR(); + switch (ch) { + case 'S': + if (last) { + return 5; + } + goto ACCESS_CONTROL_ALLOW_CREDENTIALS; + case 's': + if (last) { + return 5; + } + goto ACCESS_CONTROL_ALLOW_CREDENTIALS; + default: + return -1; + } + +ACCESS_CONTROL_ALLOW_H: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_HE; + case 'e': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_HE; + default: + return -1; + } + +ACCESS_CONTROL_ALLOW_HE: + NEXT_CHAR(); + switch (ch) { + case 'A': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_HEA; + case 'a': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_HEA; + default: + return -1; + } + +ACCESS_CONTROL_ALLOW_HEA: + NEXT_CHAR(); + switch (ch) { + case 'D': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_HEAD; + case 'd': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_HEAD; + default: + return -1; + } + +ACCESS_CONTROL_ALLOW_HEAD: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_HEADE; + case 'e': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_HEADE; + default: + return -1; + } + +ACCESS_CONTROL_ALLOW_HEADE: + NEXT_CHAR(); + switch (ch) { + case 'R': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_HEADER; + case 'r': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_HEADER; + default: + return -1; + } + +ACCESS_CONTROL_ALLOW_HEADER: + NEXT_CHAR(); + switch (ch) { + case 'S': + if (last) { + return 6; + } + goto ACCESS_CONTROL_ALLOW_HEADERS; + case 's': + if (last) { + return 6; + } + goto ACCESS_CONTROL_ALLOW_HEADERS; + default: + return -1; + } + +ACCESS_CONTROL_ALLOW_M: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_ME; + case 'e': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_ME; + default: + return -1; + } + +ACCESS_CONTROL_ALLOW_ME: + NEXT_CHAR(); + switch (ch) { + case 'T': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_MET; + case 't': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_MET; + default: + return -1; + } + +ACCESS_CONTROL_ALLOW_MET: + NEXT_CHAR(); + switch (ch) { + case 'H': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_METH; + case 'h': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_METH; + default: + return -1; + } + +ACCESS_CONTROL_ALLOW_METH: + NEXT_CHAR(); + switch (ch) { + case 'O': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_METHO; + case 'o': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_METHO; + default: + return -1; + } + +ACCESS_CONTROL_ALLOW_METHO: + NEXT_CHAR(); + switch (ch) { + case 'D': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_METHOD; + case 'd': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_METHOD; + default: + return -1; + } + +ACCESS_CONTROL_ALLOW_METHOD: + NEXT_CHAR(); + switch (ch) { + case 'S': + if (last) { + return 7; + } + goto ACCESS_CONTROL_ALLOW_METHODS; + case 's': + if (last) { + return 7; + } + goto ACCESS_CONTROL_ALLOW_METHODS; + default: + return -1; + } + +ACCESS_CONTROL_ALLOW_O: + NEXT_CHAR(); + switch (ch) { + case 'R': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_OR; + case 'r': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_OR; + default: + return -1; + } + +ACCESS_CONTROL_ALLOW_OR: + NEXT_CHAR(); + switch (ch) { + case 'I': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_ORI; + case 'i': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_ORI; + default: + return -1; + } + +ACCESS_CONTROL_ALLOW_ORI: + NEXT_CHAR(); + switch (ch) { + case 'G': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_ORIG; + case 'g': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_ORIG; + default: + return -1; + } + +ACCESS_CONTROL_ALLOW_ORIG: + NEXT_CHAR(); + switch (ch) { + case 'I': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_ORIGI; + case 'i': + if (last) { + return -1; + } + goto ACCESS_CONTROL_ALLOW_ORIGI; + default: + return -1; + } + +ACCESS_CONTROL_ALLOW_ORIGI: + NEXT_CHAR(); + switch (ch) { + case 'N': + if (last) { + return 8; + } + goto ACCESS_CONTROL_ALLOW_ORIGIN; + case 'n': + if (last) { + return 8; + } + goto ACCESS_CONTROL_ALLOW_ORIGIN; + default: + return -1; + } + +ACCESS_CONTROL_E: + NEXT_CHAR(); + switch (ch) { + case 'X': + if (last) { + return -1; + } + goto ACCESS_CONTROL_EX; + case 'x': + if (last) { + return -1; + } + goto ACCESS_CONTROL_EX; + default: + return -1; + } + +ACCESS_CONTROL_EX: + NEXT_CHAR(); + switch (ch) { + case 'P': + if (last) { + return -1; + } + goto ACCESS_CONTROL_EXP; + case 'p': + if (last) { + return -1; + } + goto ACCESS_CONTROL_EXP; + default: + return -1; + } + +ACCESS_CONTROL_EXP: + NEXT_CHAR(); + switch (ch) { + case 'O': + if (last) { + return -1; + } + goto ACCESS_CONTROL_EXPO; + case 'o': + if (last) { + return -1; + } + goto ACCESS_CONTROL_EXPO; + default: + return -1; + } + +ACCESS_CONTROL_EXPO: + NEXT_CHAR(); + switch (ch) { + case 'S': + if (last) { + return -1; + } + goto ACCESS_CONTROL_EXPOS; + case 's': + if (last) { + return -1; + } + goto ACCESS_CONTROL_EXPOS; + default: + return -1; + } + +ACCESS_CONTROL_EXPOS: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto ACCESS_CONTROL_EXPOSE; + case 'e': + if (last) { + return -1; + } + goto ACCESS_CONTROL_EXPOSE; + default: + return -1; + } + +ACCESS_CONTROL_EXPOSE: + NEXT_CHAR(); + switch (ch) { + case '-': + if (last) { + return -1; + } + goto ACCESS_CONTROL_EXPOSE_; + default: + return -1; + } + +ACCESS_CONTROL_EXPOSE_: + NEXT_CHAR(); + switch (ch) { + case 'H': + if (last) { + return -1; + } + goto ACCESS_CONTROL_EXPOSE_H; + case 'h': + if (last) { + return -1; + } + goto ACCESS_CONTROL_EXPOSE_H; + default: + return -1; + } + +ACCESS_CONTROL_EXPOSE_H: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto ACCESS_CONTROL_EXPOSE_HE; + case 'e': + if (last) { + return -1; + } + goto ACCESS_CONTROL_EXPOSE_HE; + default: + return -1; + } + +ACCESS_CONTROL_EXPOSE_HE: + NEXT_CHAR(); + switch (ch) { + case 'A': + if (last) { + return -1; + } + goto ACCESS_CONTROL_EXPOSE_HEA; + case 'a': + if (last) { + return -1; + } + goto ACCESS_CONTROL_EXPOSE_HEA; + default: + return -1; + } + +ACCESS_CONTROL_EXPOSE_HEA: + NEXT_CHAR(); + switch (ch) { + case 'D': + if (last) { + return -1; + } + goto ACCESS_CONTROL_EXPOSE_HEAD; + case 'd': + if (last) { + return -1; + } + goto ACCESS_CONTROL_EXPOSE_HEAD; + default: + return -1; + } + +ACCESS_CONTROL_EXPOSE_HEAD: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto ACCESS_CONTROL_EXPOSE_HEADE; + case 'e': + if (last) { + return -1; + } + goto ACCESS_CONTROL_EXPOSE_HEADE; + default: + return -1; + } + +ACCESS_CONTROL_EXPOSE_HEADE: + NEXT_CHAR(); + switch (ch) { + case 'R': + if (last) { + return -1; + } + goto ACCESS_CONTROL_EXPOSE_HEADER; + case 'r': + if (last) { + return -1; + } + goto ACCESS_CONTROL_EXPOSE_HEADER; + default: + return -1; + } + +ACCESS_CONTROL_EXPOSE_HEADER: + NEXT_CHAR(); + switch (ch) { + case 'S': + if (last) { + return 9; + } + goto ACCESS_CONTROL_EXPOSE_HEADERS; + case 's': + if (last) { + return 9; + } + goto ACCESS_CONTROL_EXPOSE_HEADERS; + default: + return -1; + } + +ACCESS_CONTROL_M: + NEXT_CHAR(); + switch (ch) { + case 'A': + if (last) { + return -1; + } + goto ACCESS_CONTROL_MA; + case 'a': + if (last) { + return -1; + } + goto ACCESS_CONTROL_MA; + default: + return -1; + } + +ACCESS_CONTROL_MA: + NEXT_CHAR(); + switch (ch) { + case 'X': + if (last) { + return -1; + } + goto ACCESS_CONTROL_MAX; + case 'x': + if (last) { + return -1; + } + goto ACCESS_CONTROL_MAX; + default: + return -1; + } + +ACCESS_CONTROL_MAX: + NEXT_CHAR(); + switch (ch) { + case '-': + if (last) { + return -1; + } + goto ACCESS_CONTROL_MAX_; + default: + return -1; + } + +ACCESS_CONTROL_MAX_: + NEXT_CHAR(); + switch (ch) { + case 'A': + if (last) { + return -1; + } + goto ACCESS_CONTROL_MAX_A; + case 'a': + if (last) { + return -1; + } + goto ACCESS_CONTROL_MAX_A; + default: + return -1; + } + +ACCESS_CONTROL_MAX_A: + NEXT_CHAR(); + switch (ch) { + case 'G': + if (last) { + return -1; + } + goto ACCESS_CONTROL_MAX_AG; + case 'g': + if (last) { + return -1; + } + goto ACCESS_CONTROL_MAX_AG; + default: + return -1; + } + +ACCESS_CONTROL_MAX_AG: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return 10; + } + goto ACCESS_CONTROL_MAX_AGE; + case 'e': + if (last) { + return 10; + } + goto ACCESS_CONTROL_MAX_AGE; + default: + return -1; + } + +ACCESS_CONTROL_R: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto ACCESS_CONTROL_RE; + case 'e': + if (last) { + return -1; + } + goto ACCESS_CONTROL_RE; + default: + return -1; + } + +ACCESS_CONTROL_RE: + NEXT_CHAR(); + switch (ch) { + case 'Q': + if (last) { + return -1; + } + goto ACCESS_CONTROL_REQ; + case 'q': + if (last) { + return -1; + } + goto ACCESS_CONTROL_REQ; + default: + return -1; + } + +ACCESS_CONTROL_REQ: + NEXT_CHAR(); + switch (ch) { + case 'U': + if (last) { + return -1; + } + goto ACCESS_CONTROL_REQU; + case 'u': + if (last) { + return -1; + } + goto ACCESS_CONTROL_REQU; + default: + return -1; + } + +ACCESS_CONTROL_REQU: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto ACCESS_CONTROL_REQUE; + case 'e': + if (last) { + return -1; + } + goto ACCESS_CONTROL_REQUE; + default: + return -1; + } + +ACCESS_CONTROL_REQUE: + NEXT_CHAR(); + switch (ch) { + case 'S': + if (last) { + return -1; + } + goto ACCESS_CONTROL_REQUES; + case 's': + if (last) { + return -1; + } + goto ACCESS_CONTROL_REQUES; + default: + return -1; + } + +ACCESS_CONTROL_REQUES: + NEXT_CHAR(); + switch (ch) { + case 'T': + if (last) { + return -1; + } + goto ACCESS_CONTROL_REQUEST; + case 't': + if (last) { + return -1; + } + goto ACCESS_CONTROL_REQUEST; + default: + return -1; + } + +ACCESS_CONTROL_REQUEST: + NEXT_CHAR(); + switch (ch) { + case '-': + if (last) { + return -1; + } + goto ACCESS_CONTROL_REQUEST_; + default: + return -1; + } + +ACCESS_CONTROL_REQUEST_: + NEXT_CHAR(); + switch (ch) { + case 'H': + if (last) { + return -1; + } + goto ACCESS_CONTROL_REQUEST_H; + case 'h': + if (last) { + return -1; + } + goto ACCESS_CONTROL_REQUEST_H; + case 'M': + if (last) { + return -1; + } + goto ACCESS_CONTROL_REQUEST_M; + case 'm': + if (last) { + return -1; + } + goto ACCESS_CONTROL_REQUEST_M; + default: + return -1; + } + +ACCESS_CONTROL_REQUEST_H: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto ACCESS_CONTROL_REQUEST_HE; + case 'e': + if (last) { + return -1; + } + goto ACCESS_CONTROL_REQUEST_HE; + default: + return -1; + } + +ACCESS_CONTROL_REQUEST_HE: + NEXT_CHAR(); + switch (ch) { + case 'A': + if (last) { + return -1; + } + goto ACCESS_CONTROL_REQUEST_HEA; + case 'a': + if (last) { + return -1; + } + goto ACCESS_CONTROL_REQUEST_HEA; + default: + return -1; + } + +ACCESS_CONTROL_REQUEST_HEA: + NEXT_CHAR(); + switch (ch) { + case 'D': + if (last) { + return -1; + } + goto ACCESS_CONTROL_REQUEST_HEAD; + case 'd': + if (last) { + return -1; + } + goto ACCESS_CONTROL_REQUEST_HEAD; + default: + return -1; + } + +ACCESS_CONTROL_REQUEST_HEAD: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto ACCESS_CONTROL_REQUEST_HEADE; + case 'e': + if (last) { + return -1; + } + goto ACCESS_CONTROL_REQUEST_HEADE; + default: + return -1; + } + +ACCESS_CONTROL_REQUEST_HEADE: + NEXT_CHAR(); + switch (ch) { + case 'R': + if (last) { + return -1; + } + goto ACCESS_CONTROL_REQUEST_HEADER; + case 'r': + if (last) { + return -1; + } + goto ACCESS_CONTROL_REQUEST_HEADER; + default: + return -1; + } + +ACCESS_CONTROL_REQUEST_HEADER: + NEXT_CHAR(); + switch (ch) { + case 'S': + if (last) { + return 11; + } + goto ACCESS_CONTROL_REQUEST_HEADERS; + case 's': + if (last) { + return 11; + } + goto ACCESS_CONTROL_REQUEST_HEADERS; + default: + return -1; + } + +ACCESS_CONTROL_REQUEST_M: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto ACCESS_CONTROL_REQUEST_ME; + case 'e': + if (last) { + return -1; + } + goto ACCESS_CONTROL_REQUEST_ME; + default: + return -1; + } + +ACCESS_CONTROL_REQUEST_ME: + NEXT_CHAR(); + switch (ch) { + case 'T': + if (last) { + return -1; + } + goto ACCESS_CONTROL_REQUEST_MET; + case 't': + if (last) { + return -1; + } + goto ACCESS_CONTROL_REQUEST_MET; + default: + return -1; + } + +ACCESS_CONTROL_REQUEST_MET: + NEXT_CHAR(); + switch (ch) { + case 'H': + if (last) { + return -1; + } + goto ACCESS_CONTROL_REQUEST_METH; + case 'h': + if (last) { + return -1; + } + goto ACCESS_CONTROL_REQUEST_METH; + default: + return -1; + } + +ACCESS_CONTROL_REQUEST_METH: + NEXT_CHAR(); + switch (ch) { + case 'O': + if (last) { + return -1; + } + goto ACCESS_CONTROL_REQUEST_METHO; + case 'o': + if (last) { + return -1; + } + goto ACCESS_CONTROL_REQUEST_METHO; + default: + return -1; + } + +ACCESS_CONTROL_REQUEST_METHO: + NEXT_CHAR(); + switch (ch) { + case 'D': + if (last) { + return 12; + } + goto ACCESS_CONTROL_REQUEST_METHOD; + case 'd': + if (last) { + return 12; + } + goto ACCESS_CONTROL_REQUEST_METHOD; + default: + return -1; + } + +AG: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return 13; + } + goto AGE; + case 'e': + if (last) { + return 13; + } + goto AGE; + default: + return -1; + } + +AL: + NEXT_CHAR(); + switch (ch) { + case 'L': + if (last) { + return -1; + } + goto ALL; + case 'l': + if (last) { + return -1; + } + goto ALL; + default: + return -1; + } + +ALL: + NEXT_CHAR(); + switch (ch) { + case 'O': + if (last) { + return -1; + } + goto ALLO; + case 'o': + if (last) { + return -1; + } + goto ALLO; + default: + return -1; + } + +ALLO: + NEXT_CHAR(); + switch (ch) { + case 'W': + if (last) { + return 14; + } + goto ALLOW; + case 'w': + if (last) { + return 14; + } + goto ALLOW; + default: + return -1; + } + +AU: + NEXT_CHAR(); + switch (ch) { + case 'T': + if (last) { + return -1; + } + goto AUT; + case 't': + if (last) { + return -1; + } + goto AUT; + default: + return -1; + } + +AUT: + NEXT_CHAR(); + switch (ch) { + case 'H': + if (last) { + return -1; + } + goto AUTH; + case 'h': + if (last) { + return -1; + } + goto AUTH; + default: + return -1; + } + +AUTH: + NEXT_CHAR(); + switch (ch) { + case 'O': + if (last) { + return -1; + } + goto AUTHO; + case 'o': + if (last) { + return -1; + } + goto AUTHO; + default: + return -1; + } + +AUTHO: + NEXT_CHAR(); + switch (ch) { + case 'R': + if (last) { + return -1; + } + goto AUTHOR; + case 'r': + if (last) { + return -1; + } + goto AUTHOR; + default: + return -1; + } + +AUTHOR: + NEXT_CHAR(); + switch (ch) { + case 'I': + if (last) { + return -1; + } + goto AUTHORI; + case 'i': + if (last) { + return -1; + } + goto AUTHORI; + default: + return -1; + } + +AUTHORI: + NEXT_CHAR(); + switch (ch) { + case 'Z': + if (last) { + return -1; + } + goto AUTHORIZ; + case 'z': + if (last) { + return -1; + } + goto AUTHORIZ; + default: + return -1; + } + +AUTHORIZ: + NEXT_CHAR(); + switch (ch) { + case 'A': + if (last) { + return -1; + } + goto AUTHORIZA; + case 'a': + if (last) { + return -1; + } + goto AUTHORIZA; + default: + return -1; + } + +AUTHORIZA: + NEXT_CHAR(); + switch (ch) { + case 'T': + if (last) { + return -1; + } + goto AUTHORIZAT; + case 't': + if (last) { + return -1; + } + goto AUTHORIZAT; + default: + return -1; + } + +AUTHORIZAT: + NEXT_CHAR(); + switch (ch) { + case 'I': + if (last) { + return -1; + } + goto AUTHORIZATI; + case 'i': + if (last) { + return -1; + } + goto AUTHORIZATI; + default: + return -1; + } + +AUTHORIZATI: + NEXT_CHAR(); + switch (ch) { + case 'O': + if (last) { + return -1; + } + goto AUTHORIZATIO; + case 'o': + if (last) { + return -1; + } + goto AUTHORIZATIO; + default: + return -1; + } + +AUTHORIZATIO: + NEXT_CHAR(); + switch (ch) { + case 'N': + if (last) { + return 15; + } + goto AUTHORIZATION; + case 'n': + if (last) { + return 15; + } + goto AUTHORIZATION; + default: + return -1; + } + +C: + NEXT_CHAR(); + switch (ch) { + case 'A': + if (last) { + return -1; + } + goto CA; + case 'a': + if (last) { + return -1; + } + goto CA; + case 'O': + if (last) { + return -1; + } + goto CO; + case 'o': + if (last) { + return -1; + } + goto CO; + default: + return -1; + } + +CA: + NEXT_CHAR(); + switch (ch) { + case 'C': + if (last) { + return -1; + } + goto CAC; + case 'c': + if (last) { + return -1; + } + goto CAC; + default: + return -1; + } + +CAC: + NEXT_CHAR(); + switch (ch) { + case 'H': + if (last) { + return -1; + } + goto CACH; + case 'h': + if (last) { + return -1; + } + goto CACH; + default: + return -1; + } + +CACH: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto CACHE; + case 'e': + if (last) { + return -1; + } + goto CACHE; + default: + return -1; + } + +CACHE: + NEXT_CHAR(); + switch (ch) { + case '-': + if (last) { + return -1; + } + goto CACHE_; + default: + return -1; + } + +CACHE_: + NEXT_CHAR(); + switch (ch) { + case 'C': + if (last) { + return -1; + } + goto CACHE_C; + case 'c': + if (last) { + return -1; + } + goto CACHE_C; + default: + return -1; + } + +CACHE_C: + NEXT_CHAR(); + switch (ch) { + case 'O': + if (last) { + return -1; + } + goto CACHE_CO; + case 'o': + if (last) { + return -1; + } + goto CACHE_CO; + default: + return -1; + } + +CACHE_CO: + NEXT_CHAR(); + switch (ch) { + case 'N': + if (last) { + return -1; + } + goto CACHE_CON; + case 'n': + if (last) { + return -1; + } + goto CACHE_CON; + default: + return -1; + } + +CACHE_CON: + NEXT_CHAR(); + switch (ch) { + case 'T': + if (last) { + return -1; + } + goto CACHE_CONT; + case 't': + if (last) { + return -1; + } + goto CACHE_CONT; + default: + return -1; + } + +CACHE_CONT: + NEXT_CHAR(); + switch (ch) { + case 'R': + if (last) { + return -1; + } + goto CACHE_CONTR; + case 'r': + if (last) { + return -1; + } + goto CACHE_CONTR; + default: + return -1; + } + +CACHE_CONTR: + NEXT_CHAR(); + switch (ch) { + case 'O': + if (last) { + return -1; + } + goto CACHE_CONTRO; + case 'o': + if (last) { + return -1; + } + goto CACHE_CONTRO; + default: + return -1; + } + +CACHE_CONTRO: + NEXT_CHAR(); + switch (ch) { + case 'L': + if (last) { + return 16; + } + goto CACHE_CONTROL; + case 'l': + if (last) { + return 16; + } + goto CACHE_CONTROL; + default: + return -1; + } + +CO: + NEXT_CHAR(); + switch (ch) { + case 'N': + if (last) { + return -1; + } + goto CON; + case 'n': + if (last) { + return -1; + } + goto CON; + case 'O': + if (last) { + return -1; + } + goto COO; + case 'o': + if (last) { + return -1; + } + goto COO; + default: + return -1; + } + +CON: + NEXT_CHAR(); + switch (ch) { + case 'N': + if (last) { + return -1; + } + goto CONN; + case 'n': + if (last) { + return -1; + } + goto CONN; + case 'T': + if (last) { + return -1; + } + goto CONT; + case 't': + if (last) { + return -1; + } + goto CONT; + default: + return -1; + } + +CONN: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto CONNE; + case 'e': + if (last) { + return -1; + } + goto CONNE; + default: + return -1; + } + +CONNE: + NEXT_CHAR(); + switch (ch) { + case 'C': + if (last) { + return -1; + } + goto CONNEC; + case 'c': + if (last) { + return -1; + } + goto CONNEC; + default: + return -1; + } + +CONNEC: + NEXT_CHAR(); + switch (ch) { + case 'T': + if (last) { + return -1; + } + goto CONNECT; + case 't': + if (last) { + return -1; + } + goto CONNECT; + default: + return -1; + } + +CONNECT: + NEXT_CHAR(); + switch (ch) { + case 'I': + if (last) { + return -1; + } + goto CONNECTI; + case 'i': + if (last) { + return -1; + } + goto CONNECTI; + default: + return -1; + } + +CONNECTI: + NEXT_CHAR(); + switch (ch) { + case 'O': + if (last) { + return -1; + } + goto CONNECTIO; + case 'o': + if (last) { + return -1; + } + goto CONNECTIO; + default: + return -1; + } + +CONNECTIO: + NEXT_CHAR(); + switch (ch) { + case 'N': + if (last) { + return 17; + } + goto CONNECTION; + case 'n': + if (last) { + return 17; + } + goto CONNECTION; + default: + return -1; + } + +CONT: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto CONTE; + case 'e': + if (last) { + return -1; + } + goto CONTE; + default: + return -1; + } + +CONTE: + NEXT_CHAR(); + switch (ch) { + case 'N': + if (last) { + return -1; + } + goto CONTEN; + case 'n': + if (last) { + return -1; + } + goto CONTEN; + default: + return -1; + } + +CONTEN: + NEXT_CHAR(); + switch (ch) { + case 'T': + if (last) { + return -1; + } + goto CONTENT; + case 't': + if (last) { + return -1; + } + goto CONTENT; + default: + return -1; + } + +CONTENT: + NEXT_CHAR(); + switch (ch) { + case '-': + if (last) { + return -1; + } + goto CONTENT_; + default: + return -1; + } + +CONTENT_: + NEXT_CHAR(); + switch (ch) { + case 'D': + if (last) { + return -1; + } + goto CONTENT_D; + case 'd': + if (last) { + return -1; + } + goto CONTENT_D; + case 'E': + if (last) { + return -1; + } + goto CONTENT_E; + case 'e': + if (last) { + return -1; + } + goto CONTENT_E; + case 'L': + if (last) { + return -1; + } + goto CONTENT_L; + case 'l': + if (last) { + return -1; + } + goto CONTENT_L; + case 'M': + if (last) { + return -1; + } + goto CONTENT_M; + case 'm': + if (last) { + return -1; + } + goto CONTENT_M; + case 'R': + if (last) { + return -1; + } + goto CONTENT_R; + case 'r': + if (last) { + return -1; + } + goto CONTENT_R; + case 'T': + if (last) { + return -1; + } + goto CONTENT_T; + case 't': + if (last) { + return -1; + } + goto CONTENT_T; + default: + return -1; + } + +CONTENT_D: + NEXT_CHAR(); + switch (ch) { + case 'I': + if (last) { + return -1; + } + goto CONTENT_DI; + case 'i': + if (last) { + return -1; + } + goto CONTENT_DI; + default: + return -1; + } + +CONTENT_DI: + NEXT_CHAR(); + switch (ch) { + case 'S': + if (last) { + return -1; + } + goto CONTENT_DIS; + case 's': + if (last) { + return -1; + } + goto CONTENT_DIS; + default: + return -1; + } + +CONTENT_DIS: + NEXT_CHAR(); + switch (ch) { + case 'P': + if (last) { + return -1; + } + goto CONTENT_DISP; + case 'p': + if (last) { + return -1; + } + goto CONTENT_DISP; + default: + return -1; + } + +CONTENT_DISP: + NEXT_CHAR(); + switch (ch) { + case 'O': + if (last) { + return -1; + } + goto CONTENT_DISPO; + case 'o': + if (last) { + return -1; + } + goto CONTENT_DISPO; + default: + return -1; + } + +CONTENT_DISPO: + NEXT_CHAR(); + switch (ch) { + case 'S': + if (last) { + return -1; + } + goto CONTENT_DISPOS; + case 's': + if (last) { + return -1; + } + goto CONTENT_DISPOS; + default: + return -1; + } + +CONTENT_DISPOS: + NEXT_CHAR(); + switch (ch) { + case 'I': + if (last) { + return -1; + } + goto CONTENT_DISPOSI; + case 'i': + if (last) { + return -1; + } + goto CONTENT_DISPOSI; + default: + return -1; + } + +CONTENT_DISPOSI: + NEXT_CHAR(); + switch (ch) { + case 'T': + if (last) { + return -1; + } + goto CONTENT_DISPOSIT; + case 't': + if (last) { + return -1; + } + goto CONTENT_DISPOSIT; + default: + return -1; + } + +CONTENT_DISPOSIT: + NEXT_CHAR(); + switch (ch) { + case 'I': + if (last) { + return -1; + } + goto CONTENT_DISPOSITI; + case 'i': + if (last) { + return -1; + } + goto CONTENT_DISPOSITI; + default: + return -1; + } + +CONTENT_DISPOSITI: + NEXT_CHAR(); + switch (ch) { + case 'O': + if (last) { + return -1; + } + goto CONTENT_DISPOSITIO; + case 'o': + if (last) { + return -1; + } + goto CONTENT_DISPOSITIO; + default: + return -1; + } + +CONTENT_DISPOSITIO: + NEXT_CHAR(); + switch (ch) { + case 'N': + if (last) { + return 18; + } + goto CONTENT_DISPOSITION; + case 'n': + if (last) { + return 18; + } + goto CONTENT_DISPOSITION; + default: + return -1; + } + +CONTENT_E: + NEXT_CHAR(); + switch (ch) { + case 'N': + if (last) { + return -1; + } + goto CONTENT_EN; + case 'n': + if (last) { + return -1; + } + goto CONTENT_EN; + default: + return -1; + } + +CONTENT_EN: + NEXT_CHAR(); + switch (ch) { + case 'C': + if (last) { + return -1; + } + goto CONTENT_ENC; + case 'c': + if (last) { + return -1; + } + goto CONTENT_ENC; + default: + return -1; + } + +CONTENT_ENC: + NEXT_CHAR(); + switch (ch) { + case 'O': + if (last) { + return -1; + } + goto CONTENT_ENCO; + case 'o': + if (last) { + return -1; + } + goto CONTENT_ENCO; + default: + return -1; + } + +CONTENT_ENCO: + NEXT_CHAR(); + switch (ch) { + case 'D': + if (last) { + return -1; + } + goto CONTENT_ENCOD; + case 'd': + if (last) { + return -1; + } + goto CONTENT_ENCOD; + default: + return -1; + } + +CONTENT_ENCOD: + NEXT_CHAR(); + switch (ch) { + case 'I': + if (last) { + return -1; + } + goto CONTENT_ENCODI; + case 'i': + if (last) { + return -1; + } + goto CONTENT_ENCODI; + default: + return -1; + } + +CONTENT_ENCODI: + NEXT_CHAR(); + switch (ch) { + case 'N': + if (last) { + return -1; + } + goto CONTENT_ENCODIN; + case 'n': + if (last) { + return -1; + } + goto CONTENT_ENCODIN; + default: + return -1; + } + +CONTENT_ENCODIN: + NEXT_CHAR(); + switch (ch) { + case 'G': + if (last) { + return 19; + } + goto CONTENT_ENCODING; + case 'g': + if (last) { + return 19; + } + goto CONTENT_ENCODING; + default: + return -1; + } + +CONTENT_L: + NEXT_CHAR(); + switch (ch) { + case 'A': + if (last) { + return -1; + } + goto CONTENT_LA; + case 'a': + if (last) { + return -1; + } + goto CONTENT_LA; + case 'E': + if (last) { + return -1; + } + goto CONTENT_LE; + case 'e': + if (last) { + return -1; + } + goto CONTENT_LE; + case 'O': + if (last) { + return -1; + } + goto CONTENT_LO; + case 'o': + if (last) { + return -1; + } + goto CONTENT_LO; + default: + return -1; + } + +CONTENT_LA: + NEXT_CHAR(); + switch (ch) { + case 'N': + if (last) { + return -1; + } + goto CONTENT_LAN; + case 'n': + if (last) { + return -1; + } + goto CONTENT_LAN; + default: + return -1; + } + +CONTENT_LAN: + NEXT_CHAR(); + switch (ch) { + case 'G': + if (last) { + return -1; + } + goto CONTENT_LANG; + case 'g': + if (last) { + return -1; + } + goto CONTENT_LANG; + default: + return -1; + } + +CONTENT_LANG: + NEXT_CHAR(); + switch (ch) { + case 'U': + if (last) { + return -1; + } + goto CONTENT_LANGU; + case 'u': + if (last) { + return -1; + } + goto CONTENT_LANGU; + default: + return -1; + } + +CONTENT_LANGU: + NEXT_CHAR(); + switch (ch) { + case 'A': + if (last) { + return -1; + } + goto CONTENT_LANGUA; + case 'a': + if (last) { + return -1; + } + goto CONTENT_LANGUA; + default: + return -1; + } + +CONTENT_LANGUA: + NEXT_CHAR(); + switch (ch) { + case 'G': + if (last) { + return -1; + } + goto CONTENT_LANGUAG; + case 'g': + if (last) { + return -1; + } + goto CONTENT_LANGUAG; + default: + return -1; + } + +CONTENT_LANGUAG: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return 20; + } + goto CONTENT_LANGUAGE; + case 'e': + if (last) { + return 20; + } + goto CONTENT_LANGUAGE; + default: + return -1; + } + +CONTENT_LE: + NEXT_CHAR(); + switch (ch) { + case 'N': + if (last) { + return -1; + } + goto CONTENT_LEN; + case 'n': + if (last) { + return -1; + } + goto CONTENT_LEN; + default: + return -1; + } + +CONTENT_LEN: + NEXT_CHAR(); + switch (ch) { + case 'G': + if (last) { + return -1; + } + goto CONTENT_LENG; + case 'g': + if (last) { + return -1; + } + goto CONTENT_LENG; + default: + return -1; + } + +CONTENT_LENG: + NEXT_CHAR(); + switch (ch) { + case 'T': + if (last) { + return -1; + } + goto CONTENT_LENGT; + case 't': + if (last) { + return -1; + } + goto CONTENT_LENGT; + default: + return -1; + } + +CONTENT_LENGT: + NEXT_CHAR(); + switch (ch) { + case 'H': + if (last) { + return 21; + } + goto CONTENT_LENGTH; + case 'h': + if (last) { + return 21; + } + goto CONTENT_LENGTH; + default: + return -1; + } + +CONTENT_LO: + NEXT_CHAR(); + switch (ch) { + case 'C': + if (last) { + return -1; + } + goto CONTENT_LOC; + case 'c': + if (last) { + return -1; + } + goto CONTENT_LOC; + default: + return -1; + } + +CONTENT_LOC: + NEXT_CHAR(); + switch (ch) { + case 'A': + if (last) { + return -1; + } + goto CONTENT_LOCA; + case 'a': + if (last) { + return -1; + } + goto CONTENT_LOCA; + default: + return -1; + } + +CONTENT_LOCA: + NEXT_CHAR(); + switch (ch) { + case 'T': + if (last) { + return -1; + } + goto CONTENT_LOCAT; + case 't': + if (last) { + return -1; + } + goto CONTENT_LOCAT; + default: + return -1; + } + +CONTENT_LOCAT: + NEXT_CHAR(); + switch (ch) { + case 'I': + if (last) { + return -1; + } + goto CONTENT_LOCATI; + case 'i': + if (last) { + return -1; + } + goto CONTENT_LOCATI; + default: + return -1; + } + +CONTENT_LOCATI: + NEXT_CHAR(); + switch (ch) { + case 'O': + if (last) { + return -1; + } + goto CONTENT_LOCATIO; + case 'o': + if (last) { + return -1; + } + goto CONTENT_LOCATIO; + default: + return -1; + } + +CONTENT_LOCATIO: + NEXT_CHAR(); + switch (ch) { + case 'N': + if (last) { + return 22; + } + goto CONTENT_LOCATION; + case 'n': + if (last) { + return 22; + } + goto CONTENT_LOCATION; + default: + return -1; + } + +CONTENT_M: + NEXT_CHAR(); + switch (ch) { + case 'D': + if (last) { + return -1; + } + goto CONTENT_MD; + case 'd': + if (last) { + return -1; + } + goto CONTENT_MD; + default: + return -1; + } + +CONTENT_MD: + NEXT_CHAR(); + switch (ch) { + case '5': + if (last) { + return 23; + } + goto CONTENT_MD5; + default: + return -1; + } + +CONTENT_R: + NEXT_CHAR(); + switch (ch) { + case 'A': + if (last) { + return -1; + } + goto CONTENT_RA; + case 'a': + if (last) { + return -1; + } + goto CONTENT_RA; + default: + return -1; + } + +CONTENT_RA: + NEXT_CHAR(); + switch (ch) { + case 'N': + if (last) { + return -1; + } + goto CONTENT_RAN; + case 'n': + if (last) { + return -1; + } + goto CONTENT_RAN; + default: + return -1; + } + +CONTENT_RAN: + NEXT_CHAR(); + switch (ch) { + case 'G': + if (last) { + return -1; + } + goto CONTENT_RANG; + case 'g': + if (last) { + return -1; + } + goto CONTENT_RANG; + default: + return -1; + } + +CONTENT_RANG: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return 24; + } + goto CONTENT_RANGE; + case 'e': + if (last) { + return 24; + } + goto CONTENT_RANGE; + default: + return -1; + } + +CONTENT_T: + NEXT_CHAR(); + switch (ch) { + case 'R': + if (last) { + return -1; + } + goto CONTENT_TR; + case 'r': + if (last) { + return -1; + } + goto CONTENT_TR; + case 'Y': + if (last) { + return -1; + } + goto CONTENT_TY; + case 'y': + if (last) { + return -1; + } + goto CONTENT_TY; + default: + return -1; + } + +CONTENT_TR: + NEXT_CHAR(); + switch (ch) { + case 'A': + if (last) { + return -1; + } + goto CONTENT_TRA; + case 'a': + if (last) { + return -1; + } + goto CONTENT_TRA; + default: + return -1; + } + +CONTENT_TRA: + NEXT_CHAR(); + switch (ch) { + case 'N': + if (last) { + return -1; + } + goto CONTENT_TRAN; + case 'n': + if (last) { + return -1; + } + goto CONTENT_TRAN; + default: + return -1; + } + +CONTENT_TRAN: + NEXT_CHAR(); + switch (ch) { + case 'S': + if (last) { + return -1; + } + goto CONTENT_TRANS; + case 's': + if (last) { + return -1; + } + goto CONTENT_TRANS; + default: + return -1; + } + +CONTENT_TRANS: + NEXT_CHAR(); + switch (ch) { + case 'F': + if (last) { + return -1; + } + goto CONTENT_TRANSF; + case 'f': + if (last) { + return -1; + } + goto CONTENT_TRANSF; + default: + return -1; + } + +CONTENT_TRANSF: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto CONTENT_TRANSFE; + case 'e': + if (last) { + return -1; + } + goto CONTENT_TRANSFE; + default: + return -1; + } + +CONTENT_TRANSFE: + NEXT_CHAR(); + switch (ch) { + case 'R': + if (last) { + return -1; + } + goto CONTENT_TRANSFER; + case 'r': + if (last) { + return -1; + } + goto CONTENT_TRANSFER; + default: + return -1; + } + +CONTENT_TRANSFER: + NEXT_CHAR(); + switch (ch) { + case '-': + if (last) { + return -1; + } + goto CONTENT_TRANSFER_; + default: + return -1; + } + +CONTENT_TRANSFER_: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto CONTENT_TRANSFER_E; + case 'e': + if (last) { + return -1; + } + goto CONTENT_TRANSFER_E; + default: + return -1; + } + +CONTENT_TRANSFER_E: + NEXT_CHAR(); + switch (ch) { + case 'N': + if (last) { + return -1; + } + goto CONTENT_TRANSFER_EN; + case 'n': + if (last) { + return -1; + } + goto CONTENT_TRANSFER_EN; + default: + return -1; + } + +CONTENT_TRANSFER_EN: + NEXT_CHAR(); + switch (ch) { + case 'C': + if (last) { + return -1; + } + goto CONTENT_TRANSFER_ENC; + case 'c': + if (last) { + return -1; + } + goto CONTENT_TRANSFER_ENC; + default: + return -1; + } + +CONTENT_TRANSFER_ENC: + NEXT_CHAR(); + switch (ch) { + case 'O': + if (last) { + return -1; + } + goto CONTENT_TRANSFER_ENCO; + case 'o': + if (last) { + return -1; + } + goto CONTENT_TRANSFER_ENCO; + default: + return -1; + } + +CONTENT_TRANSFER_ENCO: + NEXT_CHAR(); + switch (ch) { + case 'D': + if (last) { + return -1; + } + goto CONTENT_TRANSFER_ENCOD; + case 'd': + if (last) { + return -1; + } + goto CONTENT_TRANSFER_ENCOD; + default: + return -1; + } + +CONTENT_TRANSFER_ENCOD: + NEXT_CHAR(); + switch (ch) { + case 'I': + if (last) { + return -1; + } + goto CONTENT_TRANSFER_ENCODI; + case 'i': + if (last) { + return -1; + } + goto CONTENT_TRANSFER_ENCODI; + default: + return -1; + } + +CONTENT_TRANSFER_ENCODI: + NEXT_CHAR(); + switch (ch) { + case 'N': + if (last) { + return -1; + } + goto CONTENT_TRANSFER_ENCODIN; + case 'n': + if (last) { + return -1; + } + goto CONTENT_TRANSFER_ENCODIN; + default: + return -1; + } + +CONTENT_TRANSFER_ENCODIN: + NEXT_CHAR(); + switch (ch) { + case 'G': + if (last) { + return 25; + } + goto CONTENT_TRANSFER_ENCODING; + case 'g': + if (last) { + return 25; + } + goto CONTENT_TRANSFER_ENCODING; + default: + return -1; + } + +CONTENT_TY: + NEXT_CHAR(); + switch (ch) { + case 'P': + if (last) { + return -1; + } + goto CONTENT_TYP; + case 'p': + if (last) { + return -1; + } + goto CONTENT_TYP; + default: + return -1; + } + +CONTENT_TYP: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return 26; + } + goto CONTENT_TYPE; + case 'e': + if (last) { + return 26; + } + goto CONTENT_TYPE; + default: + return -1; + } + +COO: + NEXT_CHAR(); + switch (ch) { + case 'K': + if (last) { + return -1; + } + goto COOK; + case 'k': + if (last) { + return -1; + } + goto COOK; + default: + return -1; + } + +COOK: + NEXT_CHAR(); + switch (ch) { + case 'I': + if (last) { + return -1; + } + goto COOKI; + case 'i': + if (last) { + return -1; + } + goto COOKI; + default: + return -1; + } + +COOKI: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return 27; + } + goto COOKIE; + case 'e': + if (last) { + return 27; + } + goto COOKIE; + default: + return -1; + } + +D: + NEXT_CHAR(); + switch (ch) { + case 'A': + if (last) { + return -1; + } + goto DA; + case 'a': + if (last) { + return -1; + } + goto DA; + case 'E': + if (last) { + return -1; + } + goto DE; + case 'e': + if (last) { + return -1; + } + goto DE; + case 'I': + if (last) { + return -1; + } + goto DI; + case 'i': + if (last) { + return -1; + } + goto DI; + default: + return -1; + } + +DA: + NEXT_CHAR(); + switch (ch) { + case 'T': + if (last) { + return -1; + } + goto DAT; + case 't': + if (last) { + return -1; + } + goto DAT; + default: + return -1; + } + +DAT: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return 28; + } + goto DATE; + case 'e': + if (last) { + return 28; + } + goto DATE; + default: + return -1; + } + +DE: + NEXT_CHAR(); + switch (ch) { + case 'S': + if (last) { + return -1; + } + goto DES; + case 's': + if (last) { + return -1; + } + goto DES; + default: + return -1; + } + +DES: + NEXT_CHAR(); + switch (ch) { + case 'T': + if (last) { + return -1; + } + goto DEST; + case 't': + if (last) { + return -1; + } + goto DEST; + default: + return -1; + } + +DEST: + NEXT_CHAR(); + switch (ch) { + case 'I': + if (last) { + return -1; + } + goto DESTI; + case 'i': + if (last) { + return -1; + } + goto DESTI; + default: + return -1; + } + +DESTI: + NEXT_CHAR(); + switch (ch) { + case 'N': + if (last) { + return -1; + } + goto DESTIN; + case 'n': + if (last) { + return -1; + } + goto DESTIN; + default: + return -1; + } + +DESTIN: + NEXT_CHAR(); + switch (ch) { + case 'A': + if (last) { + return -1; + } + goto DESTINA; + case 'a': + if (last) { + return -1; + } + goto DESTINA; + default: + return -1; + } + +DESTINA: + NEXT_CHAR(); + switch (ch) { + case 'T': + if (last) { + return -1; + } + goto DESTINAT; + case 't': + if (last) { + return -1; + } + goto DESTINAT; + default: + return -1; + } + +DESTINAT: + NEXT_CHAR(); + switch (ch) { + case 'I': + if (last) { + return -1; + } + goto DESTINATI; + case 'i': + if (last) { + return -1; + } + goto DESTINATI; + default: + return -1; + } + +DESTINATI: + NEXT_CHAR(); + switch (ch) { + case 'O': + if (last) { + return -1; + } + goto DESTINATIO; + case 'o': + if (last) { + return -1; + } + goto DESTINATIO; + default: + return -1; + } + +DESTINATIO: + NEXT_CHAR(); + switch (ch) { + case 'N': + if (last) { + return 29; + } + goto DESTINATION; + case 'n': + if (last) { + return 29; + } + goto DESTINATION; + default: + return -1; + } + +DI: + NEXT_CHAR(); + switch (ch) { + case 'G': + if (last) { + return -1; + } + goto DIG; + case 'g': + if (last) { + return -1; + } + goto DIG; + default: + return -1; + } + +DIG: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto DIGE; + case 'e': + if (last) { + return -1; + } + goto DIGE; + default: + return -1; + } + +DIGE: + NEXT_CHAR(); + switch (ch) { + case 'S': + if (last) { + return -1; + } + goto DIGES; + case 's': + if (last) { + return -1; + } + goto DIGES; + default: + return -1; + } + +DIGES: + NEXT_CHAR(); + switch (ch) { + case 'T': + if (last) { + return 30; + } + goto DIGEST; + case 't': + if (last) { + return 30; + } + goto DIGEST; + default: + return -1; + } + +E: + NEXT_CHAR(); + switch (ch) { + case 'T': + if (last) { + return -1; + } + goto ET; + case 't': + if (last) { + return -1; + } + goto ET; + case 'X': + if (last) { + return -1; + } + goto EX; + case 'x': + if (last) { + return -1; + } + goto EX; + default: + return -1; + } + +ET: + NEXT_CHAR(); + switch (ch) { + case 'A': + if (last) { + return -1; + } + goto ETA; + case 'a': + if (last) { + return -1; + } + goto ETA; + default: + return -1; + } + +ETA: + NEXT_CHAR(); + switch (ch) { + case 'G': + if (last) { + return 31; + } + goto ETAG; + case 'g': + if (last) { + return 31; + } + goto ETAG; + default: + return -1; + } + +EX: + NEXT_CHAR(); + switch (ch) { + case 'P': + if (last) { + return -1; + } + goto EXP; + case 'p': + if (last) { + return -1; + } + goto EXP; + default: + return -1; + } + +EXP: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto EXPE; + case 'e': + if (last) { + return -1; + } + goto EXPE; + case 'I': + if (last) { + return -1; + } + goto EXPI; + case 'i': + if (last) { + return -1; + } + goto EXPI; + default: + return -1; + } + +EXPE: + NEXT_CHAR(); + switch (ch) { + case 'C': + if (last) { + return -1; + } + goto EXPEC; + case 'c': + if (last) { + return -1; + } + goto EXPEC; + default: + return -1; + } + +EXPEC: + NEXT_CHAR(); + switch (ch) { + case 'T': + if (last) { + return 32; + } + goto EXPECT; + case 't': + if (last) { + return 32; + } + goto EXPECT; + default: + return -1; + } + +EXPI: + NEXT_CHAR(); + switch (ch) { + case 'R': + if (last) { + return -1; + } + goto EXPIR; + case 'r': + if (last) { + return -1; + } + goto EXPIR; + default: + return -1; + } + +EXPIR: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto EXPIRE; + case 'e': + if (last) { + return -1; + } + goto EXPIRE; + default: + return -1; + } + +EXPIRE: + NEXT_CHAR(); + switch (ch) { + case 'S': + if (last) { + return 33; + } + goto EXPIRES; + case 's': + if (last) { + return 33; + } + goto EXPIRES; + default: + return -1; + } + +F: + NEXT_CHAR(); + switch (ch) { + case 'O': + if (last) { + return -1; + } + goto FO; + case 'o': + if (last) { + return -1; + } + goto FO; + case 'R': + if (last) { + return -1; + } + goto FR; + case 'r': + if (last) { + return -1; + } + goto FR; + default: + return -1; + } + +FO: + NEXT_CHAR(); + switch (ch) { + case 'R': + if (last) { + return -1; + } + goto FOR; + case 'r': + if (last) { + return -1; + } + goto FOR; + default: + return -1; + } + +FOR: + NEXT_CHAR(); + switch (ch) { + case 'W': + if (last) { + return -1; + } + goto FORW; + case 'w': + if (last) { + return -1; + } + goto FORW; + default: + return -1; + } + +FORW: + NEXT_CHAR(); + switch (ch) { + case 'A': + if (last) { + return -1; + } + goto FORWA; + case 'a': + if (last) { + return -1; + } + goto FORWA; + default: + return -1; + } + +FORWA: + NEXT_CHAR(); + switch (ch) { + case 'R': + if (last) { + return -1; + } + goto FORWAR; + case 'r': + if (last) { + return -1; + } + goto FORWAR; + default: + return -1; + } + +FORWAR: + NEXT_CHAR(); + switch (ch) { + case 'D': + if (last) { + return -1; + } + goto FORWARD; + case 'd': + if (last) { + return -1; + } + goto FORWARD; + default: + return -1; + } + +FORWARD: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto FORWARDE; + case 'e': + if (last) { + return -1; + } + goto FORWARDE; + default: + return -1; + } + +FORWARDE: + NEXT_CHAR(); + switch (ch) { + case 'D': + if (last) { + return 34; + } + goto FORWARDED; + case 'd': + if (last) { + return 34; + } + goto FORWARDED; + default: + return -1; + } + +FR: + NEXT_CHAR(); + switch (ch) { + case 'O': + if (last) { + return -1; + } + goto FRO; + case 'o': + if (last) { + return -1; + } + goto FRO; + default: + return -1; + } + +FRO: + NEXT_CHAR(); + switch (ch) { + case 'M': + if (last) { + return 35; + } + goto FROM; + case 'm': + if (last) { + return 35; + } + goto FROM; + default: + return -1; + } + +H: + NEXT_CHAR(); + switch (ch) { + case 'O': + if (last) { + return -1; + } + goto HO; + case 'o': + if (last) { + return -1; + } + goto HO; + default: + return -1; + } + +HO: + NEXT_CHAR(); + switch (ch) { + case 'S': + if (last) { + return -1; + } + goto HOS; + case 's': + if (last) { + return -1; + } + goto HOS; + default: + return -1; + } + +HOS: + NEXT_CHAR(); + switch (ch) { + case 'T': + if (last) { + return 36; + } + goto HOST; + case 't': + if (last) { + return 36; + } + goto HOST; + default: + return -1; + } + +I: + NEXT_CHAR(); + switch (ch) { + case 'F': + if (last) { + return -1; + } + goto IF; + case 'f': + if (last) { + return -1; + } + goto IF; + default: + return -1; + } + +IF: + NEXT_CHAR(); + switch (ch) { + case '-': + if (last) { + return -1; + } + goto IF_; + default: + return -1; + } + +IF_: + NEXT_CHAR(); + switch (ch) { + case 'M': + if (last) { + return -1; + } + goto IF_M; + case 'm': + if (last) { + return -1; + } + goto IF_M; + case 'N': + if (last) { + return -1; + } + goto IF_N; + case 'n': + if (last) { + return -1; + } + goto IF_N; + case 'R': + if (last) { + return -1; + } + goto IF_R; + case 'r': + if (last) { + return -1; + } + goto IF_R; + case 'U': + if (last) { + return -1; + } + goto IF_U; + case 'u': + if (last) { + return -1; + } + goto IF_U; + default: + return -1; + } + +IF_M: + NEXT_CHAR(); + switch (ch) { + case 'A': + if (last) { + return -1; + } + goto IF_MA; + case 'a': + if (last) { + return -1; + } + goto IF_MA; + case 'O': + if (last) { + return -1; + } + goto IF_MO; + case 'o': + if (last) { + return -1; + } + goto IF_MO; + default: + return -1; + } + +IF_MA: + NEXT_CHAR(); + switch (ch) { + case 'T': + if (last) { + return -1; + } + goto IF_MAT; + case 't': + if (last) { + return -1; + } + goto IF_MAT; + default: + return -1; + } + +IF_MAT: + NEXT_CHAR(); + switch (ch) { + case 'C': + if (last) { + return -1; + } + goto IF_MATC; + case 'c': + if (last) { + return -1; + } + goto IF_MATC; + default: + return -1; + } + +IF_MATC: + NEXT_CHAR(); + switch (ch) { + case 'H': + if (last) { + return 37; + } + goto IF_MATCH; + case 'h': + if (last) { + return 37; + } + goto IF_MATCH; + default: + return -1; + } + +IF_MO: + NEXT_CHAR(); + switch (ch) { + case 'D': + if (last) { + return -1; + } + goto IF_MOD; + case 'd': + if (last) { + return -1; + } + goto IF_MOD; + default: + return -1; + } + +IF_MOD: + NEXT_CHAR(); + switch (ch) { + case 'I': + if (last) { + return -1; + } + goto IF_MODI; + case 'i': + if (last) { + return -1; + } + goto IF_MODI; + default: + return -1; + } + +IF_MODI: + NEXT_CHAR(); + switch (ch) { + case 'F': + if (last) { + return -1; + } + goto IF_MODIF; + case 'f': + if (last) { + return -1; + } + goto IF_MODIF; + default: + return -1; + } + +IF_MODIF: + NEXT_CHAR(); + switch (ch) { + case 'I': + if (last) { + return -1; + } + goto IF_MODIFI; + case 'i': + if (last) { + return -1; + } + goto IF_MODIFI; + default: + return -1; + } + +IF_MODIFI: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto IF_MODIFIE; + case 'e': + if (last) { + return -1; + } + goto IF_MODIFIE; + default: + return -1; + } + +IF_MODIFIE: + NEXT_CHAR(); + switch (ch) { + case 'D': + if (last) { + return -1; + } + goto IF_MODIFIED; + case 'd': + if (last) { + return -1; + } + goto IF_MODIFIED; + default: + return -1; + } + +IF_MODIFIED: + NEXT_CHAR(); + switch (ch) { + case '-': + if (last) { + return -1; + } + goto IF_MODIFIED_; + default: + return -1; + } + +IF_MODIFIED_: + NEXT_CHAR(); + switch (ch) { + case 'S': + if (last) { + return -1; + } + goto IF_MODIFIED_S; + case 's': + if (last) { + return -1; + } + goto IF_MODIFIED_S; + default: + return -1; + } + +IF_MODIFIED_S: + NEXT_CHAR(); + switch (ch) { + case 'I': + if (last) { + return -1; + } + goto IF_MODIFIED_SI; + case 'i': + if (last) { + return -1; + } + goto IF_MODIFIED_SI; + default: + return -1; + } + +IF_MODIFIED_SI: + NEXT_CHAR(); + switch (ch) { + case 'N': + if (last) { + return -1; + } + goto IF_MODIFIED_SIN; + case 'n': + if (last) { + return -1; + } + goto IF_MODIFIED_SIN; + default: + return -1; + } + +IF_MODIFIED_SIN: + NEXT_CHAR(); + switch (ch) { + case 'C': + if (last) { + return -1; + } + goto IF_MODIFIED_SINC; + case 'c': + if (last) { + return -1; + } + goto IF_MODIFIED_SINC; + default: + return -1; + } + +IF_MODIFIED_SINC: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return 38; + } + goto IF_MODIFIED_SINCE; + case 'e': + if (last) { + return 38; + } + goto IF_MODIFIED_SINCE; + default: + return -1; + } + +IF_N: + NEXT_CHAR(); + switch (ch) { + case 'O': + if (last) { + return -1; + } + goto IF_NO; + case 'o': + if (last) { + return -1; + } + goto IF_NO; + default: + return -1; + } + +IF_NO: + NEXT_CHAR(); + switch (ch) { + case 'N': + if (last) { + return -1; + } + goto IF_NON; + case 'n': + if (last) { + return -1; + } + goto IF_NON; + default: + return -1; + } + +IF_NON: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto IF_NONE; + case 'e': + if (last) { + return -1; + } + goto IF_NONE; + default: + return -1; + } + +IF_NONE: + NEXT_CHAR(); + switch (ch) { + case '-': + if (last) { + return -1; + } + goto IF_NONE_; + default: + return -1; + } + +IF_NONE_: + NEXT_CHAR(); + switch (ch) { + case 'M': + if (last) { + return -1; + } + goto IF_NONE_M; + case 'm': + if (last) { + return -1; + } + goto IF_NONE_M; + default: + return -1; + } + +IF_NONE_M: + NEXT_CHAR(); + switch (ch) { + case 'A': + if (last) { + return -1; + } + goto IF_NONE_MA; + case 'a': + if (last) { + return -1; + } + goto IF_NONE_MA; + default: + return -1; + } + +IF_NONE_MA: + NEXT_CHAR(); + switch (ch) { + case 'T': + if (last) { + return -1; + } + goto IF_NONE_MAT; + case 't': + if (last) { + return -1; + } + goto IF_NONE_MAT; + default: + return -1; + } + +IF_NONE_MAT: + NEXT_CHAR(); + switch (ch) { + case 'C': + if (last) { + return -1; + } + goto IF_NONE_MATC; + case 'c': + if (last) { + return -1; + } + goto IF_NONE_MATC; + default: + return -1; + } + +IF_NONE_MATC: + NEXT_CHAR(); + switch (ch) { + case 'H': + if (last) { + return 39; + } + goto IF_NONE_MATCH; + case 'h': + if (last) { + return 39; + } + goto IF_NONE_MATCH; + default: + return -1; + } + +IF_R: + NEXT_CHAR(); + switch (ch) { + case 'A': + if (last) { + return -1; + } + goto IF_RA; + case 'a': + if (last) { + return -1; + } + goto IF_RA; + default: + return -1; + } + +IF_RA: + NEXT_CHAR(); + switch (ch) { + case 'N': + if (last) { + return -1; + } + goto IF_RAN; + case 'n': + if (last) { + return -1; + } + goto IF_RAN; + default: + return -1; + } + +IF_RAN: + NEXT_CHAR(); + switch (ch) { + case 'G': + if (last) { + return -1; + } + goto IF_RANG; + case 'g': + if (last) { + return -1; + } + goto IF_RANG; + default: + return -1; + } + +IF_RANG: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return 40; + } + goto IF_RANGE; + case 'e': + if (last) { + return 40; + } + goto IF_RANGE; + default: + return -1; + } + +IF_U: + NEXT_CHAR(); + switch (ch) { + case 'N': + if (last) { + return -1; + } + goto IF_UN; + case 'n': + if (last) { + return -1; + } + goto IF_UN; + default: + return -1; + } + +IF_UN: + NEXT_CHAR(); + switch (ch) { + case 'M': + if (last) { + return -1; + } + goto IF_UNM; + case 'm': + if (last) { + return -1; + } + goto IF_UNM; + default: + return -1; + } + +IF_UNM: + NEXT_CHAR(); + switch (ch) { + case 'O': + if (last) { + return -1; + } + goto IF_UNMO; + case 'o': + if (last) { + return -1; + } + goto IF_UNMO; + default: + return -1; + } + +IF_UNMO: + NEXT_CHAR(); + switch (ch) { + case 'D': + if (last) { + return -1; + } + goto IF_UNMOD; + case 'd': + if (last) { + return -1; + } + goto IF_UNMOD; + default: + return -1; + } + +IF_UNMOD: + NEXT_CHAR(); + switch (ch) { + case 'I': + if (last) { + return -1; + } + goto IF_UNMODI; + case 'i': + if (last) { + return -1; + } + goto IF_UNMODI; + default: + return -1; + } + +IF_UNMODI: + NEXT_CHAR(); + switch (ch) { + case 'F': + if (last) { + return -1; + } + goto IF_UNMODIF; + case 'f': + if (last) { + return -1; + } + goto IF_UNMODIF; + default: + return -1; + } + +IF_UNMODIF: + NEXT_CHAR(); + switch (ch) { + case 'I': + if (last) { + return -1; + } + goto IF_UNMODIFI; + case 'i': + if (last) { + return -1; + } + goto IF_UNMODIFI; + default: + return -1; + } + +IF_UNMODIFI: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto IF_UNMODIFIE; + case 'e': + if (last) { + return -1; + } + goto IF_UNMODIFIE; + default: + return -1; + } + +IF_UNMODIFIE: + NEXT_CHAR(); + switch (ch) { + case 'D': + if (last) { + return -1; + } + goto IF_UNMODIFIED; + case 'd': + if (last) { + return -1; + } + goto IF_UNMODIFIED; + default: + return -1; + } + +IF_UNMODIFIED: + NEXT_CHAR(); + switch (ch) { + case '-': + if (last) { + return -1; + } + goto IF_UNMODIFIED_; + default: + return -1; + } + +IF_UNMODIFIED_: + NEXT_CHAR(); + switch (ch) { + case 'S': + if (last) { + return -1; + } + goto IF_UNMODIFIED_S; + case 's': + if (last) { + return -1; + } + goto IF_UNMODIFIED_S; + default: + return -1; + } + +IF_UNMODIFIED_S: + NEXT_CHAR(); + switch (ch) { + case 'I': + if (last) { + return -1; + } + goto IF_UNMODIFIED_SI; + case 'i': + if (last) { + return -1; + } + goto IF_UNMODIFIED_SI; + default: + return -1; + } + +IF_UNMODIFIED_SI: + NEXT_CHAR(); + switch (ch) { + case 'N': + if (last) { + return -1; + } + goto IF_UNMODIFIED_SIN; + case 'n': + if (last) { + return -1; + } + goto IF_UNMODIFIED_SIN; + default: + return -1; + } + +IF_UNMODIFIED_SIN: + NEXT_CHAR(); + switch (ch) { + case 'C': + if (last) { + return -1; + } + goto IF_UNMODIFIED_SINC; + case 'c': + if (last) { + return -1; + } + goto IF_UNMODIFIED_SINC; + default: + return -1; + } + +IF_UNMODIFIED_SINC: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return 41; + } + goto IF_UNMODIFIED_SINCE; + case 'e': + if (last) { + return 41; + } + goto IF_UNMODIFIED_SINCE; + default: + return -1; + } + +K: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto KE; + case 'e': + if (last) { + return -1; + } + goto KE; + default: + return -1; + } + +KE: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto KEE; + case 'e': + if (last) { + return -1; + } + goto KEE; + default: + return -1; + } + +KEE: + NEXT_CHAR(); + switch (ch) { + case 'P': + if (last) { + return -1; + } + goto KEEP; + case 'p': + if (last) { + return -1; + } + goto KEEP; + default: + return -1; + } + +KEEP: + NEXT_CHAR(); + switch (ch) { + case '-': + if (last) { + return -1; + } + goto KEEP_; + default: + return -1; + } + +KEEP_: + NEXT_CHAR(); + switch (ch) { + case 'A': + if (last) { + return -1; + } + goto KEEP_A; + case 'a': + if (last) { + return -1; + } + goto KEEP_A; + default: + return -1; + } + +KEEP_A: + NEXT_CHAR(); + switch (ch) { + case 'L': + if (last) { + return -1; + } + goto KEEP_AL; + case 'l': + if (last) { + return -1; + } + goto KEEP_AL; + default: + return -1; + } + +KEEP_AL: + NEXT_CHAR(); + switch (ch) { + case 'I': + if (last) { + return -1; + } + goto KEEP_ALI; + case 'i': + if (last) { + return -1; + } + goto KEEP_ALI; + default: + return -1; + } + +KEEP_ALI: + NEXT_CHAR(); + switch (ch) { + case 'V': + if (last) { + return -1; + } + goto KEEP_ALIV; + case 'v': + if (last) { + return -1; + } + goto KEEP_ALIV; + default: + return -1; + } + +KEEP_ALIV: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return 42; + } + goto KEEP_ALIVE; + case 'e': + if (last) { + return 42; + } + goto KEEP_ALIVE; + default: + return -1; + } + +L: + NEXT_CHAR(); + switch (ch) { + case 'A': + if (last) { + return -1; + } + goto LA; + case 'a': + if (last) { + return -1; + } + goto LA; + case 'I': + if (last) { + return -1; + } + goto LI; + case 'i': + if (last) { + return -1; + } + goto LI; + case 'O': + if (last) { + return -1; + } + goto LO; + case 'o': + if (last) { + return -1; + } + goto LO; + default: + return -1; + } + +LA: + NEXT_CHAR(); + switch (ch) { + case 'S': + if (last) { + return -1; + } + goto LAS; + case 's': + if (last) { + return -1; + } + goto LAS; + default: + return -1; + } + +LAS: + NEXT_CHAR(); + switch (ch) { + case 'T': + if (last) { + return -1; + } + goto LAST; + case 't': + if (last) { + return -1; + } + goto LAST; + default: + return -1; + } + +LAST: + NEXT_CHAR(); + switch (ch) { + case '-': + if (last) { + return -1; + } + goto LAST_; + default: + return -1; + } + +LAST_: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto LAST_E; + case 'e': + if (last) { + return -1; + } + goto LAST_E; + case 'M': + if (last) { + return -1; + } + goto LAST_M; + case 'm': + if (last) { + return -1; + } + goto LAST_M; + default: + return -1; + } + +LAST_E: + NEXT_CHAR(); + switch (ch) { + case 'V': + if (last) { + return -1; + } + goto LAST_EV; + case 'v': + if (last) { + return -1; + } + goto LAST_EV; + default: + return -1; + } + +LAST_EV: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto LAST_EVE; + case 'e': + if (last) { + return -1; + } + goto LAST_EVE; + default: + return -1; + } + +LAST_EVE: + NEXT_CHAR(); + switch (ch) { + case 'N': + if (last) { + return -1; + } + goto LAST_EVEN; + case 'n': + if (last) { + return -1; + } + goto LAST_EVEN; + default: + return -1; + } + +LAST_EVEN: + NEXT_CHAR(); + switch (ch) { + case 'T': + if (last) { + return -1; + } + goto LAST_EVENT; + case 't': + if (last) { + return -1; + } + goto LAST_EVENT; + default: + return -1; + } + +LAST_EVENT: + NEXT_CHAR(); + switch (ch) { + case '-': + if (last) { + return -1; + } + goto LAST_EVENT_; + default: + return -1; + } + +LAST_EVENT_: + NEXT_CHAR(); + switch (ch) { + case 'I': + if (last) { + return -1; + } + goto LAST_EVENT_I; + case 'i': + if (last) { + return -1; + } + goto LAST_EVENT_I; + default: + return -1; + } + +LAST_EVENT_I: + NEXT_CHAR(); + switch (ch) { + case 'D': + if (last) { + return 43; + } + goto LAST_EVENT_ID; + case 'd': + if (last) { + return 43; + } + goto LAST_EVENT_ID; + default: + return -1; + } + +LAST_M: + NEXT_CHAR(); + switch (ch) { + case 'O': + if (last) { + return -1; + } + goto LAST_MO; + case 'o': + if (last) { + return -1; + } + goto LAST_MO; + default: + return -1; + } + +LAST_MO: + NEXT_CHAR(); + switch (ch) { + case 'D': + if (last) { + return -1; + } + goto LAST_MOD; + case 'd': + if (last) { + return -1; + } + goto LAST_MOD; + default: + return -1; + } + +LAST_MOD: + NEXT_CHAR(); + switch (ch) { + case 'I': + if (last) { + return -1; + } + goto LAST_MODI; + case 'i': + if (last) { + return -1; + } + goto LAST_MODI; + default: + return -1; + } + +LAST_MODI: + NEXT_CHAR(); + switch (ch) { + case 'F': + if (last) { + return -1; + } + goto LAST_MODIF; + case 'f': + if (last) { + return -1; + } + goto LAST_MODIF; + default: + return -1; + } + +LAST_MODIF: + NEXT_CHAR(); + switch (ch) { + case 'I': + if (last) { + return -1; + } + goto LAST_MODIFI; + case 'i': + if (last) { + return -1; + } + goto LAST_MODIFI; + default: + return -1; + } + +LAST_MODIFI: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto LAST_MODIFIE; + case 'e': + if (last) { + return -1; + } + goto LAST_MODIFIE; + default: + return -1; + } + +LAST_MODIFIE: + NEXT_CHAR(); + switch (ch) { + case 'D': + if (last) { + return 44; + } + goto LAST_MODIFIED; + case 'd': + if (last) { + return 44; + } + goto LAST_MODIFIED; + default: + return -1; + } + +LI: + NEXT_CHAR(); + switch (ch) { + case 'N': + if (last) { + return -1; + } + goto LIN; + case 'n': + if (last) { + return -1; + } + goto LIN; + default: + return -1; + } + +LIN: + NEXT_CHAR(); + switch (ch) { + case 'K': + if (last) { + return 45; + } + goto LINK; + case 'k': + if (last) { + return 45; + } + goto LINK; + default: + return -1; + } + +LO: + NEXT_CHAR(); + switch (ch) { + case 'C': + if (last) { + return -1; + } + goto LOC; + case 'c': + if (last) { + return -1; + } + goto LOC; + default: + return -1; + } + +LOC: + NEXT_CHAR(); + switch (ch) { + case 'A': + if (last) { + return -1; + } + goto LOCA; + case 'a': + if (last) { + return -1; + } + goto LOCA; + default: + return -1; + } + +LOCA: + NEXT_CHAR(); + switch (ch) { + case 'T': + if (last) { + return -1; + } + goto LOCAT; + case 't': + if (last) { + return -1; + } + goto LOCAT; + default: + return -1; + } + +LOCAT: + NEXT_CHAR(); + switch (ch) { + case 'I': + if (last) { + return -1; + } + goto LOCATI; + case 'i': + if (last) { + return -1; + } + goto LOCATI; + default: + return -1; + } + +LOCATI: + NEXT_CHAR(); + switch (ch) { + case 'O': + if (last) { + return -1; + } + goto LOCATIO; + case 'o': + if (last) { + return -1; + } + goto LOCATIO; + default: + return -1; + } + +LOCATIO: + NEXT_CHAR(); + switch (ch) { + case 'N': + if (last) { + return 46; + } + goto LOCATION; + case 'n': + if (last) { + return 46; + } + goto LOCATION; + default: + return -1; + } + +M: + NEXT_CHAR(); + switch (ch) { + case 'A': + if (last) { + return -1; + } + goto MA; + case 'a': + if (last) { + return -1; + } + goto MA; + default: + return -1; + } + +MA: + NEXT_CHAR(); + switch (ch) { + case 'X': + if (last) { + return -1; + } + goto MAX; + case 'x': + if (last) { + return -1; + } + goto MAX; + default: + return -1; + } + +MAX: + NEXT_CHAR(); + switch (ch) { + case '-': + if (last) { + return -1; + } + goto MAX_; + default: + return -1; + } + +MAX_: + NEXT_CHAR(); + switch (ch) { + case 'F': + if (last) { + return -1; + } + goto MAX_F; + case 'f': + if (last) { + return -1; + } + goto MAX_F; + default: + return -1; + } + +MAX_F: + NEXT_CHAR(); + switch (ch) { + case 'O': + if (last) { + return -1; + } + goto MAX_FO; + case 'o': + if (last) { + return -1; + } + goto MAX_FO; + default: + return -1; + } + +MAX_FO: + NEXT_CHAR(); + switch (ch) { + case 'R': + if (last) { + return -1; + } + goto MAX_FOR; + case 'r': + if (last) { + return -1; + } + goto MAX_FOR; + default: + return -1; + } + +MAX_FOR: + NEXT_CHAR(); + switch (ch) { + case 'W': + if (last) { + return -1; + } + goto MAX_FORW; + case 'w': + if (last) { + return -1; + } + goto MAX_FORW; + default: + return -1; + } + +MAX_FORW: + NEXT_CHAR(); + switch (ch) { + case 'A': + if (last) { + return -1; + } + goto MAX_FORWA; + case 'a': + if (last) { + return -1; + } + goto MAX_FORWA; + default: + return -1; + } + +MAX_FORWA: + NEXT_CHAR(); + switch (ch) { + case 'R': + if (last) { + return -1; + } + goto MAX_FORWAR; + case 'r': + if (last) { + return -1; + } + goto MAX_FORWAR; + default: + return -1; + } + +MAX_FORWAR: + NEXT_CHAR(); + switch (ch) { + case 'D': + if (last) { + return -1; + } + goto MAX_FORWARD; + case 'd': + if (last) { + return -1; + } + goto MAX_FORWARD; + default: + return -1; + } + +MAX_FORWARD: + NEXT_CHAR(); + switch (ch) { + case 'S': + if (last) { + return 47; + } + goto MAX_FORWARDS; + case 's': + if (last) { + return 47; + } + goto MAX_FORWARDS; + default: + return -1; + } + +O: + NEXT_CHAR(); + switch (ch) { + case 'R': + if (last) { + return -1; + } + goto OR; + case 'r': + if (last) { + return -1; + } + goto OR; + default: + return -1; + } + +OR: + NEXT_CHAR(); + switch (ch) { + case 'I': + if (last) { + return -1; + } + goto ORI; + case 'i': + if (last) { + return -1; + } + goto ORI; + default: + return -1; + } + +ORI: + NEXT_CHAR(); + switch (ch) { + case 'G': + if (last) { + return -1; + } + goto ORIG; + case 'g': + if (last) { + return -1; + } + goto ORIG; + default: + return -1; + } + +ORIG: + NEXT_CHAR(); + switch (ch) { + case 'I': + if (last) { + return -1; + } + goto ORIGI; + case 'i': + if (last) { + return -1; + } + goto ORIGI; + default: + return -1; + } + +ORIGI: + NEXT_CHAR(); + switch (ch) { + case 'N': + if (last) { + return 48; + } + goto ORIGIN; + case 'n': + if (last) { + return 48; + } + goto ORIGIN; + default: + return -1; + } + +P: + NEXT_CHAR(); + switch (ch) { + case 'R': + if (last) { + return -1; + } + goto PR; + case 'r': + if (last) { + return -1; + } + goto PR; + default: + return -1; + } + +PR: + NEXT_CHAR(); + switch (ch) { + case 'A': + if (last) { + return -1; + } + goto PRA; + case 'a': + if (last) { + return -1; + } + goto PRA; + case 'O': + if (last) { + return -1; + } + goto PRO; + case 'o': + if (last) { + return -1; + } + goto PRO; + default: + return -1; + } + +PRA: + NEXT_CHAR(); + switch (ch) { + case 'G': + if (last) { + return -1; + } + goto PRAG; + case 'g': + if (last) { + return -1; + } + goto PRAG; + default: + return -1; + } + +PRAG: + NEXT_CHAR(); + switch (ch) { + case 'M': + if (last) { + return -1; + } + goto PRAGM; + case 'm': + if (last) { + return -1; + } + goto PRAGM; + default: + return -1; + } + +PRAGM: + NEXT_CHAR(); + switch (ch) { + case 'A': + if (last) { + return 49; + } + goto PRAGMA; + case 'a': + if (last) { + return 49; + } + goto PRAGMA; + default: + return -1; + } + +PRO: + NEXT_CHAR(); + switch (ch) { + case 'X': + if (last) { + return -1; + } + goto PROX; + case 'x': + if (last) { + return -1; + } + goto PROX; + default: + return -1; + } + +PROX: + NEXT_CHAR(); + switch (ch) { + case 'Y': + if (last) { + return -1; + } + goto PROXY; + case 'y': + if (last) { + return -1; + } + goto PROXY; + default: + return -1; + } + +PROXY: + NEXT_CHAR(); + switch (ch) { + case '-': + if (last) { + return -1; + } + goto PROXY_; + default: + return -1; + } + +PROXY_: + NEXT_CHAR(); + switch (ch) { + case 'A': + if (last) { + return -1; + } + goto PROXY_A; + case 'a': + if (last) { + return -1; + } + goto PROXY_A; + default: + return -1; + } + +PROXY_A: + NEXT_CHAR(); + switch (ch) { + case 'U': + if (last) { + return -1; + } + goto PROXY_AU; + case 'u': + if (last) { + return -1; + } + goto PROXY_AU; + default: + return -1; + } + +PROXY_AU: + NEXT_CHAR(); + switch (ch) { + case 'T': + if (last) { + return -1; + } + goto PROXY_AUT; + case 't': + if (last) { + return -1; + } + goto PROXY_AUT; + default: + return -1; + } + +PROXY_AUT: + NEXT_CHAR(); + switch (ch) { + case 'H': + if (last) { + return -1; + } + goto PROXY_AUTH; + case 'h': + if (last) { + return -1; + } + goto PROXY_AUTH; + default: + return -1; + } + +PROXY_AUTH: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto PROXY_AUTHE; + case 'e': + if (last) { + return -1; + } + goto PROXY_AUTHE; + case 'O': + if (last) { + return -1; + } + goto PROXY_AUTHO; + case 'o': + if (last) { + return -1; + } + goto PROXY_AUTHO; + default: + return -1; + } + +PROXY_AUTHE: + NEXT_CHAR(); + switch (ch) { + case 'N': + if (last) { + return -1; + } + goto PROXY_AUTHEN; + case 'n': + if (last) { + return -1; + } + goto PROXY_AUTHEN; + default: + return -1; + } + +PROXY_AUTHEN: + NEXT_CHAR(); + switch (ch) { + case 'T': + if (last) { + return -1; + } + goto PROXY_AUTHENT; + case 't': + if (last) { + return -1; + } + goto PROXY_AUTHENT; + default: + return -1; + } + +PROXY_AUTHENT: + NEXT_CHAR(); + switch (ch) { + case 'I': + if (last) { + return -1; + } + goto PROXY_AUTHENTI; + case 'i': + if (last) { + return -1; + } + goto PROXY_AUTHENTI; + default: + return -1; + } + +PROXY_AUTHENTI: + NEXT_CHAR(); + switch (ch) { + case 'C': + if (last) { + return -1; + } + goto PROXY_AUTHENTIC; + case 'c': + if (last) { + return -1; + } + goto PROXY_AUTHENTIC; + default: + return -1; + } + +PROXY_AUTHENTIC: + NEXT_CHAR(); + switch (ch) { + case 'A': + if (last) { + return -1; + } + goto PROXY_AUTHENTICA; + case 'a': + if (last) { + return -1; + } + goto PROXY_AUTHENTICA; + default: + return -1; + } + +PROXY_AUTHENTICA: + NEXT_CHAR(); + switch (ch) { + case 'T': + if (last) { + return -1; + } + goto PROXY_AUTHENTICAT; + case 't': + if (last) { + return -1; + } + goto PROXY_AUTHENTICAT; + default: + return -1; + } + +PROXY_AUTHENTICAT: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return 50; + } + goto PROXY_AUTHENTICATE; + case 'e': + if (last) { + return 50; + } + goto PROXY_AUTHENTICATE; + default: + return -1; + } + +PROXY_AUTHO: + NEXT_CHAR(); + switch (ch) { + case 'R': + if (last) { + return -1; + } + goto PROXY_AUTHOR; + case 'r': + if (last) { + return -1; + } + goto PROXY_AUTHOR; + default: + return -1; + } + +PROXY_AUTHOR: + NEXT_CHAR(); + switch (ch) { + case 'I': + if (last) { + return -1; + } + goto PROXY_AUTHORI; + case 'i': + if (last) { + return -1; + } + goto PROXY_AUTHORI; + default: + return -1; + } + +PROXY_AUTHORI: + NEXT_CHAR(); + switch (ch) { + case 'Z': + if (last) { + return -1; + } + goto PROXY_AUTHORIZ; + case 'z': + if (last) { + return -1; + } + goto PROXY_AUTHORIZ; + default: + return -1; + } + +PROXY_AUTHORIZ: + NEXT_CHAR(); + switch (ch) { + case 'A': + if (last) { + return -1; + } + goto PROXY_AUTHORIZA; + case 'a': + if (last) { + return -1; + } + goto PROXY_AUTHORIZA; + default: + return -1; + } + +PROXY_AUTHORIZA: + NEXT_CHAR(); + switch (ch) { + case 'T': + if (last) { + return -1; + } + goto PROXY_AUTHORIZAT; + case 't': + if (last) { + return -1; + } + goto PROXY_AUTHORIZAT; + default: + return -1; + } + +PROXY_AUTHORIZAT: + NEXT_CHAR(); + switch (ch) { + case 'I': + if (last) { + return -1; + } + goto PROXY_AUTHORIZATI; + case 'i': + if (last) { + return -1; + } + goto PROXY_AUTHORIZATI; + default: + return -1; + } + +PROXY_AUTHORIZATI: + NEXT_CHAR(); + switch (ch) { + case 'O': + if (last) { + return -1; + } + goto PROXY_AUTHORIZATIO; + case 'o': + if (last) { + return -1; + } + goto PROXY_AUTHORIZATIO; + default: + return -1; + } + +PROXY_AUTHORIZATIO: + NEXT_CHAR(); + switch (ch) { + case 'N': + if (last) { + return 51; + } + goto PROXY_AUTHORIZATION; + case 'n': + if (last) { + return 51; + } + goto PROXY_AUTHORIZATION; + default: + return -1; + } + +R: + NEXT_CHAR(); + switch (ch) { + case 'A': + if (last) { + return -1; + } + goto RA; + case 'a': + if (last) { + return -1; + } + goto RA; + case 'E': + if (last) { + return -1; + } + goto RE; + case 'e': + if (last) { + return -1; + } + goto RE; + default: + return -1; + } + +RA: + NEXT_CHAR(); + switch (ch) { + case 'N': + if (last) { + return -1; + } + goto RAN; + case 'n': + if (last) { + return -1; + } + goto RAN; + default: + return -1; + } + +RAN: + NEXT_CHAR(); + switch (ch) { + case 'G': + if (last) { + return -1; + } + goto RANG; + case 'g': + if (last) { + return -1; + } + goto RANG; + default: + return -1; + } + +RANG: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return 52; + } + goto RANGE; + case 'e': + if (last) { + return 52; + } + goto RANGE; + default: + return -1; + } + +RE: + NEXT_CHAR(); + switch (ch) { + case 'F': + if (last) { + return -1; + } + goto REF; + case 'f': + if (last) { + return -1; + } + goto REF; + case 'T': + if (last) { + return -1; + } + goto RET; + case 't': + if (last) { + return -1; + } + goto RET; + default: + return -1; + } + +REF: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto REFE; + case 'e': + if (last) { + return -1; + } + goto REFE; + default: + return -1; + } + +REFE: + NEXT_CHAR(); + switch (ch) { + case 'R': + if (last) { + return -1; + } + goto REFER; + case 'r': + if (last) { + return -1; + } + goto REFER; + default: + return -1; + } + +REFER: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto REFERE; + case 'e': + if (last) { + return -1; + } + goto REFERE; + default: + return -1; + } + +REFERE: + NEXT_CHAR(); + switch (ch) { + case 'R': + if (last) { + return 53; + } + goto REFERER; + case 'r': + if (last) { + return 53; + } + goto REFERER; + default: + return -1; + } + +RET: + NEXT_CHAR(); + switch (ch) { + case 'R': + if (last) { + return -1; + } + goto RETR; + case 'r': + if (last) { + return -1; + } + goto RETR; + default: + return -1; + } + +RETR: + NEXT_CHAR(); + switch (ch) { + case 'Y': + if (last) { + return -1; + } + goto RETRY; + case 'y': + if (last) { + return -1; + } + goto RETRY; + default: + return -1; + } + +RETRY: + NEXT_CHAR(); + switch (ch) { + case '-': + if (last) { + return -1; + } + goto RETRY_; + default: + return -1; + } + +RETRY_: + NEXT_CHAR(); + switch (ch) { + case 'A': + if (last) { + return -1; + } + goto RETRY_A; + case 'a': + if (last) { + return -1; + } + goto RETRY_A; + default: + return -1; + } + +RETRY_A: + NEXT_CHAR(); + switch (ch) { + case 'F': + if (last) { + return -1; + } + goto RETRY_AF; + case 'f': + if (last) { + return -1; + } + goto RETRY_AF; + default: + return -1; + } + +RETRY_AF: + NEXT_CHAR(); + switch (ch) { + case 'T': + if (last) { + return -1; + } + goto RETRY_AFT; + case 't': + if (last) { + return -1; + } + goto RETRY_AFT; + default: + return -1; + } + +RETRY_AFT: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto RETRY_AFTE; + case 'e': + if (last) { + return -1; + } + goto RETRY_AFTE; + default: + return -1; + } + +RETRY_AFTE: + NEXT_CHAR(); + switch (ch) { + case 'R': + if (last) { + return 54; + } + goto RETRY_AFTER; + case 'r': + if (last) { + return 54; + } + goto RETRY_AFTER; + default: + return -1; + } + +S: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto SE; + case 'e': + if (last) { + return -1; + } + goto SE; + default: + return -1; + } + +SE: + NEXT_CHAR(); + switch (ch) { + case 'C': + if (last) { + return -1; + } + goto SEC; + case 'c': + if (last) { + return -1; + } + goto SEC; + case 'R': + if (last) { + return -1; + } + goto SER; + case 'r': + if (last) { + return -1; + } + goto SER; + case 'T': + if (last) { + return -1; + } + goto SET; + case 't': + if (last) { + return -1; + } + goto SET; + default: + return -1; + } + +SEC: + NEXT_CHAR(); + switch (ch) { + case '-': + if (last) { + return -1; + } + goto SEC_; + default: + return -1; + } + +SEC_: + NEXT_CHAR(); + switch (ch) { + case 'W': + if (last) { + return -1; + } + goto SEC_W; + case 'w': + if (last) { + return -1; + } + goto SEC_W; + default: + return -1; + } + +SEC_W: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto SEC_WE; + case 'e': + if (last) { + return -1; + } + goto SEC_WE; + default: + return -1; + } + +SEC_WE: + NEXT_CHAR(); + switch (ch) { + case 'B': + if (last) { + return -1; + } + goto SEC_WEB; + case 'b': + if (last) { + return -1; + } + goto SEC_WEB; + default: + return -1; + } + +SEC_WEB: + NEXT_CHAR(); + switch (ch) { + case 'S': + if (last) { + return -1; + } + goto SEC_WEBS; + case 's': + if (last) { + return -1; + } + goto SEC_WEBS; + default: + return -1; + } + +SEC_WEBS: + NEXT_CHAR(); + switch (ch) { + case 'O': + if (last) { + return -1; + } + goto SEC_WEBSO; + case 'o': + if (last) { + return -1; + } + goto SEC_WEBSO; + default: + return -1; + } + +SEC_WEBSO: + NEXT_CHAR(); + switch (ch) { + case 'C': + if (last) { + return -1; + } + goto SEC_WEBSOC; + case 'c': + if (last) { + return -1; + } + goto SEC_WEBSOC; + default: + return -1; + } + +SEC_WEBSOC: + NEXT_CHAR(); + switch (ch) { + case 'K': + if (last) { + return -1; + } + goto SEC_WEBSOCK; + case 'k': + if (last) { + return -1; + } + goto SEC_WEBSOCK; + default: + return -1; + } + +SEC_WEBSOCK: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto SEC_WEBSOCKE; + case 'e': + if (last) { + return -1; + } + goto SEC_WEBSOCKE; + default: + return -1; + } + +SEC_WEBSOCKE: + NEXT_CHAR(); + switch (ch) { + case 'T': + if (last) { + return -1; + } + goto SEC_WEBSOCKET; + case 't': + if (last) { + return -1; + } + goto SEC_WEBSOCKET; + default: + return -1; + } + +SEC_WEBSOCKET: + NEXT_CHAR(); + switch (ch) { + case '-': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_; + default: + return -1; + } + +SEC_WEBSOCKET_: + NEXT_CHAR(); + switch (ch) { + case 'A': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_A; + case 'a': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_A; + case 'E': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_E; + case 'e': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_E; + case 'K': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_K; + case 'k': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_K; + case 'P': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_P; + case 'p': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_P; + case 'V': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_V; + case 'v': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_V; + default: + return -1; + } + +SEC_WEBSOCKET_A: + NEXT_CHAR(); + switch (ch) { + case 'C': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_AC; + case 'c': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_AC; + default: + return -1; + } + +SEC_WEBSOCKET_AC: + NEXT_CHAR(); + switch (ch) { + case 'C': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_ACC; + case 'c': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_ACC; + default: + return -1; + } + +SEC_WEBSOCKET_ACC: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_ACCE; + case 'e': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_ACCE; + default: + return -1; + } + +SEC_WEBSOCKET_ACCE: + NEXT_CHAR(); + switch (ch) { + case 'P': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_ACCEP; + case 'p': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_ACCEP; + default: + return -1; + } + +SEC_WEBSOCKET_ACCEP: + NEXT_CHAR(); + switch (ch) { + case 'T': + if (last) { + return 55; + } + goto SEC_WEBSOCKET_ACCEPT; + case 't': + if (last) { + return 55; + } + goto SEC_WEBSOCKET_ACCEPT; + default: + return -1; + } + +SEC_WEBSOCKET_E: + NEXT_CHAR(); + switch (ch) { + case 'X': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_EX; + case 'x': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_EX; + default: + return -1; + } + +SEC_WEBSOCKET_EX: + NEXT_CHAR(); + switch (ch) { + case 'T': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_EXT; + case 't': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_EXT; + default: + return -1; + } + +SEC_WEBSOCKET_EXT: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_EXTE; + case 'e': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_EXTE; + default: + return -1; + } + +SEC_WEBSOCKET_EXTE: + NEXT_CHAR(); + switch (ch) { + case 'N': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_EXTEN; + case 'n': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_EXTEN; + default: + return -1; + } + +SEC_WEBSOCKET_EXTEN: + NEXT_CHAR(); + switch (ch) { + case 'S': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_EXTENS; + case 's': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_EXTENS; + default: + return -1; + } + +SEC_WEBSOCKET_EXTENS: + NEXT_CHAR(); + switch (ch) { + case 'I': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_EXTENSI; + case 'i': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_EXTENSI; + default: + return -1; + } + +SEC_WEBSOCKET_EXTENSI: + NEXT_CHAR(); + switch (ch) { + case 'O': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_EXTENSIO; + case 'o': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_EXTENSIO; + default: + return -1; + } + +SEC_WEBSOCKET_EXTENSIO: + NEXT_CHAR(); + switch (ch) { + case 'N': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_EXTENSION; + case 'n': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_EXTENSION; + default: + return -1; + } + +SEC_WEBSOCKET_EXTENSION: + NEXT_CHAR(); + switch (ch) { + case 'S': + if (last) { + return 56; + } + goto SEC_WEBSOCKET_EXTENSIONS; + case 's': + if (last) { + return 56; + } + goto SEC_WEBSOCKET_EXTENSIONS; + default: + return -1; + } + +SEC_WEBSOCKET_K: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_KE; + case 'e': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_KE; + default: + return -1; + } + +SEC_WEBSOCKET_KE: + NEXT_CHAR(); + switch (ch) { + case 'Y': + if (last) { + return 57; + } + goto SEC_WEBSOCKET_KEY; + case 'y': + if (last) { + return 57; + } + goto SEC_WEBSOCKET_KEY; + default: + return -1; + } + +SEC_WEBSOCKET_KEY: + NEXT_CHAR(); + switch (ch) { + case '1': + if (last) { + return 58; + } + goto SEC_WEBSOCKET_KEY1; + default: + return -1; + } + +SEC_WEBSOCKET_P: + NEXT_CHAR(); + switch (ch) { + case 'R': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_PR; + case 'r': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_PR; + default: + return -1; + } + +SEC_WEBSOCKET_PR: + NEXT_CHAR(); + switch (ch) { + case 'O': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_PRO; + case 'o': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_PRO; + default: + return -1; + } + +SEC_WEBSOCKET_PRO: + NEXT_CHAR(); + switch (ch) { + case 'T': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_PROT; + case 't': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_PROT; + default: + return -1; + } + +SEC_WEBSOCKET_PROT: + NEXT_CHAR(); + switch (ch) { + case 'O': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_PROTO; + case 'o': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_PROTO; + default: + return -1; + } + +SEC_WEBSOCKET_PROTO: + NEXT_CHAR(); + switch (ch) { + case 'C': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_PROTOC; + case 'c': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_PROTOC; + default: + return -1; + } + +SEC_WEBSOCKET_PROTOC: + NEXT_CHAR(); + switch (ch) { + case 'O': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_PROTOCO; + case 'o': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_PROTOCO; + default: + return -1; + } + +SEC_WEBSOCKET_PROTOCO: + NEXT_CHAR(); + switch (ch) { + case 'L': + if (last) { + return 59; + } + goto SEC_WEBSOCKET_PROTOCOL; + case 'l': + if (last) { + return 59; + } + goto SEC_WEBSOCKET_PROTOCOL; + default: + return -1; + } + +SEC_WEBSOCKET_V: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_VE; + case 'e': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_VE; + default: + return -1; + } + +SEC_WEBSOCKET_VE: + NEXT_CHAR(); + switch (ch) { + case 'R': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_VER; + case 'r': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_VER; + default: + return -1; + } + +SEC_WEBSOCKET_VER: + NEXT_CHAR(); + switch (ch) { + case 'S': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_VERS; + case 's': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_VERS; + default: + return -1; + } + +SEC_WEBSOCKET_VERS: + NEXT_CHAR(); + switch (ch) { + case 'I': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_VERSI; + case 'i': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_VERSI; + default: + return -1; + } + +SEC_WEBSOCKET_VERSI: + NEXT_CHAR(); + switch (ch) { + case 'O': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_VERSIO; + case 'o': + if (last) { + return -1; + } + goto SEC_WEBSOCKET_VERSIO; + default: + return -1; + } + +SEC_WEBSOCKET_VERSIO: + NEXT_CHAR(); + switch (ch) { + case 'N': + if (last) { + return 60; + } + goto SEC_WEBSOCKET_VERSION; + case 'n': + if (last) { + return 60; + } + goto SEC_WEBSOCKET_VERSION; + default: + return -1; + } + +SER: + NEXT_CHAR(); + switch (ch) { + case 'V': + if (last) { + return -1; + } + goto SERV; + case 'v': + if (last) { + return -1; + } + goto SERV; + default: + return -1; + } + +SERV: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto SERVE; + case 'e': + if (last) { + return -1; + } + goto SERVE; + default: + return -1; + } + +SERVE: + NEXT_CHAR(); + switch (ch) { + case 'R': + if (last) { + return 61; + } + goto SERVER; + case 'r': + if (last) { + return 61; + } + goto SERVER; + default: + return -1; + } + +SET: + NEXT_CHAR(); + switch (ch) { + case '-': + if (last) { + return -1; + } + goto SET_; + default: + return -1; + } + +SET_: + NEXT_CHAR(); + switch (ch) { + case 'C': + if (last) { + return -1; + } + goto SET_C; + case 'c': + if (last) { + return -1; + } + goto SET_C; + default: + return -1; + } + +SET_C: + NEXT_CHAR(); + switch (ch) { + case 'O': + if (last) { + return -1; + } + goto SET_CO; + case 'o': + if (last) { + return -1; + } + goto SET_CO; + default: + return -1; + } + +SET_CO: + NEXT_CHAR(); + switch (ch) { + case 'O': + if (last) { + return -1; + } + goto SET_COO; + case 'o': + if (last) { + return -1; + } + goto SET_COO; + default: + return -1; + } + +SET_COO: + NEXT_CHAR(); + switch (ch) { + case 'K': + if (last) { + return -1; + } + goto SET_COOK; + case 'k': + if (last) { + return -1; + } + goto SET_COOK; + default: + return -1; + } + +SET_COOK: + NEXT_CHAR(); + switch (ch) { + case 'I': + if (last) { + return -1; + } + goto SET_COOKI; + case 'i': + if (last) { + return -1; + } + goto SET_COOKI; + default: + return -1; + } + +SET_COOKI: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return 62; + } + goto SET_COOKIE; + case 'e': + if (last) { + return 62; + } + goto SET_COOKIE; + default: + return -1; + } + +T: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return 63; + } + goto TE; + case 'e': + if (last) { + return 63; + } + goto TE; + case 'R': + if (last) { + return -1; + } + goto TR; + case 'r': + if (last) { + return -1; + } + goto TR; + default: + return -1; + } + +TR: + NEXT_CHAR(); + switch (ch) { + case 'A': + if (last) { + return -1; + } + goto TRA; + case 'a': + if (last) { + return -1; + } + goto TRA; + default: + return -1; + } + +TRA: + NEXT_CHAR(); + switch (ch) { + case 'I': + if (last) { + return -1; + } + goto TRAI; + case 'i': + if (last) { + return -1; + } + goto TRAI; + case 'N': + if (last) { + return -1; + } + goto TRAN; + case 'n': + if (last) { + return -1; + } + goto TRAN; + default: + return -1; + } + +TRAI: + NEXT_CHAR(); + switch (ch) { + case 'L': + if (last) { + return -1; + } + goto TRAIL; + case 'l': + if (last) { + return -1; + } + goto TRAIL; + default: + return -1; + } + +TRAIL: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto TRAILE; + case 'e': + if (last) { + return -1; + } + goto TRAILE; + default: + return -1; + } + +TRAILE: + NEXT_CHAR(); + switch (ch) { + case 'R': + if (last) { + return 64; + } + goto TRAILER; + case 'r': + if (last) { + return 64; + } + goto TRAILER; + default: + return -1; + } + +TRAN: + NEXT_CHAR(); + switch (ch) { + case 'S': + if (last) { + return -1; + } + goto TRANS; + case 's': + if (last) { + return -1; + } + goto TRANS; + default: + return -1; + } + +TRANS: + NEXT_CHAR(); + switch (ch) { + case 'F': + if (last) { + return -1; + } + goto TRANSF; + case 'f': + if (last) { + return -1; + } + goto TRANSF; + default: + return -1; + } + +TRANSF: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto TRANSFE; + case 'e': + if (last) { + return -1; + } + goto TRANSFE; + default: + return -1; + } + +TRANSFE: + NEXT_CHAR(); + switch (ch) { + case 'R': + if (last) { + return -1; + } + goto TRANSFER; + case 'r': + if (last) { + return -1; + } + goto TRANSFER; + default: + return -1; + } + +TRANSFER: + NEXT_CHAR(); + switch (ch) { + case '-': + if (last) { + return -1; + } + goto TRANSFER_; + default: + return -1; + } + +TRANSFER_: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto TRANSFER_E; + case 'e': + if (last) { + return -1; + } + goto TRANSFER_E; + default: + return -1; + } + +TRANSFER_E: + NEXT_CHAR(); + switch (ch) { + case 'N': + if (last) { + return -1; + } + goto TRANSFER_EN; + case 'n': + if (last) { + return -1; + } + goto TRANSFER_EN; + default: + return -1; + } + +TRANSFER_EN: + NEXT_CHAR(); + switch (ch) { + case 'C': + if (last) { + return -1; + } + goto TRANSFER_ENC; + case 'c': + if (last) { + return -1; + } + goto TRANSFER_ENC; + default: + return -1; + } + +TRANSFER_ENC: + NEXT_CHAR(); + switch (ch) { + case 'O': + if (last) { + return -1; + } + goto TRANSFER_ENCO; + case 'o': + if (last) { + return -1; + } + goto TRANSFER_ENCO; + default: + return -1; + } + +TRANSFER_ENCO: + NEXT_CHAR(); + switch (ch) { + case 'D': + if (last) { + return -1; + } + goto TRANSFER_ENCOD; + case 'd': + if (last) { + return -1; + } + goto TRANSFER_ENCOD; + default: + return -1; + } + +TRANSFER_ENCOD: + NEXT_CHAR(); + switch (ch) { + case 'I': + if (last) { + return -1; + } + goto TRANSFER_ENCODI; + case 'i': + if (last) { + return -1; + } + goto TRANSFER_ENCODI; + default: + return -1; + } + +TRANSFER_ENCODI: + NEXT_CHAR(); + switch (ch) { + case 'N': + if (last) { + return -1; + } + goto TRANSFER_ENCODIN; + case 'n': + if (last) { + return -1; + } + goto TRANSFER_ENCODIN; + default: + return -1; + } + +TRANSFER_ENCODIN: + NEXT_CHAR(); + switch (ch) { + case 'G': + if (last) { + return 65; + } + goto TRANSFER_ENCODING; + case 'g': + if (last) { + return 65; + } + goto TRANSFER_ENCODING; + default: + return -1; + } + +U: + NEXT_CHAR(); + switch (ch) { + case 'R': + if (last) { + return -1; + } + goto UR; + case 'r': + if (last) { + return -1; + } + goto UR; + case 'P': + if (last) { + return -1; + } + goto UP; + case 'p': + if (last) { + return -1; + } + goto UP; + case 'S': + if (last) { + return -1; + } + goto US; + case 's': + if (last) { + return -1; + } + goto US; + default: + return -1; + } + +UR: + NEXT_CHAR(); + switch (ch) { + case 'I': + if (last) { + return 66; + } + goto URI; + case 'i': + if (last) { + return 66; + } + goto URI; + default: + return -1; + } + +UP: + NEXT_CHAR(); + switch (ch) { + case 'G': + if (last) { + return -1; + } + goto UPG; + case 'g': + if (last) { + return -1; + } + goto UPG; + default: + return -1; + } + +UPG: + NEXT_CHAR(); + switch (ch) { + case 'R': + if (last) { + return -1; + } + goto UPGR; + case 'r': + if (last) { + return -1; + } + goto UPGR; + default: + return -1; + } + +UPGR: + NEXT_CHAR(); + switch (ch) { + case 'A': + if (last) { + return -1; + } + goto UPGRA; + case 'a': + if (last) { + return -1; + } + goto UPGRA; + default: + return -1; + } + +UPGRA: + NEXT_CHAR(); + switch (ch) { + case 'D': + if (last) { + return -1; + } + goto UPGRAD; + case 'd': + if (last) { + return -1; + } + goto UPGRAD; + default: + return -1; + } + +UPGRAD: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return 67; + } + goto UPGRADE; + case 'e': + if (last) { + return 67; + } + goto UPGRADE; + default: + return -1; + } + +US: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto USE; + case 'e': + if (last) { + return -1; + } + goto USE; + default: + return -1; + } + +USE: + NEXT_CHAR(); + switch (ch) { + case 'R': + if (last) { + return -1; + } + goto USER; + case 'r': + if (last) { + return -1; + } + goto USER; + default: + return -1; + } + +USER: + NEXT_CHAR(); + switch (ch) { + case '-': + if (last) { + return -1; + } + goto USER_; + default: + return -1; + } + +USER_: + NEXT_CHAR(); + switch (ch) { + case 'A': + if (last) { + return -1; + } + goto USER_A; + case 'a': + if (last) { + return -1; + } + goto USER_A; + default: + return -1; + } + +USER_A: + NEXT_CHAR(); + switch (ch) { + case 'G': + if (last) { + return -1; + } + goto USER_AG; + case 'g': + if (last) { + return -1; + } + goto USER_AG; + default: + return -1; + } + +USER_AG: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto USER_AGE; + case 'e': + if (last) { + return -1; + } + goto USER_AGE; + default: + return -1; + } + +USER_AGE: + NEXT_CHAR(); + switch (ch) { + case 'N': + if (last) { + return -1; + } + goto USER_AGEN; + case 'n': + if (last) { + return -1; + } + goto USER_AGEN; + default: + return -1; + } + +USER_AGEN: + NEXT_CHAR(); + switch (ch) { + case 'T': + if (last) { + return 68; + } + goto USER_AGENT; + case 't': + if (last) { + return 68; + } + goto USER_AGENT; + default: + return -1; + } + +V: + NEXT_CHAR(); + switch (ch) { + case 'A': + if (last) { + return -1; + } + goto VA; + case 'a': + if (last) { + return -1; + } + goto VA; + case 'I': + if (last) { + return -1; + } + goto VI; + case 'i': + if (last) { + return -1; + } + goto VI; + default: + return -1; + } + +VA: + NEXT_CHAR(); + switch (ch) { + case 'R': + if (last) { + return -1; + } + goto VAR; + case 'r': + if (last) { + return -1; + } + goto VAR; + default: + return -1; + } + +VAR: + NEXT_CHAR(); + switch (ch) { + case 'Y': + if (last) { + return 69; + } + goto VARY; + case 'y': + if (last) { + return 69; + } + goto VARY; + default: + return -1; + } + +VI: + NEXT_CHAR(); + switch (ch) { + case 'A': + if (last) { + return 70; + } + goto VIA; + case 'a': + if (last) { + return 70; + } + goto VIA; + default: + return -1; + } + +W: + NEXT_CHAR(); + switch (ch) { + case 'W': + if (last) { + return -1; + } + goto WW; + case 'w': + if (last) { + return -1; + } + goto WW; + case 'A': + if (last) { + return -1; + } + goto WA; + case 'a': + if (last) { + return -1; + } + goto WA; + default: + return -1; + } + +WW: + NEXT_CHAR(); + switch (ch) { + case 'W': + if (last) { + return -1; + } + goto WWW; + case 'w': + if (last) { + return -1; + } + goto WWW; + default: + return -1; + } + +WWW: + NEXT_CHAR(); + switch (ch) { + case '-': + if (last) { + return -1; + } + goto WWW_; + default: + return -1; + } + +WWW_: + NEXT_CHAR(); + switch (ch) { + case 'A': + if (last) { + return -1; + } + goto WWW_A; + case 'a': + if (last) { + return -1; + } + goto WWW_A; + default: + return -1; + } + +WWW_A: + NEXT_CHAR(); + switch (ch) { + case 'U': + if (last) { + return -1; + } + goto WWW_AU; + case 'u': + if (last) { + return -1; + } + goto WWW_AU; + default: + return -1; + } + +WWW_AU: + NEXT_CHAR(); + switch (ch) { + case 'T': + if (last) { + return -1; + } + goto WWW_AUT; + case 't': + if (last) { + return -1; + } + goto WWW_AUT; + default: + return -1; + } + +WWW_AUT: + NEXT_CHAR(); + switch (ch) { + case 'H': + if (last) { + return -1; + } + goto WWW_AUTH; + case 'h': + if (last) { + return -1; + } + goto WWW_AUTH; + default: + return -1; + } + +WWW_AUTH: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto WWW_AUTHE; + case 'e': + if (last) { + return -1; + } + goto WWW_AUTHE; + default: + return -1; + } + +WWW_AUTHE: + NEXT_CHAR(); + switch (ch) { + case 'N': + if (last) { + return -1; + } + goto WWW_AUTHEN; + case 'n': + if (last) { + return -1; + } + goto WWW_AUTHEN; + default: + return -1; + } + +WWW_AUTHEN: + NEXT_CHAR(); + switch (ch) { + case 'T': + if (last) { + return -1; + } + goto WWW_AUTHENT; + case 't': + if (last) { + return -1; + } + goto WWW_AUTHENT; + default: + return -1; + } + +WWW_AUTHENT: + NEXT_CHAR(); + switch (ch) { + case 'I': + if (last) { + return -1; + } + goto WWW_AUTHENTI; + case 'i': + if (last) { + return -1; + } + goto WWW_AUTHENTI; + default: + return -1; + } + +WWW_AUTHENTI: + NEXT_CHAR(); + switch (ch) { + case 'C': + if (last) { + return -1; + } + goto WWW_AUTHENTIC; + case 'c': + if (last) { + return -1; + } + goto WWW_AUTHENTIC; + default: + return -1; + } + +WWW_AUTHENTIC: + NEXT_CHAR(); + switch (ch) { + case 'A': + if (last) { + return -1; + } + goto WWW_AUTHENTICA; + case 'a': + if (last) { + return -1; + } + goto WWW_AUTHENTICA; + default: + return -1; + } + +WWW_AUTHENTICA: + NEXT_CHAR(); + switch (ch) { + case 'T': + if (last) { + return -1; + } + goto WWW_AUTHENTICAT; + case 't': + if (last) { + return -1; + } + goto WWW_AUTHENTICAT; + default: + return -1; + } + +WWW_AUTHENTICAT: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return 71; + } + goto WWW_AUTHENTICATE; + case 'e': + if (last) { + return 71; + } + goto WWW_AUTHENTICATE; + default: + return -1; + } + +WA: + NEXT_CHAR(); + switch (ch) { + case 'N': + if (last) { + return -1; + } + goto WAN; + case 'n': + if (last) { + return -1; + } + goto WAN; + case 'R': + if (last) { + return -1; + } + goto WAR; + case 'r': + if (last) { + return -1; + } + goto WAR; + default: + return -1; + } + +WAN: + NEXT_CHAR(); + switch (ch) { + case 'T': + if (last) { + return -1; + } + goto WANT; + case 't': + if (last) { + return -1; + } + goto WANT; + default: + return -1; + } + +WANT: + NEXT_CHAR(); + switch (ch) { + case '-': + if (last) { + return -1; + } + goto WANT_; + default: + return -1; + } + +WANT_: + NEXT_CHAR(); + switch (ch) { + case 'D': + if (last) { + return -1; + } + goto WANT_D; + case 'd': + if (last) { + return -1; + } + goto WANT_D; + default: + return -1; + } + +WANT_D: + NEXT_CHAR(); + switch (ch) { + case 'I': + if (last) { + return -1; + } + goto WANT_DI; + case 'i': + if (last) { + return -1; + } + goto WANT_DI; + default: + return -1; + } + +WANT_DI: + NEXT_CHAR(); + switch (ch) { + case 'G': + if (last) { + return -1; + } + goto WANT_DIG; + case 'g': + if (last) { + return -1; + } + goto WANT_DIG; + default: + return -1; + } + +WANT_DIG: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto WANT_DIGE; + case 'e': + if (last) { + return -1; + } + goto WANT_DIGE; + default: + return -1; + } + +WANT_DIGE: + NEXT_CHAR(); + switch (ch) { + case 'S': + if (last) { + return -1; + } + goto WANT_DIGES; + case 's': + if (last) { + return -1; + } + goto WANT_DIGES; + default: + return -1; + } + +WANT_DIGES: + NEXT_CHAR(); + switch (ch) { + case 'T': + if (last) { + return 72; + } + goto WANT_DIGEST; + case 't': + if (last) { + return 72; + } + goto WANT_DIGEST; + default: + return -1; + } + +WAR: + NEXT_CHAR(); + switch (ch) { + case 'N': + if (last) { + return -1; + } + goto WARN; + case 'n': + if (last) { + return -1; + } + goto WARN; + default: + return -1; + } + +WARN: + NEXT_CHAR(); + switch (ch) { + case 'I': + if (last) { + return -1; + } + goto WARNI; + case 'i': + if (last) { + return -1; + } + goto WARNI; + default: + return -1; + } + +WARNI: + NEXT_CHAR(); + switch (ch) { + case 'N': + if (last) { + return -1; + } + goto WARNIN; + case 'n': + if (last) { + return -1; + } + goto WARNIN; + default: + return -1; + } + +WARNIN: + NEXT_CHAR(); + switch (ch) { + case 'G': + if (last) { + return 73; + } + goto WARNING; + case 'g': + if (last) { + return 73; + } + goto WARNING; + default: + return -1; + } + +X: + NEXT_CHAR(); + switch (ch) { + case '-': + if (last) { + return -1; + } + goto X_; + default: + return -1; + } + +X_: + NEXT_CHAR(); + switch (ch) { + case 'F': + if (last) { + return -1; + } + goto X_F; + case 'f': + if (last) { + return -1; + } + goto X_F; + default: + return -1; + } + +X_F: + NEXT_CHAR(); + switch (ch) { + case 'O': + if (last) { + return -1; + } + goto X_FO; + case 'o': + if (last) { + return -1; + } + goto X_FO; + default: + return -1; + } + +X_FO: + NEXT_CHAR(); + switch (ch) { + case 'R': + if (last) { + return -1; + } + goto X_FOR; + case 'r': + if (last) { + return -1; + } + goto X_FOR; + default: + return -1; + } + +X_FOR: + NEXT_CHAR(); + switch (ch) { + case 'W': + if (last) { + return -1; + } + goto X_FORW; + case 'w': + if (last) { + return -1; + } + goto X_FORW; + default: + return -1; + } + +X_FORW: + NEXT_CHAR(); + switch (ch) { + case 'A': + if (last) { + return -1; + } + goto X_FORWA; + case 'a': + if (last) { + return -1; + } + goto X_FORWA; + default: + return -1; + } + +X_FORWA: + NEXT_CHAR(); + switch (ch) { + case 'R': + if (last) { + return -1; + } + goto X_FORWAR; + case 'r': + if (last) { + return -1; + } + goto X_FORWAR; + default: + return -1; + } + +X_FORWAR: + NEXT_CHAR(); + switch (ch) { + case 'D': + if (last) { + return -1; + } + goto X_FORWARD; + case 'd': + if (last) { + return -1; + } + goto X_FORWARD; + default: + return -1; + } + +X_FORWARD: + NEXT_CHAR(); + switch (ch) { + case 'E': + if (last) { + return -1; + } + goto X_FORWARDE; + case 'e': + if (last) { + return -1; + } + goto X_FORWARDE; + default: + return -1; + } + +X_FORWARDE: + NEXT_CHAR(); + switch (ch) { + case 'D': + if (last) { + return -1; + } + goto X_FORWARDED; + case 'd': + if (last) { + return -1; + } + goto X_FORWARDED; + default: + return -1; + } + +X_FORWARDED: + NEXT_CHAR(); + switch (ch) { + case '-': + if (last) { + return -1; + } + goto X_FORWARDED_; + default: + return -1; + } + +X_FORWARDED_: + NEXT_CHAR(); + switch (ch) { + case 'F': + if (last) { + return -1; + } + goto X_FORWARDED_F; + case 'f': + if (last) { + return -1; + } + goto X_FORWARDED_F; + case 'H': + if (last) { + return -1; + } + goto X_FORWARDED_H; + case 'h': + if (last) { + return -1; + } + goto X_FORWARDED_H; + case 'P': + if (last) { + return -1; + } + goto X_FORWARDED_P; + case 'p': + if (last) { + return -1; + } + goto X_FORWARDED_P; + default: + return -1; + } + +X_FORWARDED_F: + NEXT_CHAR(); + switch (ch) { + case 'O': + if (last) { + return -1; + } + goto X_FORWARDED_FO; + case 'o': + if (last) { + return -1; + } + goto X_FORWARDED_FO; + default: + return -1; + } + +X_FORWARDED_FO: + NEXT_CHAR(); + switch (ch) { + case 'R': + if (last) { + return 74; + } + goto X_FORWARDED_FOR; + case 'r': + if (last) { + return 74; + } + goto X_FORWARDED_FOR; + default: + return -1; + } + +X_FORWARDED_H: + NEXT_CHAR(); + switch (ch) { + case 'O': + if (last) { + return -1; + } + goto X_FORWARDED_HO; + case 'o': + if (last) { + return -1; + } + goto X_FORWARDED_HO; + default: + return -1; + } + +X_FORWARDED_HO: + NEXT_CHAR(); + switch (ch) { + case 'S': + if (last) { + return -1; + } + goto X_FORWARDED_HOS; + case 's': + if (last) { + return -1; + } + goto X_FORWARDED_HOS; + default: + return -1; + } + +X_FORWARDED_HOS: + NEXT_CHAR(); + switch (ch) { + case 'T': + if (last) { + return 75; + } + goto X_FORWARDED_HOST; + case 't': + if (last) { + return 75; + } + goto X_FORWARDED_HOST; + default: + return -1; + } + +X_FORWARDED_P: + NEXT_CHAR(); + switch (ch) { + case 'R': + if (last) { + return -1; + } + goto X_FORWARDED_PR; + case 'r': + if (last) { + return -1; + } + goto X_FORWARDED_PR; + default: + return -1; + } + +X_FORWARDED_PR: + NEXT_CHAR(); + switch (ch) { + case 'O': + if (last) { + return -1; + } + goto X_FORWARDED_PRO; + case 'o': + if (last) { + return -1; + } + goto X_FORWARDED_PRO; + default: + return -1; + } + +X_FORWARDED_PRO: + NEXT_CHAR(); + switch (ch) { + case 'T': + if (last) { + return -1; + } + goto X_FORWARDED_PROT; + case 't': + if (last) { + return -1; + } + goto X_FORWARDED_PROT; + default: + return -1; + } + +X_FORWARDED_PROT: + NEXT_CHAR(); + switch (ch) { + case 'O': + if (last) { + return 76; + } + goto X_FORWARDED_PROTO; + case 'o': + if (last) { + return 76; + } + goto X_FORWARDED_PROTO; + default: + return -1; + } + +ACCEPT_CHARSET: +ACCEPT_ENCODING: +ACCEPT_LANGUAGE: +ACCEPT_RANGES: +ACCESS_CONTROL_ALLOW_CREDENTIALS: +ACCESS_CONTROL_ALLOW_HEADERS: +ACCESS_CONTROL_ALLOW_METHODS: +ACCESS_CONTROL_ALLOW_ORIGIN: +ACCESS_CONTROL_EXPOSE_HEADERS: +ACCESS_CONTROL_MAX_AGE: +ACCESS_CONTROL_REQUEST_HEADERS: +ACCESS_CONTROL_REQUEST_METHOD: +AGE: +ALLOW: +AUTHORIZATION: +CACHE_CONTROL: +CONNECTION: +CONTENT_DISPOSITION: +CONTENT_ENCODING: +CONTENT_LANGUAGE: +CONTENT_LENGTH: +CONTENT_LOCATION: +CONTENT_MD5: +CONTENT_RANGE: +CONTENT_TRANSFER_ENCODING: +CONTENT_TYPE: +COOKIE: +DATE: +DESTINATION: +DIGEST: +ETAG: +EXPECT: +EXPIRES: +FORWARDED: +FROM: +HOST: +IF_MATCH: +IF_MODIFIED_SINCE: +IF_NONE_MATCH: +IF_RANGE: +IF_UNMODIFIED_SINCE: +KEEP_ALIVE: +LAST_EVENT_ID: +LAST_MODIFIED: +LINK: +LOCATION: +MAX_FORWARDS: +ORIGIN: +PRAGMA: +PROXY_AUTHENTICATE: +PROXY_AUTHORIZATION: +RANGE: +REFERER: +RETRY_AFTER: +SEC_WEBSOCKET_ACCEPT: +SEC_WEBSOCKET_EXTENSIONS: +SEC_WEBSOCKET_KEY1: +SEC_WEBSOCKET_PROTOCOL: +SEC_WEBSOCKET_VERSION: +SERVER: +SET_COOKIE: +TE: +TRAILER: +TRANSFER_ENCODING: +UPGRADE: +URI: +USER_AGENT: +VARY: +VIA: +WANT_DIGEST: +WARNING: +WWW_AUTHENTICATE: +X_FORWARDED_FOR: +X_FORWARDED_HOST: +X_FORWARDED_PROTO: +missing: + /* nothing found */ + return -1; +} diff --git a/contrib/python/aiohttp/aiohttp/_find_header.h b/contrib/python/aiohttp/aiohttp/_find_header.h new file mode 100644 index 0000000000..99b7b4f828 --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/_find_header.h @@ -0,0 +1,14 @@ +#ifndef _FIND_HEADERS_H +#define _FIND_HEADERS_H + +#ifdef __cplusplus +extern "C" { +#endif + +int find_header(const char *str, int size); + + +#ifdef __cplusplus +} +#endif +#endif diff --git a/contrib/python/aiohttp/aiohttp/_find_header.pxd b/contrib/python/aiohttp/aiohttp/_find_header.pxd new file mode 100644 index 0000000000..37a6c37268 --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/_find_header.pxd @@ -0,0 +1,2 @@ +cdef extern from "_find_header.h": + int find_header(char *, int) diff --git a/contrib/python/aiohttp/aiohttp/_headers.pxi b/contrib/python/aiohttp/aiohttp/_headers.pxi new file mode 100644 index 0000000000..3744721d47 --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/_headers.pxi @@ -0,0 +1,83 @@ +# The file is autogenerated from aiohttp/hdrs.py +# Run ./tools/gen.py to update it after the origin changing. + +from . import hdrs +cdef tuple headers = ( + hdrs.ACCEPT, + hdrs.ACCEPT_CHARSET, + hdrs.ACCEPT_ENCODING, + hdrs.ACCEPT_LANGUAGE, + hdrs.ACCEPT_RANGES, + hdrs.ACCESS_CONTROL_ALLOW_CREDENTIALS, + hdrs.ACCESS_CONTROL_ALLOW_HEADERS, + hdrs.ACCESS_CONTROL_ALLOW_METHODS, + hdrs.ACCESS_CONTROL_ALLOW_ORIGIN, + hdrs.ACCESS_CONTROL_EXPOSE_HEADERS, + hdrs.ACCESS_CONTROL_MAX_AGE, + hdrs.ACCESS_CONTROL_REQUEST_HEADERS, + hdrs.ACCESS_CONTROL_REQUEST_METHOD, + hdrs.AGE, + hdrs.ALLOW, + hdrs.AUTHORIZATION, + hdrs.CACHE_CONTROL, + hdrs.CONNECTION, + hdrs.CONTENT_DISPOSITION, + hdrs.CONTENT_ENCODING, + hdrs.CONTENT_LANGUAGE, + hdrs.CONTENT_LENGTH, + hdrs.CONTENT_LOCATION, + hdrs.CONTENT_MD5, + hdrs.CONTENT_RANGE, + hdrs.CONTENT_TRANSFER_ENCODING, + hdrs.CONTENT_TYPE, + hdrs.COOKIE, + hdrs.DATE, + hdrs.DESTINATION, + hdrs.DIGEST, + hdrs.ETAG, + hdrs.EXPECT, + hdrs.EXPIRES, + hdrs.FORWARDED, + hdrs.FROM, + hdrs.HOST, + hdrs.IF_MATCH, + hdrs.IF_MODIFIED_SINCE, + hdrs.IF_NONE_MATCH, + hdrs.IF_RANGE, + hdrs.IF_UNMODIFIED_SINCE, + hdrs.KEEP_ALIVE, + hdrs.LAST_EVENT_ID, + hdrs.LAST_MODIFIED, + hdrs.LINK, + hdrs.LOCATION, + hdrs.MAX_FORWARDS, + hdrs.ORIGIN, + hdrs.PRAGMA, + hdrs.PROXY_AUTHENTICATE, + hdrs.PROXY_AUTHORIZATION, + hdrs.RANGE, + hdrs.REFERER, + hdrs.RETRY_AFTER, + hdrs.SEC_WEBSOCKET_ACCEPT, + hdrs.SEC_WEBSOCKET_EXTENSIONS, + hdrs.SEC_WEBSOCKET_KEY, + hdrs.SEC_WEBSOCKET_KEY1, + hdrs.SEC_WEBSOCKET_PROTOCOL, + hdrs.SEC_WEBSOCKET_VERSION, + hdrs.SERVER, + hdrs.SET_COOKIE, + hdrs.TE, + hdrs.TRAILER, + hdrs.TRANSFER_ENCODING, + hdrs.URI, + hdrs.UPGRADE, + hdrs.USER_AGENT, + hdrs.VARY, + hdrs.VIA, + hdrs.WWW_AUTHENTICATE, + hdrs.WANT_DIGEST, + hdrs.WARNING, + hdrs.X_FORWARDED_FOR, + hdrs.X_FORWARDED_HOST, + hdrs.X_FORWARDED_PROTO, +) diff --git a/contrib/python/aiohttp/aiohttp/_helpers.pyx b/contrib/python/aiohttp/aiohttp/_helpers.pyx new file mode 100644 index 0000000000..665f367c5d --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/_helpers.pyx @@ -0,0 +1,35 @@ +cdef class reify: + """Use as a class method decorator. It operates almost exactly like + the Python `@property` decorator, but it puts the result of the + method it decorates into the instance dict after the first call, + effectively replacing the function it decorates with an instance + variable. It is, in Python parlance, a data descriptor. + + """ + + cdef object wrapped + cdef object name + + def __init__(self, wrapped): + self.wrapped = wrapped + self.name = wrapped.__name__ + + @property + def __doc__(self): + return self.wrapped.__doc__ + + def __get__(self, inst, owner): + try: + try: + return inst._cache[self.name] + except KeyError: + val = self.wrapped(inst) + inst._cache[self.name] = val + return val + except AttributeError: + if inst is None: + return self + raise + + def __set__(self, inst, value): + raise AttributeError("reified property is read-only") diff --git a/contrib/python/aiohttp/aiohttp/_http_parser.pyx b/contrib/python/aiohttp/aiohttp/_http_parser.pyx new file mode 100644 index 0000000000..77bf0aa598 --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/_http_parser.pyx @@ -0,0 +1,818 @@ +#cython: language_level=3 +# +# Based on https://github.com/MagicStack/httptools +# +from __future__ import absolute_import, print_function + +from cpython cimport ( + Py_buffer, + PyBUF_SIMPLE, + PyBuffer_Release, + PyBytes_AsString, + PyBytes_AsStringAndSize, + PyObject_GetBuffer, +) +from cpython.mem cimport PyMem_Free, PyMem_Malloc +from libc.limits cimport ULLONG_MAX +from libc.string cimport memcpy + +from multidict import CIMultiDict as _CIMultiDict, CIMultiDictProxy as _CIMultiDictProxy +from yarl import URL as _URL + +from aiohttp import hdrs + +from .http_exceptions import ( + BadHttpMessage, + BadStatusLine, + ContentLengthError, + InvalidHeader, + InvalidURLError, + LineTooLong, + PayloadEncodingError, + TransferEncodingError, +) +from .http_parser import DeflateBuffer as _DeflateBuffer +from .http_writer import ( + HttpVersion as _HttpVersion, + HttpVersion10 as _HttpVersion10, + HttpVersion11 as _HttpVersion11, +) +from .streams import EMPTY_PAYLOAD as _EMPTY_PAYLOAD, StreamReader as _StreamReader + +cimport cython + +from aiohttp cimport _cparser as cparser + +include "_headers.pxi" + +from aiohttp cimport _find_header + +DEF DEFAULT_FREELIST_SIZE = 250 + +cdef extern from "Python.h": + int PyByteArray_Resize(object, Py_ssize_t) except -1 + Py_ssize_t PyByteArray_Size(object) except -1 + char* PyByteArray_AsString(object) + +__all__ = ('HttpRequestParser', 'HttpResponseParser', + 'RawRequestMessage', 'RawResponseMessage') + +cdef object URL = _URL +cdef object URL_build = URL.build +cdef object CIMultiDict = _CIMultiDict +cdef object CIMultiDictProxy = _CIMultiDictProxy +cdef object HttpVersion = _HttpVersion +cdef object HttpVersion10 = _HttpVersion10 +cdef object HttpVersion11 = _HttpVersion11 +cdef object SEC_WEBSOCKET_KEY1 = hdrs.SEC_WEBSOCKET_KEY1 +cdef object CONTENT_ENCODING = hdrs.CONTENT_ENCODING +cdef object EMPTY_PAYLOAD = _EMPTY_PAYLOAD +cdef object StreamReader = _StreamReader +cdef object DeflateBuffer = _DeflateBuffer + + +cdef inline object extend(object buf, const char* at, size_t length): + cdef Py_ssize_t s + cdef char* ptr + s = PyByteArray_Size(buf) + PyByteArray_Resize(buf, s + length) + ptr = PyByteArray_AsString(buf) + memcpy(ptr + s, at, length) + + +DEF METHODS_COUNT = 46; + +cdef list _http_method = [] + +for i in range(METHODS_COUNT): + _http_method.append( + cparser.llhttp_method_name(<cparser.llhttp_method_t> i).decode('ascii')) + + +cdef inline str http_method_str(int i): + if i < METHODS_COUNT: + return <str>_http_method[i] + else: + return "<unknown>" + +cdef inline object find_header(bytes raw_header): + cdef Py_ssize_t size + cdef char *buf + cdef int idx + PyBytes_AsStringAndSize(raw_header, &buf, &size) + idx = _find_header.find_header(buf, size) + if idx == -1: + return raw_header.decode('utf-8', 'surrogateescape') + return headers[idx] + + +@cython.freelist(DEFAULT_FREELIST_SIZE) +cdef class RawRequestMessage: + cdef readonly str method + cdef readonly str path + cdef readonly object version # HttpVersion + cdef readonly object headers # CIMultiDict + cdef readonly object raw_headers # tuple + cdef readonly object should_close + cdef readonly object compression + cdef readonly object upgrade + cdef readonly object chunked + cdef readonly object url # yarl.URL + + def __init__(self, method, path, version, headers, raw_headers, + should_close, compression, upgrade, chunked, url): + self.method = method + self.path = path + self.version = version + self.headers = headers + self.raw_headers = raw_headers + self.should_close = should_close + self.compression = compression + self.upgrade = upgrade + self.chunked = chunked + self.url = url + + def __repr__(self): + info = [] + info.append(("method", self.method)) + info.append(("path", self.path)) + info.append(("version", self.version)) + info.append(("headers", self.headers)) + info.append(("raw_headers", self.raw_headers)) + info.append(("should_close", self.should_close)) + info.append(("compression", self.compression)) + info.append(("upgrade", self.upgrade)) + info.append(("chunked", self.chunked)) + info.append(("url", self.url)) + sinfo = ', '.join(name + '=' + repr(val) for name, val in info) + return '<RawRequestMessage(' + sinfo + ')>' + + def _replace(self, **dct): + cdef RawRequestMessage ret + ret = _new_request_message(self.method, + self.path, + self.version, + self.headers, + self.raw_headers, + self.should_close, + self.compression, + self.upgrade, + self.chunked, + self.url) + if "method" in dct: + ret.method = dct["method"] + if "path" in dct: + ret.path = dct["path"] + if "version" in dct: + ret.version = dct["version"] + if "headers" in dct: + ret.headers = dct["headers"] + if "raw_headers" in dct: + ret.raw_headers = dct["raw_headers"] + if "should_close" in dct: + ret.should_close = dct["should_close"] + if "compression" in dct: + ret.compression = dct["compression"] + if "upgrade" in dct: + ret.upgrade = dct["upgrade"] + if "chunked" in dct: + ret.chunked = dct["chunked"] + if "url" in dct: + ret.url = dct["url"] + return ret + +cdef _new_request_message(str method, + str path, + object version, + object headers, + object raw_headers, + bint should_close, + object compression, + bint upgrade, + bint chunked, + object url): + cdef RawRequestMessage ret + ret = RawRequestMessage.__new__(RawRequestMessage) + ret.method = method + ret.path = path + ret.version = version + ret.headers = headers + ret.raw_headers = raw_headers + ret.should_close = should_close + ret.compression = compression + ret.upgrade = upgrade + ret.chunked = chunked + ret.url = url + return ret + + +@cython.freelist(DEFAULT_FREELIST_SIZE) +cdef class RawResponseMessage: + cdef readonly object version # HttpVersion + cdef readonly int code + cdef readonly str reason + cdef readonly object headers # CIMultiDict + cdef readonly object raw_headers # tuple + cdef readonly object should_close + cdef readonly object compression + cdef readonly object upgrade + cdef readonly object chunked + + def __init__(self, version, code, reason, headers, raw_headers, + should_close, compression, upgrade, chunked): + self.version = version + self.code = code + self.reason = reason + self.headers = headers + self.raw_headers = raw_headers + self.should_close = should_close + self.compression = compression + self.upgrade = upgrade + self.chunked = chunked + + def __repr__(self): + info = [] + info.append(("version", self.version)) + info.append(("code", self.code)) + info.append(("reason", self.reason)) + info.append(("headers", self.headers)) + info.append(("raw_headers", self.raw_headers)) + info.append(("should_close", self.should_close)) + info.append(("compression", self.compression)) + info.append(("upgrade", self.upgrade)) + info.append(("chunked", self.chunked)) + sinfo = ', '.join(name + '=' + repr(val) for name, val in info) + return '<RawResponseMessage(' + sinfo + ')>' + + +cdef _new_response_message(object version, + int code, + str reason, + object headers, + object raw_headers, + bint should_close, + object compression, + bint upgrade, + bint chunked): + cdef RawResponseMessage ret + ret = RawResponseMessage.__new__(RawResponseMessage) + ret.version = version + ret.code = code + ret.reason = reason + ret.headers = headers + ret.raw_headers = raw_headers + ret.should_close = should_close + ret.compression = compression + ret.upgrade = upgrade + ret.chunked = chunked + return ret + + +@cython.internal +cdef class HttpParser: + + cdef: + cparser.llhttp_t* _cparser + cparser.llhttp_settings_t* _csettings + + bytearray _raw_name + bytearray _raw_value + bint _has_value + + object _protocol + object _loop + object _timer + + size_t _max_line_size + size_t _max_field_size + size_t _max_headers + bint _response_with_body + bint _read_until_eof + + bint _started + object _url + bytearray _buf + str _path + str _reason + object _headers + list _raw_headers + bint _upgraded + list _messages + object _payload + bint _payload_error + object _payload_exception + object _last_error + bint _auto_decompress + int _limit + + str _content_encoding + + Py_buffer py_buf + + def __cinit__(self): + self._cparser = <cparser.llhttp_t*> \ + PyMem_Malloc(sizeof(cparser.llhttp_t)) + if self._cparser is NULL: + raise MemoryError() + + self._csettings = <cparser.llhttp_settings_t*> \ + PyMem_Malloc(sizeof(cparser.llhttp_settings_t)) + if self._csettings is NULL: + raise MemoryError() + + def __dealloc__(self): + PyMem_Free(self._cparser) + PyMem_Free(self._csettings) + + cdef _init( + self, cparser.llhttp_type mode, + object protocol, object loop, int limit, + object timer=None, + size_t max_line_size=8190, size_t max_headers=32768, + size_t max_field_size=8190, payload_exception=None, + bint response_with_body=True, bint read_until_eof=False, + bint auto_decompress=True, + ): + cparser.llhttp_settings_init(self._csettings) + cparser.llhttp_init(self._cparser, mode, self._csettings) + self._cparser.data = <void*>self + self._cparser.content_length = 0 + + self._protocol = protocol + self._loop = loop + self._timer = timer + + self._buf = bytearray() + self._payload = None + self._payload_error = 0 + self._payload_exception = payload_exception + self._messages = [] + + self._raw_name = bytearray() + self._raw_value = bytearray() + self._has_value = False + + self._max_line_size = max_line_size + self._max_headers = max_headers + self._max_field_size = max_field_size + self._response_with_body = response_with_body + self._read_until_eof = read_until_eof + self._upgraded = False + self._auto_decompress = auto_decompress + self._content_encoding = None + + self._csettings.on_url = cb_on_url + self._csettings.on_status = cb_on_status + self._csettings.on_header_field = cb_on_header_field + self._csettings.on_header_value = cb_on_header_value + self._csettings.on_headers_complete = cb_on_headers_complete + self._csettings.on_body = cb_on_body + self._csettings.on_message_begin = cb_on_message_begin + self._csettings.on_message_complete = cb_on_message_complete + self._csettings.on_chunk_header = cb_on_chunk_header + self._csettings.on_chunk_complete = cb_on_chunk_complete + + self._last_error = None + self._limit = limit + + cdef _process_header(self): + if self._raw_name: + raw_name = bytes(self._raw_name) + raw_value = bytes(self._raw_value) + + name = find_header(raw_name) + value = raw_value.decode('utf-8', 'surrogateescape') + + self._headers.add(name, value) + + if name is CONTENT_ENCODING: + self._content_encoding = value + + PyByteArray_Resize(self._raw_name, 0) + PyByteArray_Resize(self._raw_value, 0) + self._has_value = False + self._raw_headers.append((raw_name, raw_value)) + + cdef _on_header_field(self, const char* at, size_t length): + cdef Py_ssize_t size + cdef char *buf + if self._has_value: + self._process_header() + + size = PyByteArray_Size(self._raw_name) + PyByteArray_Resize(self._raw_name, size + length) + buf = PyByteArray_AsString(self._raw_name) + memcpy(buf + size, at, length) + + cdef _on_header_value(self, const char* at, size_t length): + cdef Py_ssize_t size + cdef char *buf + + size = PyByteArray_Size(self._raw_value) + PyByteArray_Resize(self._raw_value, size + length) + buf = PyByteArray_AsString(self._raw_value) + memcpy(buf + size, at, length) + self._has_value = True + + cdef _on_headers_complete(self): + self._process_header() + + method = http_method_str(self._cparser.method) + should_close = not cparser.llhttp_should_keep_alive(self._cparser) + upgrade = self._cparser.upgrade + chunked = self._cparser.flags & cparser.F_CHUNKED + + raw_headers = tuple(self._raw_headers) + headers = CIMultiDictProxy(self._headers) + + if upgrade or self._cparser.method == 5: # cparser.CONNECT: + self._upgraded = True + + # do not support old websocket spec + if SEC_WEBSOCKET_KEY1 in headers: + raise InvalidHeader(SEC_WEBSOCKET_KEY1) + + encoding = None + enc = self._content_encoding + if enc is not None: + self._content_encoding = None + enc = enc.lower() + if enc in ('gzip', 'deflate', 'br'): + encoding = enc + + if self._cparser.type == cparser.HTTP_REQUEST: + msg = _new_request_message( + method, self._path, + self.http_version(), headers, raw_headers, + should_close, encoding, upgrade, chunked, self._url) + else: + msg = _new_response_message( + self.http_version(), self._cparser.status_code, self._reason, + headers, raw_headers, should_close, encoding, + upgrade, chunked) + + if ( + ULLONG_MAX > self._cparser.content_length > 0 or chunked or + self._cparser.method == 5 or # CONNECT: 5 + (self._cparser.status_code >= 199 and + self._cparser.content_length == 0 and + self._read_until_eof) + ): + payload = StreamReader( + self._protocol, timer=self._timer, loop=self._loop, + limit=self._limit) + else: + payload = EMPTY_PAYLOAD + + self._payload = payload + if encoding is not None and self._auto_decompress: + self._payload = DeflateBuffer(payload, encoding) + + if not self._response_with_body: + payload = EMPTY_PAYLOAD + + self._messages.append((msg, payload)) + + cdef _on_message_complete(self): + self._payload.feed_eof() + self._payload = None + + cdef _on_chunk_header(self): + self._payload.begin_http_chunk_receiving() + + cdef _on_chunk_complete(self): + self._payload.end_http_chunk_receiving() + + cdef object _on_status_complete(self): + pass + + cdef inline http_version(self): + cdef cparser.llhttp_t* parser = self._cparser + + if parser.http_major == 1: + if parser.http_minor == 0: + return HttpVersion10 + elif parser.http_minor == 1: + return HttpVersion11 + + return HttpVersion(parser.http_major, parser.http_minor) + + ### Public API ### + + def feed_eof(self): + cdef bytes desc + + if self._payload is not None: + if self._cparser.flags & cparser.F_CHUNKED: + raise TransferEncodingError( + "Not enough data for satisfy transfer length header.") + elif self._cparser.flags & cparser.F_CONTENT_LENGTH: + raise ContentLengthError( + "Not enough data for satisfy content length header.") + elif cparser.llhttp_get_errno(self._cparser) != cparser.HPE_OK: + desc = cparser.llhttp_get_error_reason(self._cparser) + raise PayloadEncodingError(desc.decode('latin-1')) + else: + self._payload.feed_eof() + elif self._started: + self._on_headers_complete() + if self._messages: + return self._messages[-1][0] + + def feed_data(self, data): + cdef: + size_t data_len + size_t nb + cdef cparser.llhttp_errno_t errno + + PyObject_GetBuffer(data, &self.py_buf, PyBUF_SIMPLE) + data_len = <size_t>self.py_buf.len + + errno = cparser.llhttp_execute( + self._cparser, + <char*>self.py_buf.buf, + data_len) + + if errno is cparser.HPE_PAUSED_UPGRADE: + cparser.llhttp_resume_after_upgrade(self._cparser) + + nb = cparser.llhttp_get_error_pos(self._cparser) - <char*>self.py_buf.buf + + PyBuffer_Release(&self.py_buf) + + if errno not in (cparser.HPE_OK, cparser.HPE_PAUSED_UPGRADE): + if self._payload_error == 0: + if self._last_error is not None: + ex = self._last_error + self._last_error = None + else: + ex = parser_error_from_errno(self._cparser) + self._payload = None + raise ex + + if self._messages: + messages = self._messages + self._messages = [] + else: + messages = () + + if self._upgraded: + return messages, True, data[nb:] + else: + return messages, False, b'' + + def set_upgraded(self, val): + self._upgraded = val + + +cdef class HttpRequestParser(HttpParser): + + def __init__( + self, protocol, loop, int limit, timer=None, + size_t max_line_size=8190, size_t max_headers=32768, + size_t max_field_size=8190, payload_exception=None, + bint response_with_body=True, bint read_until_eof=False, + bint auto_decompress=True, + ): + self._init(cparser.HTTP_REQUEST, protocol, loop, limit, timer, + max_line_size, max_headers, max_field_size, + payload_exception, response_with_body, read_until_eof, + auto_decompress) + + cdef object _on_status_complete(self): + cdef int idx1, idx2 + if not self._buf: + return + self._path = self._buf.decode('utf-8', 'surrogateescape') + try: + idx3 = len(self._path) + idx1 = self._path.find("?") + if idx1 == -1: + query = "" + idx2 = self._path.find("#") + if idx2 == -1: + path = self._path + fragment = "" + else: + path = self._path[0: idx2] + fragment = self._path[idx2+1:] + + else: + path = self._path[0:idx1] + idx1 += 1 + idx2 = self._path.find("#", idx1+1) + if idx2 == -1: + query = self._path[idx1:] + fragment = "" + else: + query = self._path[idx1: idx2] + fragment = self._path[idx2+1:] + + self._url = URL.build( + path=path, + query_string=query, + fragment=fragment, + encoded=True, + ) + finally: + PyByteArray_Resize(self._buf, 0) + + +cdef class HttpResponseParser(HttpParser): + + def __init__( + self, protocol, loop, int limit, timer=None, + size_t max_line_size=8190, size_t max_headers=32768, + size_t max_field_size=8190, payload_exception=None, + bint response_with_body=True, bint read_until_eof=False, + bint auto_decompress=True + ): + self._init(cparser.HTTP_RESPONSE, protocol, loop, limit, timer, + max_line_size, max_headers, max_field_size, + payload_exception, response_with_body, read_until_eof, + auto_decompress) + + cdef object _on_status_complete(self): + if self._buf: + self._reason = self._buf.decode('utf-8', 'surrogateescape') + PyByteArray_Resize(self._buf, 0) + else: + self._reason = self._reason or '' + +cdef int cb_on_message_begin(cparser.llhttp_t* parser) except -1: + cdef HttpParser pyparser = <HttpParser>parser.data + + pyparser._started = True + pyparser._headers = CIMultiDict() + pyparser._raw_headers = [] + PyByteArray_Resize(pyparser._buf, 0) + pyparser._path = None + pyparser._reason = None + return 0 + + +cdef int cb_on_url(cparser.llhttp_t* parser, + const char *at, size_t length) except -1: + cdef HttpParser pyparser = <HttpParser>parser.data + try: + if length > pyparser._max_line_size: + raise LineTooLong( + 'Status line is too long', pyparser._max_line_size, length) + extend(pyparser._buf, at, length) + except BaseException as ex: + pyparser._last_error = ex + return -1 + else: + return 0 + + +cdef int cb_on_status(cparser.llhttp_t* parser, + const char *at, size_t length) except -1: + cdef HttpParser pyparser = <HttpParser>parser.data + cdef str reason + try: + if length > pyparser._max_line_size: + raise LineTooLong( + 'Status line is too long', pyparser._max_line_size, length) + extend(pyparser._buf, at, length) + except BaseException as ex: + pyparser._last_error = ex + return -1 + else: + return 0 + + +cdef int cb_on_header_field(cparser.llhttp_t* parser, + const char *at, size_t length) except -1: + cdef HttpParser pyparser = <HttpParser>parser.data + cdef Py_ssize_t size + try: + pyparser._on_status_complete() + size = len(pyparser._raw_name) + length + if size > pyparser._max_field_size: + raise LineTooLong( + 'Header name is too long', pyparser._max_field_size, size) + pyparser._on_header_field(at, length) + except BaseException as ex: + pyparser._last_error = ex + return -1 + else: + return 0 + + +cdef int cb_on_header_value(cparser.llhttp_t* parser, + const char *at, size_t length) except -1: + cdef HttpParser pyparser = <HttpParser>parser.data + cdef Py_ssize_t size + try: + size = len(pyparser._raw_value) + length + if size > pyparser._max_field_size: + raise LineTooLong( + 'Header value is too long', pyparser._max_field_size, size) + pyparser._on_header_value(at, length) + except BaseException as ex: + pyparser._last_error = ex + return -1 + else: + return 0 + + +cdef int cb_on_headers_complete(cparser.llhttp_t* parser) except -1: + cdef HttpParser pyparser = <HttpParser>parser.data + try: + pyparser._on_status_complete() + pyparser._on_headers_complete() + except BaseException as exc: + pyparser._last_error = exc + return -1 + else: + if pyparser._cparser.upgrade or pyparser._cparser.method == 5: # CONNECT + return 2 + else: + return 0 + + +cdef int cb_on_body(cparser.llhttp_t* parser, + const char *at, size_t length) except -1: + cdef HttpParser pyparser = <HttpParser>parser.data + cdef bytes body = at[:length] + try: + pyparser._payload.feed_data(body, length) + except BaseException as exc: + if pyparser._payload_exception is not None: + pyparser._payload.set_exception(pyparser._payload_exception(str(exc))) + else: + pyparser._payload.set_exception(exc) + pyparser._payload_error = 1 + return -1 + else: + return 0 + + +cdef int cb_on_message_complete(cparser.llhttp_t* parser) except -1: + cdef HttpParser pyparser = <HttpParser>parser.data + try: + pyparser._started = False + pyparser._on_message_complete() + except BaseException as exc: + pyparser._last_error = exc + return -1 + else: + return 0 + + +cdef int cb_on_chunk_header(cparser.llhttp_t* parser) except -1: + cdef HttpParser pyparser = <HttpParser>parser.data + try: + pyparser._on_chunk_header() + except BaseException as exc: + pyparser._last_error = exc + return -1 + else: + return 0 + + +cdef int cb_on_chunk_complete(cparser.llhttp_t* parser) except -1: + cdef HttpParser pyparser = <HttpParser>parser.data + try: + pyparser._on_chunk_complete() + except BaseException as exc: + pyparser._last_error = exc + return -1 + else: + return 0 + + +cdef parser_error_from_errno(cparser.llhttp_t* parser): + cdef cparser.llhttp_errno_t errno = cparser.llhttp_get_errno(parser) + cdef bytes desc = cparser.llhttp_get_error_reason(parser) + + if errno in (cparser.HPE_CB_MESSAGE_BEGIN, + cparser.HPE_CB_HEADERS_COMPLETE, + cparser.HPE_CB_MESSAGE_COMPLETE, + cparser.HPE_CB_CHUNK_HEADER, + cparser.HPE_CB_CHUNK_COMPLETE, + cparser.HPE_INVALID_CONSTANT, + cparser.HPE_INVALID_HEADER_TOKEN, + cparser.HPE_INVALID_CONTENT_LENGTH, + cparser.HPE_INVALID_CHUNK_SIZE, + cparser.HPE_INVALID_EOF_STATE, + cparser.HPE_INVALID_TRANSFER_ENCODING): + cls = BadHttpMessage + + elif errno == cparser.HPE_INVALID_STATUS: + cls = BadStatusLine + + elif errno == cparser.HPE_INVALID_METHOD: + cls = BadStatusLine + + elif errno == cparser.HPE_INVALID_VERSION: + cls = BadStatusLine + + elif errno == cparser.HPE_INVALID_URL: + cls = InvalidURLError + + else: + cls = BadHttpMessage + + return cls(desc.decode('latin-1')) diff --git a/contrib/python/aiohttp/aiohttp/_http_writer.pyx b/contrib/python/aiohttp/aiohttp/_http_writer.pyx new file mode 100644 index 0000000000..eff8521958 --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/_http_writer.pyx @@ -0,0 +1,163 @@ +from cpython.bytes cimport PyBytes_FromStringAndSize +from cpython.exc cimport PyErr_NoMemory +from cpython.mem cimport PyMem_Free, PyMem_Malloc, PyMem_Realloc +from cpython.object cimport PyObject_Str +from libc.stdint cimport uint8_t, uint64_t +from libc.string cimport memcpy + +from multidict import istr + +DEF BUF_SIZE = 16 * 1024 # 16KiB +cdef char BUFFER[BUF_SIZE] + +cdef object _istr = istr + + +# ----------------- writer --------------------------- + +cdef struct Writer: + char *buf + Py_ssize_t size + Py_ssize_t pos + + +cdef inline void _init_writer(Writer* writer): + writer.buf = &BUFFER[0] + writer.size = BUF_SIZE + writer.pos = 0 + + +cdef inline void _release_writer(Writer* writer): + if writer.buf != BUFFER: + PyMem_Free(writer.buf) + + +cdef inline int _write_byte(Writer* writer, uint8_t ch): + cdef char * buf + cdef Py_ssize_t size + + if writer.pos == writer.size: + # reallocate + size = writer.size + BUF_SIZE + if writer.buf == BUFFER: + buf = <char*>PyMem_Malloc(size) + if buf == NULL: + PyErr_NoMemory() + return -1 + memcpy(buf, writer.buf, writer.size) + else: + buf = <char*>PyMem_Realloc(writer.buf, size) + if buf == NULL: + PyErr_NoMemory() + return -1 + writer.buf = buf + writer.size = size + writer.buf[writer.pos] = <char>ch + writer.pos += 1 + return 0 + + +cdef inline int _write_utf8(Writer* writer, Py_UCS4 symbol): + cdef uint64_t utf = <uint64_t> symbol + + if utf < 0x80: + return _write_byte(writer, <uint8_t>utf) + elif utf < 0x800: + if _write_byte(writer, <uint8_t>(0xc0 | (utf >> 6))) < 0: + return -1 + return _write_byte(writer, <uint8_t>(0x80 | (utf & 0x3f))) + elif 0xD800 <= utf <= 0xDFFF: + # surogate pair, ignored + return 0 + elif utf < 0x10000: + if _write_byte(writer, <uint8_t>(0xe0 | (utf >> 12))) < 0: + return -1 + if _write_byte(writer, <uint8_t>(0x80 | ((utf >> 6) & 0x3f))) < 0: + return -1 + return _write_byte(writer, <uint8_t>(0x80 | (utf & 0x3f))) + elif utf > 0x10FFFF: + # symbol is too large + return 0 + else: + if _write_byte(writer, <uint8_t>(0xf0 | (utf >> 18))) < 0: + return -1 + if _write_byte(writer, + <uint8_t>(0x80 | ((utf >> 12) & 0x3f))) < 0: + return -1 + if _write_byte(writer, + <uint8_t>(0x80 | ((utf >> 6) & 0x3f))) < 0: + return -1 + return _write_byte(writer, <uint8_t>(0x80 | (utf & 0x3f))) + + +cdef inline int _write_str(Writer* writer, str s): + cdef Py_UCS4 ch + for ch in s: + if _write_utf8(writer, ch) < 0: + return -1 + + +# --------------- _serialize_headers ---------------------- + +cdef str to_str(object s): + typ = type(s) + if typ is str: + return <str>s + elif typ is _istr: + return PyObject_Str(s) + elif not isinstance(s, str): + raise TypeError("Cannot serialize non-str key {!r}".format(s)) + else: + return str(s) + + +cdef void _safe_header(str string) except *: + if "\r" in string or "\n" in string: + raise ValueError( + "Newline or carriage return character detected in HTTP status message or " + "header. This is a potential security issue." + ) + + +def _serialize_headers(str status_line, headers): + cdef Writer writer + cdef object key + cdef object val + cdef bytes ret + + _init_writer(&writer) + + for key, val in headers.items(): + _safe_header(to_str(key)) + _safe_header(to_str(val)) + + try: + if _write_str(&writer, status_line) < 0: + raise + if _write_byte(&writer, b'\r') < 0: + raise + if _write_byte(&writer, b'\n') < 0: + raise + + for key, val in headers.items(): + if _write_str(&writer, to_str(key)) < 0: + raise + if _write_byte(&writer, b':') < 0: + raise + if _write_byte(&writer, b' ') < 0: + raise + if _write_str(&writer, to_str(val)) < 0: + raise + if _write_byte(&writer, b'\r') < 0: + raise + if _write_byte(&writer, b'\n') < 0: + raise + + if _write_byte(&writer, b'\r') < 0: + raise + if _write_byte(&writer, b'\n') < 0: + raise + + return PyBytes_FromStringAndSize(writer.buf, writer.pos) + finally: + _release_writer(&writer) diff --git a/contrib/python/aiohttp/aiohttp/_websocket.pyx b/contrib/python/aiohttp/aiohttp/_websocket.pyx new file mode 100644 index 0000000000..94318d2b1b --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/_websocket.pyx @@ -0,0 +1,56 @@ +from cpython cimport PyBytes_AsString + + +#from cpython cimport PyByteArray_AsString # cython still not exports that +cdef extern from "Python.h": + char* PyByteArray_AsString(bytearray ba) except NULL + +from libc.stdint cimport uint32_t, uint64_t, uintmax_t + + +def _websocket_mask_cython(object mask, object data): + """Note, this function mutates its `data` argument + """ + cdef: + Py_ssize_t data_len, i + # bit operations on signed integers are implementation-specific + unsigned char * in_buf + const unsigned char * mask_buf + uint32_t uint32_msk + uint64_t uint64_msk + + assert len(mask) == 4 + + if not isinstance(mask, bytes): + mask = bytes(mask) + + if isinstance(data, bytearray): + data = <bytearray>data + else: + data = bytearray(data) + + data_len = len(data) + in_buf = <unsigned char*>PyByteArray_AsString(data) + mask_buf = <const unsigned char*>PyBytes_AsString(mask) + uint32_msk = (<uint32_t*>mask_buf)[0] + + # TODO: align in_data ptr to achieve even faster speeds + # does it need in python ?! malloc() always aligns to sizeof(long) bytes + + if sizeof(size_t) >= 8: + uint64_msk = uint32_msk + uint64_msk = (uint64_msk << 32) | uint32_msk + + while data_len >= 8: + (<uint64_t*>in_buf)[0] ^= uint64_msk + in_buf += 8 + data_len -= 8 + + + while data_len >= 4: + (<uint32_t*>in_buf)[0] ^= uint32_msk + in_buf += 4 + data_len -= 4 + + for i in range(0, data_len): + in_buf[i] ^= mask_buf[i] diff --git a/contrib/python/aiohttp/aiohttp/abc.py b/contrib/python/aiohttp/aiohttp/abc.py new file mode 100644 index 0000000000..06fc831638 --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/abc.py @@ -0,0 +1,207 @@ +import asyncio +import logging +from abc import ABC, abstractmethod +from collections.abc import Sized +from http.cookies import BaseCookie, Morsel +from typing import ( + TYPE_CHECKING, + Any, + Awaitable, + Callable, + Dict, + Generator, + Iterable, + List, + Optional, + Tuple, +) + +from multidict import CIMultiDict +from yarl import URL + +from .helpers import get_running_loop +from .typedefs import LooseCookies + +if TYPE_CHECKING: # pragma: no cover + from .web_app import Application + from .web_exceptions import HTTPException + from .web_request import BaseRequest, Request + from .web_response import StreamResponse +else: + BaseRequest = Request = Application = StreamResponse = None + HTTPException = None + + +class AbstractRouter(ABC): + def __init__(self) -> None: + self._frozen = False + + def post_init(self, app: Application) -> None: + """Post init stage. + + Not an abstract method for sake of backward compatibility, + but if the router wants to be aware of the application + it can override this. + """ + + @property + def frozen(self) -> bool: + return self._frozen + + def freeze(self) -> None: + """Freeze router.""" + self._frozen = True + + @abstractmethod + async def resolve(self, request: Request) -> "AbstractMatchInfo": + """Return MATCH_INFO for given request""" + + +class AbstractMatchInfo(ABC): + @property # pragma: no branch + @abstractmethod + def handler(self) -> Callable[[Request], Awaitable[StreamResponse]]: + """Execute matched request handler""" + + @property + @abstractmethod + def expect_handler(self) -> Callable[[Request], Awaitable[None]]: + """Expect handler for 100-continue processing""" + + @property # pragma: no branch + @abstractmethod + def http_exception(self) -> Optional[HTTPException]: + """HTTPException instance raised on router's resolving, or None""" + + @abstractmethod # pragma: no branch + def get_info(self) -> Dict[str, Any]: + """Return a dict with additional info useful for introspection""" + + @property # pragma: no branch + @abstractmethod + def apps(self) -> Tuple[Application, ...]: + """Stack of nested applications. + + Top level application is left-most element. + + """ + + @abstractmethod + def add_app(self, app: Application) -> None: + """Add application to the nested apps stack.""" + + @abstractmethod + def freeze(self) -> None: + """Freeze the match info. + + The method is called after route resolution. + + After the call .add_app() is forbidden. + + """ + + +class AbstractView(ABC): + """Abstract class based view.""" + + def __init__(self, request: Request) -> None: + self._request = request + + @property + def request(self) -> Request: + """Request instance.""" + return self._request + + @abstractmethod + def __await__(self) -> Generator[Any, None, StreamResponse]: + """Execute the view handler.""" + + +class AbstractResolver(ABC): + """Abstract DNS resolver.""" + + @abstractmethod + async def resolve(self, host: str, port: int, family: int) -> List[Dict[str, Any]]: + """Return IP address for given hostname""" + + @abstractmethod + async def close(self) -> None: + """Release resolver""" + + +if TYPE_CHECKING: # pragma: no cover + IterableBase = Iterable[Morsel[str]] +else: + IterableBase = Iterable + + +ClearCookiePredicate = Callable[["Morsel[str]"], bool] + + +class AbstractCookieJar(Sized, IterableBase): + """Abstract Cookie Jar.""" + + def __init__(self, *, loop: Optional[asyncio.AbstractEventLoop] = None) -> None: + self._loop = get_running_loop(loop) + + @abstractmethod + def clear(self, predicate: Optional[ClearCookiePredicate] = None) -> None: + """Clear all cookies if no predicate is passed.""" + + @abstractmethod + def clear_domain(self, domain: str) -> None: + """Clear all cookies for domain and all subdomains.""" + + @abstractmethod + def update_cookies(self, cookies: LooseCookies, response_url: URL = URL()) -> None: + """Update cookies.""" + + @abstractmethod + def filter_cookies(self, request_url: URL) -> "BaseCookie[str]": + """Return the jar's cookies filtered by their attributes.""" + + +class AbstractStreamWriter(ABC): + """Abstract stream writer.""" + + buffer_size = 0 + output_size = 0 + length = 0 # type: Optional[int] + + @abstractmethod + async def write(self, chunk: bytes) -> None: + """Write chunk into stream.""" + + @abstractmethod + async def write_eof(self, chunk: bytes = b"") -> None: + """Write last chunk.""" + + @abstractmethod + async def drain(self) -> None: + """Flush the write buffer.""" + + @abstractmethod + def enable_compression(self, encoding: str = "deflate") -> None: + """Enable HTTP body compression""" + + @abstractmethod + def enable_chunking(self) -> None: + """Enable HTTP chunked mode""" + + @abstractmethod + async def write_headers( + self, status_line: str, headers: "CIMultiDict[str]" + ) -> None: + """Write HTTP headers""" + + +class AbstractAccessLogger(ABC): + """Abstract writer to access log.""" + + def __init__(self, logger: logging.Logger, log_format: str) -> None: + self.logger = logger + self.log_format = log_format + + @abstractmethod + def log(self, request: BaseRequest, response: StreamResponse, time: float) -> None: + """Emit log to logger.""" diff --git a/contrib/python/aiohttp/aiohttp/base_protocol.py b/contrib/python/aiohttp/aiohttp/base_protocol.py new file mode 100644 index 0000000000..fff4610a1e --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/base_protocol.py @@ -0,0 +1,87 @@ +import asyncio +from typing import Optional, cast + +from .tcp_helpers import tcp_nodelay + + +class BaseProtocol(asyncio.Protocol): + __slots__ = ( + "_loop", + "_paused", + "_drain_waiter", + "_connection_lost", + "_reading_paused", + "transport", + ) + + def __init__(self, loop: asyncio.AbstractEventLoop) -> None: + self._loop = loop # type: asyncio.AbstractEventLoop + self._paused = False + self._drain_waiter = None # type: Optional[asyncio.Future[None]] + self._connection_lost = False + self._reading_paused = False + + self.transport = None # type: Optional[asyncio.Transport] + + def pause_writing(self) -> None: + assert not self._paused + self._paused = True + + def resume_writing(self) -> None: + assert self._paused + self._paused = False + + waiter = self._drain_waiter + if waiter is not None: + self._drain_waiter = None + if not waiter.done(): + waiter.set_result(None) + + def pause_reading(self) -> None: + if not self._reading_paused and self.transport is not None: + try: + self.transport.pause_reading() + except (AttributeError, NotImplementedError, RuntimeError): + pass + self._reading_paused = True + + def resume_reading(self) -> None: + if self._reading_paused and self.transport is not None: + try: + self.transport.resume_reading() + except (AttributeError, NotImplementedError, RuntimeError): + pass + self._reading_paused = False + + def connection_made(self, transport: asyncio.BaseTransport) -> None: + tr = cast(asyncio.Transport, transport) + tcp_nodelay(tr, True) + self.transport = tr + + def connection_lost(self, exc: Optional[BaseException]) -> None: + self._connection_lost = True + # Wake up the writer if currently paused. + self.transport = None + if not self._paused: + return + waiter = self._drain_waiter + if waiter is None: + return + self._drain_waiter = None + if waiter.done(): + return + if exc is None: + waiter.set_result(None) + else: + waiter.set_exception(exc) + + async def _drain_helper(self) -> None: + if self._connection_lost: + raise ConnectionResetError("Connection lost") + if not self._paused: + return + waiter = self._drain_waiter + if waiter is None: + waiter = self._loop.create_future() + self._drain_waiter = waiter + await asyncio.shield(waiter) diff --git a/contrib/python/aiohttp/aiohttp/client.py b/contrib/python/aiohttp/aiohttp/client.py new file mode 100644 index 0000000000..6ae9549db9 --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/client.py @@ -0,0 +1,1304 @@ +"""HTTP Client for asyncio.""" + +import asyncio +import base64 +import hashlib +import json +import os +import sys +import traceback +import warnings +from contextlib import suppress +from types import SimpleNamespace, TracebackType +from typing import ( + Any, + Awaitable, + Callable, + Coroutine, + FrozenSet, + Generator, + Generic, + Iterable, + List, + Mapping, + Optional, + Set, + Tuple, + Type, + TypeVar, + Union, +) + +import attr +from multidict import CIMultiDict, MultiDict, MultiDictProxy, istr +from yarl import URL + +from . import hdrs, http, payload +from .abc import AbstractCookieJar +from .client_exceptions import ( + ClientConnectionError as ClientConnectionError, + ClientConnectorCertificateError as ClientConnectorCertificateError, + ClientConnectorError as ClientConnectorError, + ClientConnectorSSLError as ClientConnectorSSLError, + ClientError as ClientError, + ClientHttpProxyError as ClientHttpProxyError, + ClientOSError as ClientOSError, + ClientPayloadError as ClientPayloadError, + ClientProxyConnectionError as ClientProxyConnectionError, + ClientResponseError as ClientResponseError, + ClientSSLError as ClientSSLError, + ContentTypeError as ContentTypeError, + InvalidURL as InvalidURL, + ServerConnectionError as ServerConnectionError, + ServerDisconnectedError as ServerDisconnectedError, + ServerFingerprintMismatch as ServerFingerprintMismatch, + ServerTimeoutError as ServerTimeoutError, + TooManyRedirects as TooManyRedirects, + WSServerHandshakeError as WSServerHandshakeError, +) +from .client_reqrep import ( + ClientRequest as ClientRequest, + ClientResponse as ClientResponse, + Fingerprint as Fingerprint, + RequestInfo as RequestInfo, + _merge_ssl_params, +) +from .client_ws import ClientWebSocketResponse as ClientWebSocketResponse +from .connector import ( + BaseConnector as BaseConnector, + NamedPipeConnector as NamedPipeConnector, + TCPConnector as TCPConnector, + UnixConnector as UnixConnector, +) +from .cookiejar import CookieJar +from .helpers import ( + DEBUG, + PY_36, + BasicAuth, + TimeoutHandle, + ceil_timeout, + get_env_proxy_for_url, + get_running_loop, + sentinel, + strip_auth_from_url, +) +from .http import WS_KEY, HttpVersion, WebSocketReader, WebSocketWriter +from .http_websocket import WSHandshakeError, WSMessage, ws_ext_gen, ws_ext_parse +from .streams import FlowControlDataQueue +from .tracing import Trace, TraceConfig +from .typedefs import Final, JSONEncoder, LooseCookies, LooseHeaders, StrOrURL + +__all__ = ( + # client_exceptions + "ClientConnectionError", + "ClientConnectorCertificateError", + "ClientConnectorError", + "ClientConnectorSSLError", + "ClientError", + "ClientHttpProxyError", + "ClientOSError", + "ClientPayloadError", + "ClientProxyConnectionError", + "ClientResponseError", + "ClientSSLError", + "ContentTypeError", + "InvalidURL", + "ServerConnectionError", + "ServerDisconnectedError", + "ServerFingerprintMismatch", + "ServerTimeoutError", + "TooManyRedirects", + "WSServerHandshakeError", + # client_reqrep + "ClientRequest", + "ClientResponse", + "Fingerprint", + "RequestInfo", + # connector + "BaseConnector", + "TCPConnector", + "UnixConnector", + "NamedPipeConnector", + # client_ws + "ClientWebSocketResponse", + # client + "ClientSession", + "ClientTimeout", + "request", +) + + +try: + from ssl import SSLContext +except ImportError: # pragma: no cover + SSLContext = object # type: ignore[misc,assignment] + + +@attr.s(auto_attribs=True, frozen=True, slots=True) +class ClientTimeout: + total: Optional[float] = None + connect: Optional[float] = None + sock_read: Optional[float] = None + sock_connect: Optional[float] = None + + # pool_queue_timeout: Optional[float] = None + # dns_resolution_timeout: Optional[float] = None + # socket_connect_timeout: Optional[float] = None + # connection_acquiring_timeout: Optional[float] = None + # new_connection_timeout: Optional[float] = None + # http_header_timeout: Optional[float] = None + # response_body_timeout: Optional[float] = None + + # to create a timeout specific for a single request, either + # - create a completely new one to overwrite the default + # - or use http://www.attrs.org/en/stable/api.html#attr.evolve + # to overwrite the defaults + + +# 5 Minute default read timeout +DEFAULT_TIMEOUT: Final[ClientTimeout] = ClientTimeout(total=5 * 60) + +_RetType = TypeVar("_RetType") + + +class ClientSession: + """First-class interface for making HTTP requests.""" + + ATTRS = frozenset( + [ + "_base_url", + "_source_traceback", + "_connector", + "requote_redirect_url", + "_loop", + "_cookie_jar", + "_connector_owner", + "_default_auth", + "_version", + "_json_serialize", + "_requote_redirect_url", + "_timeout", + "_raise_for_status", + "_auto_decompress", + "_trust_env", + "_default_headers", + "_skip_auto_headers", + "_request_class", + "_response_class", + "_ws_response_class", + "_trace_configs", + "_read_bufsize", + ] + ) + + _source_traceback = None + + def __init__( + self, + base_url: Optional[StrOrURL] = None, + *, + connector: Optional[BaseConnector] = None, + loop: Optional[asyncio.AbstractEventLoop] = None, + cookies: Optional[LooseCookies] = None, + headers: Optional[LooseHeaders] = None, + skip_auto_headers: Optional[Iterable[str]] = None, + auth: Optional[BasicAuth] = None, + json_serialize: JSONEncoder = json.dumps, + request_class: Type[ClientRequest] = ClientRequest, + response_class: Type[ClientResponse] = ClientResponse, + ws_response_class: Type[ClientWebSocketResponse] = ClientWebSocketResponse, + version: HttpVersion = http.HttpVersion11, + cookie_jar: Optional[AbstractCookieJar] = None, + connector_owner: bool = True, + raise_for_status: bool = False, + read_timeout: Union[float, object] = sentinel, + conn_timeout: Optional[float] = None, + timeout: Union[object, ClientTimeout] = sentinel, + auto_decompress: bool = True, + trust_env: bool = False, + requote_redirect_url: bool = True, + trace_configs: Optional[List[TraceConfig]] = None, + read_bufsize: int = 2 ** 16, + ) -> None: + if loop is None: + if connector is not None: + loop = connector._loop + + loop = get_running_loop(loop) + + if base_url is None or isinstance(base_url, URL): + self._base_url: Optional[URL] = base_url + else: + self._base_url = URL(base_url) + assert ( + self._base_url.origin() == self._base_url + ), "Only absolute URLs without path part are supported" + + if connector is None: + connector = TCPConnector(loop=loop) + + if connector._loop is not loop: + raise RuntimeError("Session and connector has to use same event loop") + + self._loop = loop + + if loop.get_debug(): + self._source_traceback = traceback.extract_stack(sys._getframe(1)) + + if cookie_jar is None: + cookie_jar = CookieJar(loop=loop) + self._cookie_jar = cookie_jar + + if cookies is not None: + self._cookie_jar.update_cookies(cookies) + + self._connector = connector # type: Optional[BaseConnector] + self._connector_owner = connector_owner + self._default_auth = auth + self._version = version + self._json_serialize = json_serialize + if timeout is sentinel: + self._timeout = DEFAULT_TIMEOUT + if read_timeout is not sentinel: + warnings.warn( + "read_timeout is deprecated, " "use timeout argument instead", + DeprecationWarning, + stacklevel=2, + ) + self._timeout = attr.evolve(self._timeout, total=read_timeout) + if conn_timeout is not None: + self._timeout = attr.evolve(self._timeout, connect=conn_timeout) + warnings.warn( + "conn_timeout is deprecated, " "use timeout argument instead", + DeprecationWarning, + stacklevel=2, + ) + else: + self._timeout = timeout # type: ignore[assignment] + if read_timeout is not sentinel: + raise ValueError( + "read_timeout and timeout parameters " + "conflict, please setup " + "timeout.read" + ) + if conn_timeout is not None: + raise ValueError( + "conn_timeout and timeout parameters " + "conflict, please setup " + "timeout.connect" + ) + self._raise_for_status = raise_for_status + self._auto_decompress = auto_decompress + self._trust_env = trust_env + self._requote_redirect_url = requote_redirect_url + self._read_bufsize = read_bufsize + + # Convert to list of tuples + if headers: + real_headers = CIMultiDict(headers) # type: CIMultiDict[str] + else: + real_headers = CIMultiDict() + self._default_headers = real_headers # type: CIMultiDict[str] + if skip_auto_headers is not None: + self._skip_auto_headers = frozenset(istr(i) for i in skip_auto_headers) + else: + self._skip_auto_headers = frozenset() + + self._request_class = request_class + self._response_class = response_class + self._ws_response_class = ws_response_class + + self._trace_configs = trace_configs or [] + for trace_config in self._trace_configs: + trace_config.freeze() + + def __init_subclass__(cls: Type["ClientSession"]) -> None: + warnings.warn( + "Inheritance class {} from ClientSession " + "is discouraged".format(cls.__name__), + DeprecationWarning, + stacklevel=2, + ) + + if DEBUG: + + def __setattr__(self, name: str, val: Any) -> None: + if name not in self.ATTRS: + warnings.warn( + "Setting custom ClientSession.{} attribute " + "is discouraged".format(name), + DeprecationWarning, + stacklevel=2, + ) + super().__setattr__(name, val) + + def __del__(self, _warnings: Any = warnings) -> None: + if not self.closed: + if PY_36: + kwargs = {"source": self} + else: + kwargs = {} + _warnings.warn( + f"Unclosed client session {self!r}", ResourceWarning, **kwargs + ) + context = {"client_session": self, "message": "Unclosed client session"} + if self._source_traceback is not None: + context["source_traceback"] = self._source_traceback + self._loop.call_exception_handler(context) + + def request( + self, method: str, url: StrOrURL, **kwargs: Any + ) -> "_RequestContextManager": + """Perform HTTP request.""" + return _RequestContextManager(self._request(method, url, **kwargs)) + + def _build_url(self, str_or_url: StrOrURL) -> URL: + url = URL(str_or_url) + if self._base_url is None: + return url + else: + assert not url.is_absolute() and url.path.startswith("/") + return self._base_url.join(url) + + async def _request( + self, + method: str, + str_or_url: StrOrURL, + *, + params: Optional[Mapping[str, str]] = None, + data: Any = None, + json: Any = None, + cookies: Optional[LooseCookies] = None, + headers: Optional[LooseHeaders] = None, + skip_auto_headers: Optional[Iterable[str]] = None, + auth: Optional[BasicAuth] = None, + allow_redirects: bool = True, + max_redirects: int = 10, + compress: Optional[str] = None, + chunked: Optional[bool] = None, + expect100: bool = False, + raise_for_status: Optional[bool] = None, + read_until_eof: bool = True, + proxy: Optional[StrOrURL] = None, + proxy_auth: Optional[BasicAuth] = None, + timeout: Union[ClientTimeout, object] = sentinel, + verify_ssl: Optional[bool] = None, + fingerprint: Optional[bytes] = None, + ssl_context: Optional[SSLContext] = None, + ssl: Optional[Union[SSLContext, bool, Fingerprint]] = None, + proxy_headers: Optional[LooseHeaders] = None, + trace_request_ctx: Optional[SimpleNamespace] = None, + read_bufsize: Optional[int] = None, + ) -> ClientResponse: + + # NOTE: timeout clamps existing connect and read timeouts. We cannot + # set the default to None because we need to detect if the user wants + # to use the existing timeouts by setting timeout to None. + + if self.closed: + raise RuntimeError("Session is closed") + + ssl = _merge_ssl_params(ssl, verify_ssl, ssl_context, fingerprint) + + if data is not None and json is not None: + raise ValueError( + "data and json parameters can not be used at the same time" + ) + elif json is not None: + data = payload.JsonPayload(json, dumps=self._json_serialize) + + if not isinstance(chunked, bool) and chunked is not None: + warnings.warn("Chunk size is deprecated #1615", DeprecationWarning) + + redirects = 0 + history = [] + version = self._version + + # Merge with default headers and transform to CIMultiDict + headers = self._prepare_headers(headers) + proxy_headers = self._prepare_headers(proxy_headers) + + try: + url = self._build_url(str_or_url) + except ValueError as e: + raise InvalidURL(str_or_url) from e + + skip_headers = set(self._skip_auto_headers) + if skip_auto_headers is not None: + for i in skip_auto_headers: + skip_headers.add(istr(i)) + + if proxy is not None: + try: + proxy = URL(proxy) + except ValueError as e: + raise InvalidURL(proxy) from e + + if timeout is sentinel: + real_timeout = self._timeout # type: ClientTimeout + else: + if not isinstance(timeout, ClientTimeout): + real_timeout = ClientTimeout(total=timeout) # type: ignore[arg-type] + else: + real_timeout = timeout + # timeout is cumulative for all request operations + # (request, redirects, responses, data consuming) + tm = TimeoutHandle(self._loop, real_timeout.total) + handle = tm.start() + + if read_bufsize is None: + read_bufsize = self._read_bufsize + + traces = [ + Trace( + self, + trace_config, + trace_config.trace_config_ctx(trace_request_ctx=trace_request_ctx), + ) + for trace_config in self._trace_configs + ] + + for trace in traces: + await trace.send_request_start(method, url.update_query(params), headers) + + timer = tm.timer() + try: + with timer: + while True: + url, auth_from_url = strip_auth_from_url(url) + if auth and auth_from_url: + raise ValueError( + "Cannot combine AUTH argument with " + "credentials encoded in URL" + ) + + if auth is None: + auth = auth_from_url + if auth is None: + auth = self._default_auth + # It would be confusing if we support explicit + # Authorization header with auth argument + if ( + headers is not None + and auth is not None + and hdrs.AUTHORIZATION in headers + ): + raise ValueError( + "Cannot combine AUTHORIZATION header " + "with AUTH argument or credentials " + "encoded in URL" + ) + + all_cookies = self._cookie_jar.filter_cookies(url) + + if cookies is not None: + tmp_cookie_jar = CookieJar() + tmp_cookie_jar.update_cookies(cookies) + req_cookies = tmp_cookie_jar.filter_cookies(url) + if req_cookies: + all_cookies.load(req_cookies) + + if proxy is not None: + proxy = URL(proxy) + elif self._trust_env: + with suppress(LookupError): + proxy, proxy_auth = get_env_proxy_for_url(url) + + req = self._request_class( + method, + url, + params=params, + headers=headers, + skip_auto_headers=skip_headers, + data=data, + cookies=all_cookies, + auth=auth, + version=version, + compress=compress, + chunked=chunked, + expect100=expect100, + loop=self._loop, + response_class=self._response_class, + proxy=proxy, + proxy_auth=proxy_auth, + timer=timer, + session=self, + ssl=ssl, + proxy_headers=proxy_headers, + traces=traces, + ) + + # connection timeout + try: + async with ceil_timeout(real_timeout.connect): + assert self._connector is not None + conn = await self._connector.connect( + req, traces=traces, timeout=real_timeout + ) + except asyncio.TimeoutError as exc: + raise ServerTimeoutError( + "Connection timeout " "to host {}".format(url) + ) from exc + + assert conn.transport is not None + + assert conn.protocol is not None + conn.protocol.set_response_params( + timer=timer, + skip_payload=method.upper() == "HEAD", + read_until_eof=read_until_eof, + auto_decompress=self._auto_decompress, + read_timeout=real_timeout.sock_read, + read_bufsize=read_bufsize, + ) + + try: + try: + resp = await req.send(conn) + try: + await resp.start(conn) + except BaseException: + resp.close() + raise + except BaseException: + conn.close() + raise + except ClientError: + raise + except OSError as exc: + if exc.errno is None and isinstance(exc, asyncio.TimeoutError): + raise + raise ClientOSError(*exc.args) from exc + + self._cookie_jar.update_cookies(resp.cookies, resp.url) + + # redirects + if resp.status in (301, 302, 303, 307, 308) and allow_redirects: + + for trace in traces: + await trace.send_request_redirect( + method, url.update_query(params), headers, resp + ) + + redirects += 1 + history.append(resp) + if max_redirects and redirects >= max_redirects: + resp.close() + raise TooManyRedirects( + history[0].request_info, tuple(history) + ) + + # For 301 and 302, mimic IE, now changed in RFC + # https://github.com/kennethreitz/requests/pull/269 + if (resp.status == 303 and resp.method != hdrs.METH_HEAD) or ( + resp.status in (301, 302) and resp.method == hdrs.METH_POST + ): + method = hdrs.METH_GET + data = None + if headers.get(hdrs.CONTENT_LENGTH): + headers.pop(hdrs.CONTENT_LENGTH) + + r_url = resp.headers.get(hdrs.LOCATION) or resp.headers.get( + hdrs.URI + ) + if r_url is None: + # see github.com/aio-libs/aiohttp/issues/2022 + break + else: + # reading from correct redirection + # response is forbidden + resp.release() + + try: + parsed_url = URL( + r_url, encoded=not self._requote_redirect_url + ) + + except ValueError as e: + raise InvalidURL(r_url) from e + + scheme = parsed_url.scheme + if scheme not in ("http", "https", ""): + resp.close() + raise ValueError("Can redirect only to http or https") + elif not scheme: + parsed_url = url.join(parsed_url) + + if url.origin() != parsed_url.origin(): + auth = None + headers.pop(hdrs.AUTHORIZATION, None) + + url = parsed_url + params = None + resp.release() + continue + + break + + # check response status + if raise_for_status is None: + raise_for_status = self._raise_for_status + if raise_for_status: + resp.raise_for_status() + + # register connection + if handle is not None: + if resp.connection is not None: + resp.connection.add_callback(handle.cancel) + else: + handle.cancel() + + resp._history = tuple(history) + + for trace in traces: + await trace.send_request_end( + method, url.update_query(params), headers, resp + ) + return resp + + except BaseException as e: + # cleanup timer + tm.close() + if handle: + handle.cancel() + handle = None + + for trace in traces: + await trace.send_request_exception( + method, url.update_query(params), headers, e + ) + raise + + def ws_connect( + self, + url: StrOrURL, + *, + method: str = hdrs.METH_GET, + protocols: Iterable[str] = (), + timeout: float = 10.0, + receive_timeout: Optional[float] = None, + autoclose: bool = True, + autoping: bool = True, + heartbeat: Optional[float] = None, + auth: Optional[BasicAuth] = None, + origin: Optional[str] = None, + params: Optional[Mapping[str, str]] = None, + headers: Optional[LooseHeaders] = None, + proxy: Optional[StrOrURL] = None, + proxy_auth: Optional[BasicAuth] = None, + ssl: Union[SSLContext, bool, None, Fingerprint] = None, + verify_ssl: Optional[bool] = None, + fingerprint: Optional[bytes] = None, + ssl_context: Optional[SSLContext] = None, + proxy_headers: Optional[LooseHeaders] = None, + compress: int = 0, + max_msg_size: int = 4 * 1024 * 1024, + ) -> "_WSRequestContextManager": + """Initiate websocket connection.""" + return _WSRequestContextManager( + self._ws_connect( + url, + method=method, + protocols=protocols, + timeout=timeout, + receive_timeout=receive_timeout, + autoclose=autoclose, + autoping=autoping, + heartbeat=heartbeat, + auth=auth, + origin=origin, + params=params, + headers=headers, + proxy=proxy, + proxy_auth=proxy_auth, + ssl=ssl, + verify_ssl=verify_ssl, + fingerprint=fingerprint, + ssl_context=ssl_context, + proxy_headers=proxy_headers, + compress=compress, + max_msg_size=max_msg_size, + ) + ) + + async def _ws_connect( + self, + url: StrOrURL, + *, + method: str = hdrs.METH_GET, + protocols: Iterable[str] = (), + timeout: float = 10.0, + receive_timeout: Optional[float] = None, + autoclose: bool = True, + autoping: bool = True, + heartbeat: Optional[float] = None, + auth: Optional[BasicAuth] = None, + origin: Optional[str] = None, + params: Optional[Mapping[str, str]] = None, + headers: Optional[LooseHeaders] = None, + proxy: Optional[StrOrURL] = None, + proxy_auth: Optional[BasicAuth] = None, + ssl: Union[SSLContext, bool, None, Fingerprint] = None, + verify_ssl: Optional[bool] = None, + fingerprint: Optional[bytes] = None, + ssl_context: Optional[SSLContext] = None, + proxy_headers: Optional[LooseHeaders] = None, + compress: int = 0, + max_msg_size: int = 4 * 1024 * 1024, + ) -> ClientWebSocketResponse: + + if headers is None: + real_headers = CIMultiDict() # type: CIMultiDict[str] + else: + real_headers = CIMultiDict(headers) + + default_headers = { + hdrs.UPGRADE: "websocket", + hdrs.CONNECTION: "upgrade", + hdrs.SEC_WEBSOCKET_VERSION: "13", + } + + for key, value in default_headers.items(): + real_headers.setdefault(key, value) + + sec_key = base64.b64encode(os.urandom(16)) + real_headers[hdrs.SEC_WEBSOCKET_KEY] = sec_key.decode() + + if protocols: + real_headers[hdrs.SEC_WEBSOCKET_PROTOCOL] = ",".join(protocols) + if origin is not None: + real_headers[hdrs.ORIGIN] = origin + if compress: + extstr = ws_ext_gen(compress=compress) + real_headers[hdrs.SEC_WEBSOCKET_EXTENSIONS] = extstr + + ssl = _merge_ssl_params(ssl, verify_ssl, ssl_context, fingerprint) + + # send request + resp = await self.request( + method, + url, + params=params, + headers=real_headers, + read_until_eof=False, + auth=auth, + proxy=proxy, + proxy_auth=proxy_auth, + ssl=ssl, + proxy_headers=proxy_headers, + ) + + try: + # check handshake + if resp.status != 101: + raise WSServerHandshakeError( + resp.request_info, + resp.history, + message="Invalid response status", + status=resp.status, + headers=resp.headers, + ) + + if resp.headers.get(hdrs.UPGRADE, "").lower() != "websocket": + raise WSServerHandshakeError( + resp.request_info, + resp.history, + message="Invalid upgrade header", + status=resp.status, + headers=resp.headers, + ) + + if resp.headers.get(hdrs.CONNECTION, "").lower() != "upgrade": + raise WSServerHandshakeError( + resp.request_info, + resp.history, + message="Invalid connection header", + status=resp.status, + headers=resp.headers, + ) + + # key calculation + r_key = resp.headers.get(hdrs.SEC_WEBSOCKET_ACCEPT, "") + match = base64.b64encode(hashlib.sha1(sec_key + WS_KEY).digest()).decode() + if r_key != match: + raise WSServerHandshakeError( + resp.request_info, + resp.history, + message="Invalid challenge response", + status=resp.status, + headers=resp.headers, + ) + + # websocket protocol + protocol = None + if protocols and hdrs.SEC_WEBSOCKET_PROTOCOL in resp.headers: + resp_protocols = [ + proto.strip() + for proto in resp.headers[hdrs.SEC_WEBSOCKET_PROTOCOL].split(",") + ] + + for proto in resp_protocols: + if proto in protocols: + protocol = proto + break + + # websocket compress + notakeover = False + if compress: + compress_hdrs = resp.headers.get(hdrs.SEC_WEBSOCKET_EXTENSIONS) + if compress_hdrs: + try: + compress, notakeover = ws_ext_parse(compress_hdrs) + except WSHandshakeError as exc: + raise WSServerHandshakeError( + resp.request_info, + resp.history, + message=exc.args[0], + status=resp.status, + headers=resp.headers, + ) from exc + else: + compress = 0 + notakeover = False + + conn = resp.connection + assert conn is not None + conn_proto = conn.protocol + assert conn_proto is not None + transport = conn.transport + assert transport is not None + reader = FlowControlDataQueue( + conn_proto, 2 ** 16, loop=self._loop + ) # type: FlowControlDataQueue[WSMessage] + conn_proto.set_parser(WebSocketReader(reader, max_msg_size), reader) + writer = WebSocketWriter( + conn_proto, + transport, + use_mask=True, + compress=compress, + notakeover=notakeover, + ) + except BaseException: + resp.close() + raise + else: + return self._ws_response_class( + reader, + writer, + protocol, + resp, + timeout, + autoclose, + autoping, + self._loop, + receive_timeout=receive_timeout, + heartbeat=heartbeat, + compress=compress, + client_notakeover=notakeover, + ) + + def _prepare_headers(self, headers: Optional[LooseHeaders]) -> "CIMultiDict[str]": + """Add default headers and transform it to CIMultiDict""" + # Convert headers to MultiDict + result = CIMultiDict(self._default_headers) + if headers: + if not isinstance(headers, (MultiDictProxy, MultiDict)): + headers = CIMultiDict(headers) + added_names = set() # type: Set[str] + for key, value in headers.items(): + if key in added_names: + result.add(key, value) + else: + result[key] = value + added_names.add(key) + return result + + def get( + self, url: StrOrURL, *, allow_redirects: bool = True, **kwargs: Any + ) -> "_RequestContextManager": + """Perform HTTP GET request.""" + return _RequestContextManager( + self._request(hdrs.METH_GET, url, allow_redirects=allow_redirects, **kwargs) + ) + + def options( + self, url: StrOrURL, *, allow_redirects: bool = True, **kwargs: Any + ) -> "_RequestContextManager": + """Perform HTTP OPTIONS request.""" + return _RequestContextManager( + self._request( + hdrs.METH_OPTIONS, url, allow_redirects=allow_redirects, **kwargs + ) + ) + + def head( + self, url: StrOrURL, *, allow_redirects: bool = False, **kwargs: Any + ) -> "_RequestContextManager": + """Perform HTTP HEAD request.""" + return _RequestContextManager( + self._request( + hdrs.METH_HEAD, url, allow_redirects=allow_redirects, **kwargs + ) + ) + + def post( + self, url: StrOrURL, *, data: Any = None, **kwargs: Any + ) -> "_RequestContextManager": + """Perform HTTP POST request.""" + return _RequestContextManager( + self._request(hdrs.METH_POST, url, data=data, **kwargs) + ) + + def put( + self, url: StrOrURL, *, data: Any = None, **kwargs: Any + ) -> "_RequestContextManager": + """Perform HTTP PUT request.""" + return _RequestContextManager( + self._request(hdrs.METH_PUT, url, data=data, **kwargs) + ) + + def patch( + self, url: StrOrURL, *, data: Any = None, **kwargs: Any + ) -> "_RequestContextManager": + """Perform HTTP PATCH request.""" + return _RequestContextManager( + self._request(hdrs.METH_PATCH, url, data=data, **kwargs) + ) + + def delete(self, url: StrOrURL, **kwargs: Any) -> "_RequestContextManager": + """Perform HTTP DELETE request.""" + return _RequestContextManager(self._request(hdrs.METH_DELETE, url, **kwargs)) + + async def close(self) -> None: + """Close underlying connector. + + Release all acquired resources. + """ + if not self.closed: + if self._connector is not None and self._connector_owner: + await self._connector.close() + self._connector = None + + @property + def closed(self) -> bool: + """Is client session closed. + + A readonly property. + """ + return self._connector is None or self._connector.closed + + @property + def connector(self) -> Optional[BaseConnector]: + """Connector instance used for the session.""" + return self._connector + + @property + def cookie_jar(self) -> AbstractCookieJar: + """The session cookies.""" + return self._cookie_jar + + @property + def version(self) -> Tuple[int, int]: + """The session HTTP protocol version.""" + return self._version + + @property + def requote_redirect_url(self) -> bool: + """Do URL requoting on redirection handling.""" + return self._requote_redirect_url + + @requote_redirect_url.setter + def requote_redirect_url(self, val: bool) -> None: + """Do URL requoting on redirection handling.""" + warnings.warn( + "session.requote_redirect_url modification " "is deprecated #2778", + DeprecationWarning, + stacklevel=2, + ) + self._requote_redirect_url = val + + @property + def loop(self) -> asyncio.AbstractEventLoop: + """Session's loop.""" + warnings.warn( + "client.loop property is deprecated", DeprecationWarning, stacklevel=2 + ) + return self._loop + + @property + def timeout(self) -> Union[object, ClientTimeout]: + """Timeout for the session.""" + return self._timeout + + @property + def headers(self) -> "CIMultiDict[str]": + """The default headers of the client session.""" + return self._default_headers + + @property + def skip_auto_headers(self) -> FrozenSet[istr]: + """Headers for which autogeneration should be skipped""" + return self._skip_auto_headers + + @property + def auth(self) -> Optional[BasicAuth]: + """An object that represents HTTP Basic Authorization""" + return self._default_auth + + @property + def json_serialize(self) -> JSONEncoder: + """Json serializer callable""" + return self._json_serialize + + @property + def connector_owner(self) -> bool: + """Should connector be closed on session closing""" + return self._connector_owner + + @property + def raise_for_status( + self, + ) -> Union[bool, Callable[[ClientResponse], Awaitable[None]]]: + """Should `ClientResponse.raise_for_status()` be called for each response.""" + return self._raise_for_status + + @property + def auto_decompress(self) -> bool: + """Should the body response be automatically decompressed.""" + return self._auto_decompress + + @property + def trust_env(self) -> bool: + """ + Should proxies information from environment or netrc be trusted. + + Information is from HTTP_PROXY / HTTPS_PROXY environment variables + or ~/.netrc file if present. + """ + return self._trust_env + + @property + def trace_configs(self) -> List[TraceConfig]: + """A list of TraceConfig instances used for client tracing""" + return self._trace_configs + + def detach(self) -> None: + """Detach connector from session without closing the former. + + Session is switched to closed state anyway. + """ + self._connector = None + + def __enter__(self) -> None: + raise TypeError("Use async with instead") + + def __exit__( + self, + exc_type: Optional[Type[BaseException]], + exc_val: Optional[BaseException], + exc_tb: Optional[TracebackType], + ) -> None: + # __exit__ should exist in pair with __enter__ but never executed + pass # pragma: no cover + + async def __aenter__(self) -> "ClientSession": + return self + + async def __aexit__( + self, + exc_type: Optional[Type[BaseException]], + exc_val: Optional[BaseException], + exc_tb: Optional[TracebackType], + ) -> None: + await self.close() + + +class _BaseRequestContextManager(Coroutine[Any, Any, _RetType], Generic[_RetType]): + + __slots__ = ("_coro", "_resp") + + def __init__(self, coro: Coroutine["asyncio.Future[Any]", None, _RetType]) -> None: + self._coro = coro + + def send(self, arg: None) -> "asyncio.Future[Any]": + return self._coro.send(arg) + + def throw(self, arg: BaseException) -> None: # type: ignore[arg-type,override] + self._coro.throw(arg) + + def close(self) -> None: + return self._coro.close() + + def __await__(self) -> Generator[Any, None, _RetType]: + ret = self._coro.__await__() + return ret + + def __iter__(self) -> Generator[Any, None, _RetType]: + return self.__await__() + + async def __aenter__(self) -> _RetType: + self._resp = await self._coro + return self._resp + + +class _RequestContextManager(_BaseRequestContextManager[ClientResponse]): + __slots__ = () + + async def __aexit__( + self, + exc_type: Optional[Type[BaseException]], + exc: Optional[BaseException], + tb: Optional[TracebackType], + ) -> None: + # We're basing behavior on the exception as it can be caused by + # user code unrelated to the status of the connection. If you + # would like to close a connection you must do that + # explicitly. Otherwise connection error handling should kick in + # and close/recycle the connection as required. + self._resp.release() + + +class _WSRequestContextManager(_BaseRequestContextManager[ClientWebSocketResponse]): + __slots__ = () + + async def __aexit__( + self, + exc_type: Optional[Type[BaseException]], + exc: Optional[BaseException], + tb: Optional[TracebackType], + ) -> None: + await self._resp.close() + + +class _SessionRequestContextManager: + + __slots__ = ("_coro", "_resp", "_session") + + def __init__( + self, + coro: Coroutine["asyncio.Future[Any]", None, ClientResponse], + session: ClientSession, + ) -> None: + self._coro = coro + self._resp = None # type: Optional[ClientResponse] + self._session = session + + async def __aenter__(self) -> ClientResponse: + try: + self._resp = await self._coro + except BaseException: + await self._session.close() + raise + else: + return self._resp + + async def __aexit__( + self, + exc_type: Optional[Type[BaseException]], + exc: Optional[BaseException], + tb: Optional[TracebackType], + ) -> None: + assert self._resp is not None + self._resp.close() + await self._session.close() + + +def request( + method: str, + url: StrOrURL, + *, + params: Optional[Mapping[str, str]] = None, + data: Any = None, + json: Any = None, + headers: Optional[LooseHeaders] = None, + skip_auto_headers: Optional[Iterable[str]] = None, + auth: Optional[BasicAuth] = None, + allow_redirects: bool = True, + max_redirects: int = 10, + compress: Optional[str] = None, + chunked: Optional[bool] = None, + expect100: bool = False, + raise_for_status: Optional[bool] = None, + read_until_eof: bool = True, + proxy: Optional[StrOrURL] = None, + proxy_auth: Optional[BasicAuth] = None, + timeout: Union[ClientTimeout, object] = sentinel, + cookies: Optional[LooseCookies] = None, + version: HttpVersion = http.HttpVersion11, + connector: Optional[BaseConnector] = None, + read_bufsize: Optional[int] = None, + loop: Optional[asyncio.AbstractEventLoop] = None, +) -> _SessionRequestContextManager: + """Constructs and sends a request. + + Returns response object. + method - HTTP method + url - request url + params - (optional) Dictionary or bytes to be sent in the query + string of the new request + data - (optional) Dictionary, bytes, or file-like object to + send in the body of the request + json - (optional) Any json compatible python object + headers - (optional) Dictionary of HTTP Headers to send with + the request + cookies - (optional) Dict object to send with the request + auth - (optional) BasicAuth named tuple represent HTTP Basic Auth + auth - aiohttp.helpers.BasicAuth + allow_redirects - (optional) If set to False, do not follow + redirects + version - Request HTTP version. + compress - Set to True if request has to be compressed + with deflate encoding. + chunked - Set to chunk size for chunked transfer encoding. + expect100 - Expect 100-continue response from server. + connector - BaseConnector sub-class instance to support + connection pooling. + read_until_eof - Read response until eof if response + does not have Content-Length header. + loop - Optional event loop. + timeout - Optional ClientTimeout settings structure, 5min + total timeout by default. + Usage:: + >>> import aiohttp + >>> resp = await aiohttp.request('GET', 'http://python.org/') + >>> resp + <ClientResponse(python.org/) [200]> + >>> data = await resp.read() + """ + connector_owner = False + if connector is None: + connector_owner = True + connector = TCPConnector(loop=loop, force_close=True) + + session = ClientSession( + loop=loop, + cookies=cookies, + version=version, + timeout=timeout, + connector=connector, + connector_owner=connector_owner, + ) + + return _SessionRequestContextManager( + session._request( + method, + url, + params=params, + data=data, + json=json, + headers=headers, + skip_auto_headers=skip_auto_headers, + auth=auth, + allow_redirects=allow_redirects, + max_redirects=max_redirects, + compress=compress, + chunked=chunked, + expect100=expect100, + raise_for_status=raise_for_status, + read_until_eof=read_until_eof, + proxy=proxy, + proxy_auth=proxy_auth, + read_bufsize=read_bufsize, + ), + session, + ) diff --git a/contrib/python/aiohttp/aiohttp/client_exceptions.py b/contrib/python/aiohttp/aiohttp/client_exceptions.py new file mode 100644 index 0000000000..dd55321054 --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/client_exceptions.py @@ -0,0 +1,342 @@ +"""HTTP related errors.""" + +import asyncio +import warnings +from typing import TYPE_CHECKING, Any, Optional, Tuple, Union + +from .http_parser import RawResponseMessage +from .typedefs import LooseHeaders + +try: + import ssl + + SSLContext = ssl.SSLContext +except ImportError: # pragma: no cover + ssl = SSLContext = None # type: ignore[assignment] + + +if TYPE_CHECKING: # pragma: no cover + from .client_reqrep import ClientResponse, ConnectionKey, Fingerprint, RequestInfo +else: + RequestInfo = ClientResponse = ConnectionKey = None + +__all__ = ( + "ClientError", + "ClientConnectionError", + "ClientOSError", + "ClientConnectorError", + "ClientProxyConnectionError", + "ClientSSLError", + "ClientConnectorSSLError", + "ClientConnectorCertificateError", + "ServerConnectionError", + "ServerTimeoutError", + "ServerDisconnectedError", + "ServerFingerprintMismatch", + "ClientResponseError", + "ClientHttpProxyError", + "WSServerHandshakeError", + "ContentTypeError", + "ClientPayloadError", + "InvalidURL", +) + + +class ClientError(Exception): + """Base class for client connection errors.""" + + +class ClientResponseError(ClientError): + """Connection error during reading response. + + request_info: instance of RequestInfo + """ + + def __init__( + self, + request_info: RequestInfo, + history: Tuple[ClientResponse, ...], + *, + code: Optional[int] = None, + status: Optional[int] = None, + message: str = "", + headers: Optional[LooseHeaders] = None, + ) -> None: + self.request_info = request_info + if code is not None: + if status is not None: + raise ValueError( + "Both code and status arguments are provided; " + "code is deprecated, use status instead" + ) + warnings.warn( + "code argument is deprecated, use status instead", + DeprecationWarning, + stacklevel=2, + ) + if status is not None: + self.status = status + elif code is not None: + self.status = code + else: + self.status = 0 + self.message = message + self.headers = headers + self.history = history + self.args = (request_info, history) + + def __str__(self) -> str: + return "{}, message={!r}, url={!r}".format( + self.status, + self.message, + self.request_info.real_url, + ) + + def __repr__(self) -> str: + args = f"{self.request_info!r}, {self.history!r}" + if self.status != 0: + args += f", status={self.status!r}" + if self.message != "": + args += f", message={self.message!r}" + if self.headers is not None: + args += f", headers={self.headers!r}" + return f"{type(self).__name__}({args})" + + @property + def code(self) -> int: + warnings.warn( + "code property is deprecated, use status instead", + DeprecationWarning, + stacklevel=2, + ) + return self.status + + @code.setter + def code(self, value: int) -> None: + warnings.warn( + "code property is deprecated, use status instead", + DeprecationWarning, + stacklevel=2, + ) + self.status = value + + +class ContentTypeError(ClientResponseError): + """ContentType found is not valid.""" + + +class WSServerHandshakeError(ClientResponseError): + """websocket server handshake error.""" + + +class ClientHttpProxyError(ClientResponseError): + """HTTP proxy error. + + Raised in :class:`aiohttp.connector.TCPConnector` if + proxy responds with status other than ``200 OK`` + on ``CONNECT`` request. + """ + + +class TooManyRedirects(ClientResponseError): + """Client was redirected too many times.""" + + +class ClientConnectionError(ClientError): + """Base class for client socket errors.""" + + +class ClientOSError(ClientConnectionError, OSError): + """OSError error.""" + + +class ClientConnectorError(ClientOSError): + """Client connector error. + + Raised in :class:`aiohttp.connector.TCPConnector` if + connection to proxy can not be established. + """ + + def __init__(self, connection_key: ConnectionKey, os_error: OSError) -> None: + self._conn_key = connection_key + self._os_error = os_error + super().__init__(os_error.errno, os_error.strerror) + self.args = (connection_key, os_error) + + @property + def os_error(self) -> OSError: + return self._os_error + + @property + def host(self) -> str: + return self._conn_key.host + + @property + def port(self) -> Optional[int]: + return self._conn_key.port + + @property + def ssl(self) -> Union[SSLContext, None, bool, "Fingerprint"]: + return self._conn_key.ssl + + def __str__(self) -> str: + return "Cannot connect to host {0.host}:{0.port} ssl:{1} [{2}]".format( + self, self.ssl if self.ssl is not None else "default", self.strerror + ) + + # OSError.__reduce__ does too much black magick + __reduce__ = BaseException.__reduce__ + + +class ClientProxyConnectionError(ClientConnectorError): + """Proxy connection error. + + Raised in :class:`aiohttp.connector.TCPConnector` if + connection to proxy can not be established. + """ + + +class UnixClientConnectorError(ClientConnectorError): + """Unix connector error. + + Raised in :py:class:`aiohttp.connector.UnixConnector` + if connection to unix socket can not be established. + """ + + def __init__( + self, path: str, connection_key: ConnectionKey, os_error: OSError + ) -> None: + self._path = path + super().__init__(connection_key, os_error) + + @property + def path(self) -> str: + return self._path + + def __str__(self) -> str: + return "Cannot connect to unix socket {0.path} ssl:{1} [{2}]".format( + self, self.ssl if self.ssl is not None else "default", self.strerror + ) + + +class ServerConnectionError(ClientConnectionError): + """Server connection errors.""" + + +class ServerDisconnectedError(ServerConnectionError): + """Server disconnected.""" + + def __init__(self, message: Union[RawResponseMessage, str, None] = None) -> None: + if message is None: + message = "Server disconnected" + + self.args = (message,) + self.message = message + + +class ServerTimeoutError(ServerConnectionError, asyncio.TimeoutError): + """Server timeout error.""" + + +class ServerFingerprintMismatch(ServerConnectionError): + """SSL certificate does not match expected fingerprint.""" + + def __init__(self, expected: bytes, got: bytes, host: str, port: int) -> None: + self.expected = expected + self.got = got + self.host = host + self.port = port + self.args = (expected, got, host, port) + + def __repr__(self) -> str: + return "<{} expected={!r} got={!r} host={!r} port={!r}>".format( + self.__class__.__name__, self.expected, self.got, self.host, self.port + ) + + +class ClientPayloadError(ClientError): + """Response payload error.""" + + +class InvalidURL(ClientError, ValueError): + """Invalid URL. + + URL used for fetching is malformed, e.g. it doesn't contains host + part. + """ + + # Derive from ValueError for backward compatibility + + def __init__(self, url: Any) -> None: + # The type of url is not yarl.URL because the exception can be raised + # on URL(url) call + super().__init__(url) + + @property + def url(self) -> Any: + return self.args[0] + + def __repr__(self) -> str: + return f"<{self.__class__.__name__} {self.url}>" + + +class ClientSSLError(ClientConnectorError): + """Base error for ssl.*Errors.""" + + +if ssl is not None: + cert_errors = (ssl.CertificateError,) + cert_errors_bases = ( + ClientSSLError, + ssl.CertificateError, + ) + + ssl_errors = (ssl.SSLError,) + ssl_error_bases = (ClientSSLError, ssl.SSLError) +else: # pragma: no cover + cert_errors = tuple() + cert_errors_bases = ( + ClientSSLError, + ValueError, + ) + + ssl_errors = tuple() + ssl_error_bases = (ClientSSLError,) + + +class ClientConnectorSSLError(*ssl_error_bases): # type: ignore[misc] + """Response ssl error.""" + + +class ClientConnectorCertificateError(*cert_errors_bases): # type: ignore[misc] + """Response certificate error.""" + + def __init__( + self, connection_key: ConnectionKey, certificate_error: Exception + ) -> None: + self._conn_key = connection_key + self._certificate_error = certificate_error + self.args = (connection_key, certificate_error) + + @property + def certificate_error(self) -> Exception: + return self._certificate_error + + @property + def host(self) -> str: + return self._conn_key.host + + @property + def port(self) -> Optional[int]: + return self._conn_key.port + + @property + def ssl(self) -> bool: + return self._conn_key.is_ssl + + def __str__(self) -> str: + return ( + "Cannot connect to host {0.host}:{0.port} ssl:{0.ssl} " + "[{0.certificate_error.__class__.__name__}: " + "{0.certificate_error.args}]".format(self) + ) diff --git a/contrib/python/aiohttp/aiohttp/client_proto.py b/contrib/python/aiohttp/aiohttp/client_proto.py new file mode 100644 index 0000000000..f36863b836 --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/client_proto.py @@ -0,0 +1,251 @@ +import asyncio +from contextlib import suppress +from typing import Any, Optional, Tuple + +from .base_protocol import BaseProtocol +from .client_exceptions import ( + ClientOSError, + ClientPayloadError, + ServerDisconnectedError, + ServerTimeoutError, +) +from .helpers import BaseTimerContext +from .http import HttpResponseParser, RawResponseMessage +from .streams import EMPTY_PAYLOAD, DataQueue, StreamReader + + +class ResponseHandler(BaseProtocol, DataQueue[Tuple[RawResponseMessage, StreamReader]]): + """Helper class to adapt between Protocol and StreamReader.""" + + def __init__(self, loop: asyncio.AbstractEventLoop) -> None: + BaseProtocol.__init__(self, loop=loop) + DataQueue.__init__(self, loop) + + self._should_close = False + + self._payload: Optional[StreamReader] = None + self._skip_payload = False + self._payload_parser = None + + self._timer = None + + self._tail = b"" + self._upgraded = False + self._parser = None # type: Optional[HttpResponseParser] + + self._read_timeout = None # type: Optional[float] + self._read_timeout_handle = None # type: Optional[asyncio.TimerHandle] + + @property + def upgraded(self) -> bool: + return self._upgraded + + @property + def should_close(self) -> bool: + if self._payload is not None and not self._payload.is_eof() or self._upgraded: + return True + + return ( + self._should_close + or self._upgraded + or self.exception() is not None + or self._payload_parser is not None + or len(self) > 0 + or bool(self._tail) + ) + + def force_close(self) -> None: + self._should_close = True + + def close(self) -> None: + transport = self.transport + if transport is not None: + transport.close() + self.transport = None + self._payload = None + self._drop_timeout() + + def is_connected(self) -> bool: + return self.transport is not None and not self.transport.is_closing() + + def connection_lost(self, exc: Optional[BaseException]) -> None: + self._drop_timeout() + + if self._payload_parser is not None: + with suppress(Exception): + self._payload_parser.feed_eof() + + uncompleted = None + if self._parser is not None: + try: + uncompleted = self._parser.feed_eof() + except Exception: + if self._payload is not None: + self._payload.set_exception( + ClientPayloadError("Response payload is not completed") + ) + + if not self.is_eof(): + if isinstance(exc, OSError): + exc = ClientOSError(*exc.args) + if exc is None: + exc = ServerDisconnectedError(uncompleted) + # assigns self._should_close to True as side effect, + # we do it anyway below + self.set_exception(exc) + + self._should_close = True + self._parser = None + self._payload = None + self._payload_parser = None + self._reading_paused = False + + super().connection_lost(exc) + + def eof_received(self) -> None: + # should call parser.feed_eof() most likely + self._drop_timeout() + + def pause_reading(self) -> None: + super().pause_reading() + self._drop_timeout() + + def resume_reading(self) -> None: + super().resume_reading() + self._reschedule_timeout() + + def set_exception(self, exc: BaseException) -> None: + self._should_close = True + self._drop_timeout() + super().set_exception(exc) + + def set_parser(self, parser: Any, payload: Any) -> None: + # TODO: actual types are: + # parser: WebSocketReader + # payload: FlowControlDataQueue + # but they are not generi enough + # Need an ABC for both types + self._payload = payload + self._payload_parser = parser + + self._drop_timeout() + + if self._tail: + data, self._tail = self._tail, b"" + self.data_received(data) + + def set_response_params( + self, + *, + timer: Optional[BaseTimerContext] = None, + skip_payload: bool = False, + read_until_eof: bool = False, + auto_decompress: bool = True, + read_timeout: Optional[float] = None, + read_bufsize: int = 2 ** 16, + ) -> None: + self._skip_payload = skip_payload + + self._read_timeout = read_timeout + self._reschedule_timeout() + + self._parser = HttpResponseParser( + self, + self._loop, + read_bufsize, + timer=timer, + payload_exception=ClientPayloadError, + response_with_body=not skip_payload, + read_until_eof=read_until_eof, + auto_decompress=auto_decompress, + ) + + if self._tail: + data, self._tail = self._tail, b"" + self.data_received(data) + + def _drop_timeout(self) -> None: + if self._read_timeout_handle is not None: + self._read_timeout_handle.cancel() + self._read_timeout_handle = None + + def _reschedule_timeout(self) -> None: + timeout = self._read_timeout + if self._read_timeout_handle is not None: + self._read_timeout_handle.cancel() + + if timeout: + self._read_timeout_handle = self._loop.call_later( + timeout, self._on_read_timeout + ) + else: + self._read_timeout_handle = None + + def _on_read_timeout(self) -> None: + exc = ServerTimeoutError("Timeout on reading data from socket") + self.set_exception(exc) + if self._payload is not None: + self._payload.set_exception(exc) + + def data_received(self, data: bytes) -> None: + self._reschedule_timeout() + + if not data: + return + + # custom payload parser + if self._payload_parser is not None: + eof, tail = self._payload_parser.feed_data(data) + if eof: + self._payload = None + self._payload_parser = None + + if tail: + self.data_received(tail) + return + else: + if self._upgraded or self._parser is None: + # i.e. websocket connection, websocket parser is not set yet + self._tail += data + else: + # parse http messages + try: + messages, upgraded, tail = self._parser.feed_data(data) + except BaseException as exc: + if self.transport is not None: + # connection.release() could be called BEFORE + # data_received(), the transport is already + # closed in this case + self.transport.close() + # should_close is True after the call + self.set_exception(exc) + return + + self._upgraded = upgraded + + payload: Optional[StreamReader] = None + for message, payload in messages: + if message.should_close: + self._should_close = True + + self._payload = payload + + if self._skip_payload or message.code in (204, 304): + self.feed_data((message, EMPTY_PAYLOAD), 0) + else: + self.feed_data((message, payload), 0) + if payload is not None: + # new message(s) was processed + # register timeout handler unsubscribing + # either on end-of-stream or immediately for + # EMPTY_PAYLOAD + if payload is not EMPTY_PAYLOAD: + payload.on_eof(self._drop_timeout) + else: + self._drop_timeout() + + if tail: + if upgraded: + self.data_received(tail) + else: + self._tail = tail diff --git a/contrib/python/aiohttp/aiohttp/client_reqrep.py b/contrib/python/aiohttp/aiohttp/client_reqrep.py new file mode 100644 index 0000000000..343002517b --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/client_reqrep.py @@ -0,0 +1,1133 @@ +import asyncio +import codecs +import functools +import io +import re +import sys +import traceback +import warnings +from hashlib import md5, sha1, sha256 +from http.cookies import CookieError, Morsel, SimpleCookie +from types import MappingProxyType, TracebackType +from typing import ( + TYPE_CHECKING, + Any, + Dict, + Iterable, + List, + Mapping, + Optional, + Tuple, + Type, + Union, + cast, +) + +import attr +from multidict import CIMultiDict, CIMultiDictProxy, MultiDict, MultiDictProxy +from yarl import URL + +from . import hdrs, helpers, http, multipart, payload +from .abc import AbstractStreamWriter +from .client_exceptions import ( + ClientConnectionError, + ClientOSError, + ClientResponseError, + ContentTypeError, + InvalidURL, + ServerFingerprintMismatch, +) +from .formdata import FormData +from .helpers import ( + PY_36, + BaseTimerContext, + BasicAuth, + HeadersMixin, + TimerNoop, + noop, + reify, + set_result, +) +from .http import SERVER_SOFTWARE, HttpVersion10, HttpVersion11, StreamWriter +from .log import client_logger +from .streams import StreamReader +from .typedefs import ( + DEFAULT_JSON_DECODER, + JSONDecoder, + LooseCookies, + LooseHeaders, + RawHeaders, +) + +try: + import ssl + from ssl import SSLContext +except ImportError: # pragma: no cover + ssl = None # type: ignore[assignment] + SSLContext = object # type: ignore[misc,assignment] + +try: + import cchardet as chardet +except ImportError: # pragma: no cover + import charset_normalizer as chardet # type: ignore[no-redef] + + +__all__ = ("ClientRequest", "ClientResponse", "RequestInfo", "Fingerprint") + + +if TYPE_CHECKING: # pragma: no cover + from .client import ClientSession + from .connector import Connection + from .tracing import Trace + + +json_re = re.compile(r"^application/(?:[\w.+-]+?\+)?json") + + +@attr.s(auto_attribs=True, frozen=True, slots=True) +class ContentDisposition: + type: Optional[str] + parameters: "MappingProxyType[str, str]" + filename: Optional[str] + + +@attr.s(auto_attribs=True, frozen=True, slots=True) +class RequestInfo: + url: URL + method: str + headers: "CIMultiDictProxy[str]" + real_url: URL = attr.ib() + + @real_url.default + def real_url_default(self) -> URL: + return self.url + + +class Fingerprint: + HASHFUNC_BY_DIGESTLEN = { + 16: md5, + 20: sha1, + 32: sha256, + } + + def __init__(self, fingerprint: bytes) -> None: + digestlen = len(fingerprint) + hashfunc = self.HASHFUNC_BY_DIGESTLEN.get(digestlen) + if not hashfunc: + raise ValueError("fingerprint has invalid length") + elif hashfunc is md5 or hashfunc is sha1: + raise ValueError( + "md5 and sha1 are insecure and " "not supported. Use sha256." + ) + self._hashfunc = hashfunc + self._fingerprint = fingerprint + + @property + def fingerprint(self) -> bytes: + return self._fingerprint + + def check(self, transport: asyncio.Transport) -> None: + if not transport.get_extra_info("sslcontext"): + return + sslobj = transport.get_extra_info("ssl_object") + cert = sslobj.getpeercert(binary_form=True) + got = self._hashfunc(cert).digest() + if got != self._fingerprint: + host, port, *_ = transport.get_extra_info("peername") + raise ServerFingerprintMismatch(self._fingerprint, got, host, port) + + +if ssl is not None: + SSL_ALLOWED_TYPES = (ssl.SSLContext, bool, Fingerprint, type(None)) +else: # pragma: no cover + SSL_ALLOWED_TYPES = type(None) + + +def _merge_ssl_params( + ssl: Union["SSLContext", bool, Fingerprint, None], + verify_ssl: Optional[bool], + ssl_context: Optional["SSLContext"], + fingerprint: Optional[bytes], +) -> Union["SSLContext", bool, Fingerprint, None]: + if verify_ssl is not None and not verify_ssl: + warnings.warn( + "verify_ssl is deprecated, use ssl=False instead", + DeprecationWarning, + stacklevel=3, + ) + if ssl is not None: + raise ValueError( + "verify_ssl, ssl_context, fingerprint and ssl " + "parameters are mutually exclusive" + ) + else: + ssl = False + if ssl_context is not None: + warnings.warn( + "ssl_context is deprecated, use ssl=context instead", + DeprecationWarning, + stacklevel=3, + ) + if ssl is not None: + raise ValueError( + "verify_ssl, ssl_context, fingerprint and ssl " + "parameters are mutually exclusive" + ) + else: + ssl = ssl_context + if fingerprint is not None: + warnings.warn( + "fingerprint is deprecated, " "use ssl=Fingerprint(fingerprint) instead", + DeprecationWarning, + stacklevel=3, + ) + if ssl is not None: + raise ValueError( + "verify_ssl, ssl_context, fingerprint and ssl " + "parameters are mutually exclusive" + ) + else: + ssl = Fingerprint(fingerprint) + if not isinstance(ssl, SSL_ALLOWED_TYPES): + raise TypeError( + "ssl should be SSLContext, bool, Fingerprint or None, " + "got {!r} instead.".format(ssl) + ) + return ssl + + +@attr.s(auto_attribs=True, slots=True, frozen=True) +class ConnectionKey: + # the key should contain an information about used proxy / TLS + # to prevent reusing wrong connections from a pool + host: str + port: Optional[int] + is_ssl: bool + ssl: Union[SSLContext, None, bool, Fingerprint] + proxy: Optional[URL] + proxy_auth: Optional[BasicAuth] + proxy_headers_hash: Optional[int] # hash(CIMultiDict) + + +def _is_expected_content_type( + response_content_type: str, expected_content_type: str +) -> bool: + if expected_content_type == "application/json": + return json_re.match(response_content_type) is not None + return expected_content_type in response_content_type + + +class ClientRequest: + GET_METHODS = { + hdrs.METH_GET, + hdrs.METH_HEAD, + hdrs.METH_OPTIONS, + hdrs.METH_TRACE, + } + POST_METHODS = {hdrs.METH_PATCH, hdrs.METH_POST, hdrs.METH_PUT} + ALL_METHODS = GET_METHODS.union(POST_METHODS).union({hdrs.METH_DELETE}) + + DEFAULT_HEADERS = { + hdrs.ACCEPT: "*/*", + hdrs.ACCEPT_ENCODING: "gzip, deflate", + } + + body = b"" + auth = None + response = None + + _writer = None # async task for streaming data + _continue = None # waiter future for '100 Continue' response + + # N.B. + # Adding __del__ method with self._writer closing doesn't make sense + # because _writer is instance method, thus it keeps a reference to self. + # Until writer has finished finalizer will not be called. + + def __init__( + self, + method: str, + url: URL, + *, + params: Optional[Mapping[str, str]] = None, + headers: Optional[LooseHeaders] = None, + skip_auto_headers: Iterable[str] = frozenset(), + data: Any = None, + cookies: Optional[LooseCookies] = None, + auth: Optional[BasicAuth] = None, + version: http.HttpVersion = http.HttpVersion11, + compress: Optional[str] = None, + chunked: Optional[bool] = None, + expect100: bool = False, + loop: Optional[asyncio.AbstractEventLoop] = None, + response_class: Optional[Type["ClientResponse"]] = None, + proxy: Optional[URL] = None, + proxy_auth: Optional[BasicAuth] = None, + timer: Optional[BaseTimerContext] = None, + session: Optional["ClientSession"] = None, + ssl: Union[SSLContext, bool, Fingerprint, None] = None, + proxy_headers: Optional[LooseHeaders] = None, + traces: Optional[List["Trace"]] = None, + ): + + if loop is None: + loop = asyncio.get_event_loop() + + assert isinstance(url, URL), url + assert isinstance(proxy, (URL, type(None))), proxy + # FIXME: session is None in tests only, need to fix tests + # assert session is not None + self._session = cast("ClientSession", session) + if params: + q = MultiDict(url.query) + url2 = url.with_query(params) + q.extend(url2.query) + url = url.with_query(q) + self.original_url = url + self.url = url.with_fragment(None) + self.method = method.upper() + self.chunked = chunked + self.compress = compress + self.loop = loop + self.length = None + if response_class is None: + real_response_class = ClientResponse + else: + real_response_class = response_class + self.response_class = real_response_class # type: Type[ClientResponse] + self._timer = timer if timer is not None else TimerNoop() + self._ssl = ssl + + if loop.get_debug(): + self._source_traceback = traceback.extract_stack(sys._getframe(1)) + + self.update_version(version) + self.update_host(url) + self.update_headers(headers) + self.update_auto_headers(skip_auto_headers) + self.update_cookies(cookies) + self.update_content_encoding(data) + self.update_auth(auth) + self.update_proxy(proxy, proxy_auth, proxy_headers) + + self.update_body_from_data(data) + if data is not None or self.method not in self.GET_METHODS: + self.update_transfer_encoding() + self.update_expect_continue(expect100) + if traces is None: + traces = [] + self._traces = traces + + def is_ssl(self) -> bool: + return self.url.scheme in ("https", "wss") + + @property + def ssl(self) -> Union["SSLContext", None, bool, Fingerprint]: + return self._ssl + + @property + def connection_key(self) -> ConnectionKey: + proxy_headers = self.proxy_headers + if proxy_headers: + h = hash( + tuple((k, v) for k, v in proxy_headers.items()) + ) # type: Optional[int] + else: + h = None + return ConnectionKey( + self.host, + self.port, + self.is_ssl(), + self.ssl, + self.proxy, + self.proxy_auth, + h, + ) + + @property + def host(self) -> str: + ret = self.url.raw_host + assert ret is not None + return ret + + @property + def port(self) -> Optional[int]: + return self.url.port + + @property + def request_info(self) -> RequestInfo: + headers = CIMultiDictProxy(self.headers) # type: CIMultiDictProxy[str] + return RequestInfo(self.url, self.method, headers, self.original_url) + + def update_host(self, url: URL) -> None: + """Update destination host, port and connection type (ssl).""" + # get host/port + if not url.raw_host: + raise InvalidURL(url) + + # basic auth info + username, password = url.user, url.password + if username: + self.auth = helpers.BasicAuth(username, password or "") + + def update_version(self, version: Union[http.HttpVersion, str]) -> None: + """Convert request version to two elements tuple. + + parser HTTP version '1.1' => (1, 1) + """ + if isinstance(version, str): + v = [part.strip() for part in version.split(".", 1)] + try: + version = http.HttpVersion(int(v[0]), int(v[1])) + except ValueError: + raise ValueError( + f"Can not parse http version number: {version}" + ) from None + self.version = version + + def update_headers(self, headers: Optional[LooseHeaders]) -> None: + """Update request headers.""" + self.headers = CIMultiDict() # type: CIMultiDict[str] + + # add host + netloc = cast(str, self.url.raw_host) + if helpers.is_ipv6_address(netloc): + netloc = f"[{netloc}]" + if self.url.port is not None and not self.url.is_default_port(): + netloc += ":" + str(self.url.port) + self.headers[hdrs.HOST] = netloc + + if headers: + if isinstance(headers, (dict, MultiDictProxy, MultiDict)): + headers = headers.items() # type: ignore[assignment] + + for key, value in headers: # type: ignore[misc] + # A special case for Host header + if key.lower() == "host": + self.headers[key] = value + else: + self.headers.add(key, value) + + def update_auto_headers(self, skip_auto_headers: Iterable[str]) -> None: + self.skip_auto_headers = CIMultiDict( + (hdr, None) for hdr in sorted(skip_auto_headers) + ) + used_headers = self.headers.copy() + used_headers.extend(self.skip_auto_headers) # type: ignore[arg-type] + + for hdr, val in self.DEFAULT_HEADERS.items(): + if hdr not in used_headers: + self.headers.add(hdr, val) + + if hdrs.USER_AGENT not in used_headers: + self.headers[hdrs.USER_AGENT] = SERVER_SOFTWARE + + def update_cookies(self, cookies: Optional[LooseCookies]) -> None: + """Update request cookies header.""" + if not cookies: + return + + c = SimpleCookie() # type: SimpleCookie[str] + if hdrs.COOKIE in self.headers: + c.load(self.headers.get(hdrs.COOKIE, "")) + del self.headers[hdrs.COOKIE] + + if isinstance(cookies, Mapping): + iter_cookies = cookies.items() + else: + iter_cookies = cookies # type: ignore[assignment] + for name, value in iter_cookies: + if isinstance(value, Morsel): + # Preserve coded_value + mrsl_val = value.get(value.key, Morsel()) + mrsl_val.set(value.key, value.value, value.coded_value) + c[name] = mrsl_val + else: + c[name] = value # type: ignore[assignment] + + self.headers[hdrs.COOKIE] = c.output(header="", sep=";").strip() + + def update_content_encoding(self, data: Any) -> None: + """Set request content encoding.""" + if data is None: + return + + enc = self.headers.get(hdrs.CONTENT_ENCODING, "").lower() + if enc: + if self.compress: + raise ValueError( + "compress can not be set " "if Content-Encoding header is set" + ) + elif self.compress: + if not isinstance(self.compress, str): + self.compress = "deflate" + self.headers[hdrs.CONTENT_ENCODING] = self.compress + self.chunked = True # enable chunked, no need to deal with length + + def update_transfer_encoding(self) -> None: + """Analyze transfer-encoding header.""" + te = self.headers.get(hdrs.TRANSFER_ENCODING, "").lower() + + if "chunked" in te: + if self.chunked: + raise ValueError( + "chunked can not be set " + 'if "Transfer-Encoding: chunked" header is set' + ) + + elif self.chunked: + if hdrs.CONTENT_LENGTH in self.headers: + raise ValueError( + "chunked can not be set " "if Content-Length header is set" + ) + + self.headers[hdrs.TRANSFER_ENCODING] = "chunked" + else: + if hdrs.CONTENT_LENGTH not in self.headers: + self.headers[hdrs.CONTENT_LENGTH] = str(len(self.body)) + + def update_auth(self, auth: Optional[BasicAuth]) -> None: + """Set basic auth.""" + if auth is None: + auth = self.auth + if auth is None: + return + + if not isinstance(auth, helpers.BasicAuth): + raise TypeError("BasicAuth() tuple is required instead") + + self.headers[hdrs.AUTHORIZATION] = auth.encode() + + def update_body_from_data(self, body: Any) -> None: + if body is None: + return + + # FormData + if isinstance(body, FormData): + body = body() + + try: + body = payload.PAYLOAD_REGISTRY.get(body, disposition=None) + except payload.LookupError: + body = FormData(body)() + + self.body = body + + # enable chunked encoding if needed + if not self.chunked: + if hdrs.CONTENT_LENGTH not in self.headers: + size = body.size + if size is None: + self.chunked = True + else: + if hdrs.CONTENT_LENGTH not in self.headers: + self.headers[hdrs.CONTENT_LENGTH] = str(size) + + # copy payload headers + assert body.headers + for (key, value) in body.headers.items(): + if key in self.headers: + continue + if key in self.skip_auto_headers: + continue + self.headers[key] = value + + def update_expect_continue(self, expect: bool = False) -> None: + if expect: + self.headers[hdrs.EXPECT] = "100-continue" + elif self.headers.get(hdrs.EXPECT, "").lower() == "100-continue": + expect = True + + if expect: + self._continue = self.loop.create_future() + + def update_proxy( + self, + proxy: Optional[URL], + proxy_auth: Optional[BasicAuth], + proxy_headers: Optional[LooseHeaders], + ) -> None: + if proxy_auth and not isinstance(proxy_auth, helpers.BasicAuth): + raise ValueError("proxy_auth must be None or BasicAuth() tuple") + self.proxy = proxy + self.proxy_auth = proxy_auth + self.proxy_headers = proxy_headers + + def keep_alive(self) -> bool: + if self.version < HttpVersion10: + # keep alive not supported at all + return False + if self.version == HttpVersion10: + if self.headers.get(hdrs.CONNECTION) == "keep-alive": + return True + else: # no headers means we close for Http 1.0 + return False + elif self.headers.get(hdrs.CONNECTION) == "close": + return False + + return True + + async def write_bytes( + self, writer: AbstractStreamWriter, conn: "Connection" + ) -> None: + """Support coroutines that yields bytes objects.""" + # 100 response + if self._continue is not None: + await writer.drain() + await self._continue + + protocol = conn.protocol + assert protocol is not None + try: + if isinstance(self.body, payload.Payload): + await self.body.write(writer) + else: + if isinstance(self.body, (bytes, bytearray)): + self.body = (self.body,) # type: ignore[assignment] + + for chunk in self.body: + await writer.write(chunk) # type: ignore[arg-type] + + await writer.write_eof() + except OSError as exc: + new_exc = ClientOSError( + exc.errno, "Can not write request body for %s" % self.url + ) + new_exc.__context__ = exc + new_exc.__cause__ = exc + protocol.set_exception(new_exc) + except asyncio.CancelledError as exc: + if not conn.closed: + protocol.set_exception(exc) + except Exception as exc: + protocol.set_exception(exc) + finally: + self._writer = None + + async def send(self, conn: "Connection") -> "ClientResponse": + # Specify request target: + # - CONNECT request must send authority form URI + # - not CONNECT proxy must send absolute form URI + # - most common is origin form URI + if self.method == hdrs.METH_CONNECT: + connect_host = self.url.raw_host + assert connect_host is not None + if helpers.is_ipv6_address(connect_host): + connect_host = f"[{connect_host}]" + path = f"{connect_host}:{self.url.port}" + elif self.proxy and not self.is_ssl(): + path = str(self.url) + else: + path = self.url.raw_path + if self.url.raw_query_string: + path += "?" + self.url.raw_query_string + + protocol = conn.protocol + assert protocol is not None + writer = StreamWriter( + protocol, + self.loop, + on_chunk_sent=functools.partial( + self._on_chunk_request_sent, self.method, self.url + ), + on_headers_sent=functools.partial( + self._on_headers_request_sent, self.method, self.url + ), + ) + + if self.compress: + writer.enable_compression(self.compress) + + if self.chunked is not None: + writer.enable_chunking() + + # set default content-type + if ( + self.method in self.POST_METHODS + and hdrs.CONTENT_TYPE not in self.skip_auto_headers + and hdrs.CONTENT_TYPE not in self.headers + ): + self.headers[hdrs.CONTENT_TYPE] = "application/octet-stream" + + # set the connection header + connection = self.headers.get(hdrs.CONNECTION) + if not connection: + if self.keep_alive(): + if self.version == HttpVersion10: + connection = "keep-alive" + else: + if self.version == HttpVersion11: + connection = "close" + + if connection is not None: + self.headers[hdrs.CONNECTION] = connection + + # status + headers + status_line = "{0} {1} HTTP/{2[0]}.{2[1]}".format( + self.method, path, self.version + ) + await writer.write_headers(status_line, self.headers) + + self._writer = self.loop.create_task(self.write_bytes(writer, conn)) + + response_class = self.response_class + assert response_class is not None + self.response = response_class( + self.method, + self.original_url, + writer=self._writer, + continue100=self._continue, + timer=self._timer, + request_info=self.request_info, + traces=self._traces, + loop=self.loop, + session=self._session, + ) + return self.response + + async def close(self) -> None: + if self._writer is not None: + try: + await self._writer + finally: + self._writer = None + + def terminate(self) -> None: + if self._writer is not None: + if not self.loop.is_closed(): + self._writer.cancel() + self._writer = None + + async def _on_chunk_request_sent(self, method: str, url: URL, chunk: bytes) -> None: + for trace in self._traces: + await trace.send_request_chunk_sent(method, url, chunk) + + async def _on_headers_request_sent( + self, method: str, url: URL, headers: "CIMultiDict[str]" + ) -> None: + for trace in self._traces: + await trace.send_request_headers(method, url, headers) + + +class ClientResponse(HeadersMixin): + + # from the Status-Line of the response + version = None # HTTP-Version + status = None # type: int # Status-Code + reason = None # Reason-Phrase + + content = None # type: StreamReader # Payload stream + _headers = None # type: CIMultiDictProxy[str] # Response headers + _raw_headers = None # type: RawHeaders # Response raw headers + + _connection = None # current connection + _source_traceback = None + # setted up by ClientRequest after ClientResponse object creation + # post-init stage allows to not change ctor signature + _closed = True # to allow __del__ for non-initialized properly response + _released = False + + def __init__( + self, + method: str, + url: URL, + *, + writer: "asyncio.Task[None]", + continue100: Optional["asyncio.Future[bool]"], + timer: BaseTimerContext, + request_info: RequestInfo, + traces: List["Trace"], + loop: asyncio.AbstractEventLoop, + session: "ClientSession", + ) -> None: + assert isinstance(url, URL) + + self.method = method + self.cookies = SimpleCookie() # type: SimpleCookie[str] + + self._real_url = url + self._url = url.with_fragment(None) + self._body = None # type: Any + self._writer = writer # type: Optional[asyncio.Task[None]] + self._continue = continue100 # None by default + self._closed = True + self._history = () # type: Tuple[ClientResponse, ...] + self._request_info = request_info + self._timer = timer if timer is not None else TimerNoop() + self._cache = {} # type: Dict[str, Any] + self._traces = traces + self._loop = loop + # store a reference to session #1985 + self._session = session # type: Optional[ClientSession] + if loop.get_debug(): + self._source_traceback = traceback.extract_stack(sys._getframe(1)) + + @reify + def url(self) -> URL: + return self._url + + @reify + def url_obj(self) -> URL: + warnings.warn("Deprecated, use .url #1654", DeprecationWarning, stacklevel=2) + return self._url + + @reify + def real_url(self) -> URL: + return self._real_url + + @reify + def host(self) -> str: + assert self._url.host is not None + return self._url.host + + @reify + def headers(self) -> "CIMultiDictProxy[str]": + return self._headers + + @reify + def raw_headers(self) -> RawHeaders: + return self._raw_headers + + @reify + def request_info(self) -> RequestInfo: + return self._request_info + + @reify + def content_disposition(self) -> Optional[ContentDisposition]: + raw = self._headers.get(hdrs.CONTENT_DISPOSITION) + if raw is None: + return None + disposition_type, params_dct = multipart.parse_content_disposition(raw) + params = MappingProxyType(params_dct) + filename = multipart.content_disposition_filename(params) + return ContentDisposition(disposition_type, params, filename) + + def __del__(self, _warnings: Any = warnings) -> None: + if self._closed: + return + + if self._connection is not None: + self._connection.release() + self._cleanup_writer() + + if self._loop.get_debug(): + if PY_36: + kwargs = {"source": self} + else: + kwargs = {} + _warnings.warn(f"Unclosed response {self!r}", ResourceWarning, **kwargs) + context = {"client_response": self, "message": "Unclosed response"} + if self._source_traceback: + context["source_traceback"] = self._source_traceback + self._loop.call_exception_handler(context) + + def __repr__(self) -> str: + out = io.StringIO() + ascii_encodable_url = str(self.url) + if self.reason: + ascii_encodable_reason = self.reason.encode( + "ascii", "backslashreplace" + ).decode("ascii") + else: + ascii_encodable_reason = self.reason + print( + "<ClientResponse({}) [{} {}]>".format( + ascii_encodable_url, self.status, ascii_encodable_reason + ), + file=out, + ) + print(self.headers, file=out) + return out.getvalue() + + @property + def connection(self) -> Optional["Connection"]: + return self._connection + + @reify + def history(self) -> Tuple["ClientResponse", ...]: + """A sequence of of responses, if redirects occurred.""" + return self._history + + @reify + def links(self) -> "MultiDictProxy[MultiDictProxy[Union[str, URL]]]": + links_str = ", ".join(self.headers.getall("link", [])) + + if not links_str: + return MultiDictProxy(MultiDict()) + + links = MultiDict() # type: MultiDict[MultiDictProxy[Union[str, URL]]] + + for val in re.split(r",(?=\s*<)", links_str): + match = re.match(r"\s*<(.*)>(.*)", val) + if match is None: # pragma: no cover + # the check exists to suppress mypy error + continue + url, params_str = match.groups() + params = params_str.split(";")[1:] + + link = MultiDict() # type: MultiDict[Union[str, URL]] + + for param in params: + match = re.match(r"^\s*(\S*)\s*=\s*(['\"]?)(.*?)(\2)\s*$", param, re.M) + if match is None: # pragma: no cover + # the check exists to suppress mypy error + continue + key, _, value, _ = match.groups() + + link.add(key, value) + + key = link.get("rel", url) # type: ignore[assignment] + + link.add("url", self.url.join(URL(url))) + + links.add(key, MultiDictProxy(link)) + + return MultiDictProxy(links) + + async def start(self, connection: "Connection") -> "ClientResponse": + """Start response processing.""" + self._closed = False + self._protocol = connection.protocol + self._connection = connection + + with self._timer: + while True: + # read response + try: + protocol = self._protocol + message, payload = await protocol.read() # type: ignore[union-attr] + except http.HttpProcessingError as exc: + raise ClientResponseError( + self.request_info, + self.history, + status=exc.code, + message=exc.message, + headers=exc.headers, + ) from exc + + if message.code < 100 or message.code > 199 or message.code == 101: + break + + if self._continue is not None: + set_result(self._continue, True) + self._continue = None + + # payload eof handler + payload.on_eof(self._response_eof) + + # response status + self.version = message.version + self.status = message.code + self.reason = message.reason + + # headers + self._headers = message.headers # type is CIMultiDictProxy + self._raw_headers = message.raw_headers # type is Tuple[bytes, bytes] + + # payload + self.content = payload + + # cookies + for hdr in self.headers.getall(hdrs.SET_COOKIE, ()): + try: + self.cookies.load(hdr) + except CookieError as exc: + client_logger.warning("Can not load response cookies: %s", exc) + return self + + def _response_eof(self) -> None: + if self._closed: + return + + if self._connection is not None: + # websocket, protocol could be None because + # connection could be detached + if ( + self._connection.protocol is not None + and self._connection.protocol.upgraded + ): + return + + self._connection.release() + self._connection = None + + self._closed = True + self._cleanup_writer() + + @property + def closed(self) -> bool: + return self._closed + + def close(self) -> None: + if not self._released: + self._notify_content() + if self._closed: + return + + self._closed = True + if self._loop is None or self._loop.is_closed(): + return + + if self._connection is not None: + self._connection.close() + self._connection = None + self._cleanup_writer() + + def release(self) -> Any: + if not self._released: + self._notify_content() + if self._closed: + return noop() + + self._closed = True + if self._connection is not None: + self._connection.release() + self._connection = None + + self._cleanup_writer() + return noop() + + @property + def ok(self) -> bool: + """Returns ``True`` if ``status`` is less than ``400``, ``False`` if not. + + This is **not** a check for ``200 OK`` but a check that the response + status is under 400. + """ + return 400 > self.status + + def raise_for_status(self) -> None: + if not self.ok: + # reason should always be not None for a started response + assert self.reason is not None + self.release() + raise ClientResponseError( + self.request_info, + self.history, + status=self.status, + message=self.reason, + headers=self.headers, + ) + + def _cleanup_writer(self) -> None: + if self._writer is not None: + self._writer.cancel() + self._writer = None + self._session = None + + def _notify_content(self) -> None: + content = self.content + if content and content.exception() is None: + content.set_exception(ClientConnectionError("Connection closed")) + self._released = True + + async def wait_for_close(self) -> None: + if self._writer is not None: + try: + await self._writer + finally: + self._writer = None + self.release() + + async def read(self) -> bytes: + """Read response payload.""" + if self._body is None: + try: + self._body = await self.content.read() + for trace in self._traces: + await trace.send_response_chunk_received( + self.method, self.url, self._body + ) + except BaseException: + self.close() + raise + elif self._released: + raise ClientConnectionError("Connection closed") + + return self._body # type: ignore[no-any-return] + + def get_encoding(self) -> str: + ctype = self.headers.get(hdrs.CONTENT_TYPE, "").lower() + mimetype = helpers.parse_mimetype(ctype) + + encoding = mimetype.parameters.get("charset") + if encoding: + try: + codecs.lookup(encoding) + except LookupError: + encoding = None + if not encoding: + if mimetype.type == "application" and ( + mimetype.subtype == "json" or mimetype.subtype == "rdap" + ): + # RFC 7159 states that the default encoding is UTF-8. + # RFC 7483 defines application/rdap+json + encoding = "utf-8" + elif self._body is None: + raise RuntimeError( + "Cannot guess the encoding of " "a not yet read body" + ) + else: + encoding = chardet.detect(self._body)["encoding"] + if not encoding: + encoding = "utf-8" + + return encoding + + async def text(self, encoding: Optional[str] = None, errors: str = "strict") -> str: + """Read response payload and decode.""" + if self._body is None: + await self.read() + + if encoding is None: + encoding = self.get_encoding() + + return self._body.decode( # type: ignore[no-any-return,union-attr] + encoding, errors=errors + ) + + async def json( + self, + *, + encoding: Optional[str] = None, + loads: JSONDecoder = DEFAULT_JSON_DECODER, + content_type: Optional[str] = "application/json", + ) -> Any: + """Read and decodes JSON response.""" + if self._body is None: + await self.read() + + if content_type: + ctype = self.headers.get(hdrs.CONTENT_TYPE, "").lower() + if not _is_expected_content_type(ctype, content_type): + raise ContentTypeError( + self.request_info, + self.history, + message=( + "Attempt to decode JSON with " "unexpected mimetype: %s" % ctype + ), + headers=self.headers, + ) + + stripped = self._body.strip() # type: ignore[union-attr] + if not stripped: + return None + + if encoding is None: + encoding = self.get_encoding() + + return loads(stripped.decode(encoding)) + + async def __aenter__(self) -> "ClientResponse": + return self + + async def __aexit__( + self, + exc_type: Optional[Type[BaseException]], + exc_val: Optional[BaseException], + exc_tb: Optional[TracebackType], + ) -> None: + # similar to _RequestContextManager, we do not need to check + # for exceptions, response object can close connection + # if state is broken + self.release() diff --git a/contrib/python/aiohttp/aiohttp/client_ws.py b/contrib/python/aiohttp/aiohttp/client_ws.py new file mode 100644 index 0000000000..7c8121f659 --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/client_ws.py @@ -0,0 +1,300 @@ +"""WebSocket client for asyncio.""" + +import asyncio +from typing import Any, Optional, cast + +import async_timeout + +from .client_exceptions import ClientError +from .client_reqrep import ClientResponse +from .helpers import call_later, set_result +from .http import ( + WS_CLOSED_MESSAGE, + WS_CLOSING_MESSAGE, + WebSocketError, + WSCloseCode, + WSMessage, + WSMsgType, +) +from .http_websocket import WebSocketWriter # WSMessage +from .streams import EofStream, FlowControlDataQueue +from .typedefs import ( + DEFAULT_JSON_DECODER, + DEFAULT_JSON_ENCODER, + JSONDecoder, + JSONEncoder, +) + + +class ClientWebSocketResponse: + def __init__( + self, + reader: "FlowControlDataQueue[WSMessage]", + writer: WebSocketWriter, + protocol: Optional[str], + response: ClientResponse, + timeout: float, + autoclose: bool, + autoping: bool, + loop: asyncio.AbstractEventLoop, + *, + receive_timeout: Optional[float] = None, + heartbeat: Optional[float] = None, + compress: int = 0, + client_notakeover: bool = False, + ) -> None: + self._response = response + self._conn = response.connection + + self._writer = writer + self._reader = reader + self._protocol = protocol + self._closed = False + self._closing = False + self._close_code = None # type: Optional[int] + self._timeout = timeout + self._receive_timeout = receive_timeout + self._autoclose = autoclose + self._autoping = autoping + self._heartbeat = heartbeat + self._heartbeat_cb: Optional[asyncio.TimerHandle] = None + if heartbeat is not None: + self._pong_heartbeat = heartbeat / 2.0 + self._pong_response_cb: Optional[asyncio.TimerHandle] = None + self._loop = loop + self._waiting = None # type: Optional[asyncio.Future[bool]] + self._exception = None # type: Optional[BaseException] + self._compress = compress + self._client_notakeover = client_notakeover + + self._reset_heartbeat() + + def _cancel_heartbeat(self) -> None: + if self._pong_response_cb is not None: + self._pong_response_cb.cancel() + self._pong_response_cb = None + + if self._heartbeat_cb is not None: + self._heartbeat_cb.cancel() + self._heartbeat_cb = None + + def _reset_heartbeat(self) -> None: + self._cancel_heartbeat() + + if self._heartbeat is not None: + self._heartbeat_cb = call_later( + self._send_heartbeat, self._heartbeat, self._loop + ) + + def _send_heartbeat(self) -> None: + if self._heartbeat is not None and not self._closed: + # fire-and-forget a task is not perfect but maybe ok for + # sending ping. Otherwise we need a long-living heartbeat + # task in the class. + self._loop.create_task(self._writer.ping()) + + if self._pong_response_cb is not None: + self._pong_response_cb.cancel() + self._pong_response_cb = call_later( + self._pong_not_received, self._pong_heartbeat, self._loop + ) + + def _pong_not_received(self) -> None: + if not self._closed: + self._closed = True + self._close_code = WSCloseCode.ABNORMAL_CLOSURE + self._exception = asyncio.TimeoutError() + self._response.close() + + @property + def closed(self) -> bool: + return self._closed + + @property + def close_code(self) -> Optional[int]: + return self._close_code + + @property + def protocol(self) -> Optional[str]: + return self._protocol + + @property + def compress(self) -> int: + return self._compress + + @property + def client_notakeover(self) -> bool: + return self._client_notakeover + + def get_extra_info(self, name: str, default: Any = None) -> Any: + """extra info from connection transport""" + conn = self._response.connection + if conn is None: + return default + transport = conn.transport + if transport is None: + return default + return transport.get_extra_info(name, default) + + def exception(self) -> Optional[BaseException]: + return self._exception + + async def ping(self, message: bytes = b"") -> None: + await self._writer.ping(message) + + async def pong(self, message: bytes = b"") -> None: + await self._writer.pong(message) + + async def send_str(self, data: str, compress: Optional[int] = None) -> None: + if not isinstance(data, str): + raise TypeError("data argument must be str (%r)" % type(data)) + await self._writer.send(data, binary=False, compress=compress) + + async def send_bytes(self, data: bytes, compress: Optional[int] = None) -> None: + if not isinstance(data, (bytes, bytearray, memoryview)): + raise TypeError("data argument must be byte-ish (%r)" % type(data)) + await self._writer.send(data, binary=True, compress=compress) + + async def send_json( + self, + data: Any, + compress: Optional[int] = None, + *, + dumps: JSONEncoder = DEFAULT_JSON_ENCODER, + ) -> None: + await self.send_str(dumps(data), compress=compress) + + async def close(self, *, code: int = WSCloseCode.OK, message: bytes = b"") -> bool: + # we need to break `receive()` cycle first, + # `close()` may be called from different task + if self._waiting is not None and not self._closed: + self._reader.feed_data(WS_CLOSING_MESSAGE, 0) + await self._waiting + + if not self._closed: + self._cancel_heartbeat() + self._closed = True + try: + await self._writer.close(code, message) + except asyncio.CancelledError: + self._close_code = WSCloseCode.ABNORMAL_CLOSURE + self._response.close() + raise + except Exception as exc: + self._close_code = WSCloseCode.ABNORMAL_CLOSURE + self._exception = exc + self._response.close() + return True + + if self._closing: + self._response.close() + return True + + while True: + try: + async with async_timeout.timeout(self._timeout): + msg = await self._reader.read() + except asyncio.CancelledError: + self._close_code = WSCloseCode.ABNORMAL_CLOSURE + self._response.close() + raise + except Exception as exc: + self._close_code = WSCloseCode.ABNORMAL_CLOSURE + self._exception = exc + self._response.close() + return True + + if msg.type == WSMsgType.CLOSE: + self._close_code = msg.data + self._response.close() + return True + else: + return False + + async def receive(self, timeout: Optional[float] = None) -> WSMessage: + while True: + if self._waiting is not None: + raise RuntimeError("Concurrent call to receive() is not allowed") + + if self._closed: + return WS_CLOSED_MESSAGE + elif self._closing: + await self.close() + return WS_CLOSED_MESSAGE + + try: + self._waiting = self._loop.create_future() + try: + async with async_timeout.timeout(timeout or self._receive_timeout): + msg = await self._reader.read() + self._reset_heartbeat() + finally: + waiter = self._waiting + self._waiting = None + set_result(waiter, True) + except (asyncio.CancelledError, asyncio.TimeoutError): + self._close_code = WSCloseCode.ABNORMAL_CLOSURE + raise + except EofStream: + self._close_code = WSCloseCode.OK + await self.close() + return WSMessage(WSMsgType.CLOSED, None, None) + except ClientError: + self._closed = True + self._close_code = WSCloseCode.ABNORMAL_CLOSURE + return WS_CLOSED_MESSAGE + except WebSocketError as exc: + self._close_code = exc.code + await self.close(code=exc.code) + return WSMessage(WSMsgType.ERROR, exc, None) + except Exception as exc: + self._exception = exc + self._closing = True + self._close_code = WSCloseCode.ABNORMAL_CLOSURE + await self.close() + return WSMessage(WSMsgType.ERROR, exc, None) + + if msg.type == WSMsgType.CLOSE: + self._closing = True + self._close_code = msg.data + if not self._closed and self._autoclose: + await self.close() + elif msg.type == WSMsgType.CLOSING: + self._closing = True + elif msg.type == WSMsgType.PING and self._autoping: + await self.pong(msg.data) + continue + elif msg.type == WSMsgType.PONG and self._autoping: + continue + + return msg + + async def receive_str(self, *, timeout: Optional[float] = None) -> str: + msg = await self.receive(timeout) + if msg.type != WSMsgType.TEXT: + raise TypeError(f"Received message {msg.type}:{msg.data!r} is not str") + return cast(str, msg.data) + + async def receive_bytes(self, *, timeout: Optional[float] = None) -> bytes: + msg = await self.receive(timeout) + if msg.type != WSMsgType.BINARY: + raise TypeError(f"Received message {msg.type}:{msg.data!r} is not bytes") + return cast(bytes, msg.data) + + async def receive_json( + self, + *, + loads: JSONDecoder = DEFAULT_JSON_DECODER, + timeout: Optional[float] = None, + ) -> Any: + data = await self.receive_str(timeout=timeout) + return loads(data) + + def __aiter__(self) -> "ClientWebSocketResponse": + return self + + async def __anext__(self) -> WSMessage: + msg = await self.receive() + if msg.type in (WSMsgType.CLOSE, WSMsgType.CLOSING, WSMsgType.CLOSED): + raise StopAsyncIteration + return msg diff --git a/contrib/python/aiohttp/aiohttp/connector.py b/contrib/python/aiohttp/aiohttp/connector.py new file mode 100644 index 0000000000..4c9a951d6e --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/connector.py @@ -0,0 +1,1451 @@ +import asyncio +import functools +import random +import sys +import traceback +import warnings +from collections import defaultdict, deque +from contextlib import suppress +from http.cookies import SimpleCookie +from itertools import cycle, islice +from time import monotonic +from types import TracebackType +from typing import ( + TYPE_CHECKING, + Any, + Awaitable, + Callable, + DefaultDict, + Dict, + Iterator, + List, + Optional, + Set, + Tuple, + Type, + Union, + cast, +) + +import attr + +from . import hdrs, helpers +from .abc import AbstractResolver +from .client_exceptions import ( + ClientConnectionError, + ClientConnectorCertificateError, + ClientConnectorError, + ClientConnectorSSLError, + ClientHttpProxyError, + ClientProxyConnectionError, + ServerFingerprintMismatch, + UnixClientConnectorError, + cert_errors, + ssl_errors, +) +from .client_proto import ResponseHandler +from .client_reqrep import ClientRequest, Fingerprint, _merge_ssl_params +from .helpers import ( + PY_36, + ceil_timeout, + get_running_loop, + is_ip_address, + noop, + sentinel, +) +from .http import RESPONSES +from .locks import EventResultOrError +from .resolver import DefaultResolver + +try: + import ssl + + SSLContext = ssl.SSLContext +except ImportError: # pragma: no cover + ssl = None # type: ignore[assignment] + SSLContext = object # type: ignore[misc,assignment] + + +__all__ = ("BaseConnector", "TCPConnector", "UnixConnector", "NamedPipeConnector") + + +if TYPE_CHECKING: # pragma: no cover + from .client import ClientTimeout + from .client_reqrep import ConnectionKey + from .tracing import Trace + + +class _DeprecationWaiter: + __slots__ = ("_awaitable", "_awaited") + + def __init__(self, awaitable: Awaitable[Any]) -> None: + self._awaitable = awaitable + self._awaited = False + + def __await__(self) -> Any: + self._awaited = True + return self._awaitable.__await__() + + def __del__(self) -> None: + if not self._awaited: + warnings.warn( + "Connector.close() is a coroutine, " + "please use await connector.close()", + DeprecationWarning, + ) + + +class Connection: + + _source_traceback = None + _transport = None + + def __init__( + self, + connector: "BaseConnector", + key: "ConnectionKey", + protocol: ResponseHandler, + loop: asyncio.AbstractEventLoop, + ) -> None: + self._key = key + self._connector = connector + self._loop = loop + self._protocol = protocol # type: Optional[ResponseHandler] + self._callbacks = [] # type: List[Callable[[], None]] + + if loop.get_debug(): + self._source_traceback = traceback.extract_stack(sys._getframe(1)) + + def __repr__(self) -> str: + return f"Connection<{self._key}>" + + def __del__(self, _warnings: Any = warnings) -> None: + if self._protocol is not None: + if PY_36: + kwargs = {"source": self} + else: + kwargs = {} + _warnings.warn(f"Unclosed connection {self!r}", ResourceWarning, **kwargs) + if self._loop.is_closed(): + return + + self._connector._release(self._key, self._protocol, should_close=True) + + context = {"client_connection": self, "message": "Unclosed connection"} + if self._source_traceback is not None: + context["source_traceback"] = self._source_traceback + self._loop.call_exception_handler(context) + + @property + def loop(self) -> asyncio.AbstractEventLoop: + warnings.warn( + "connector.loop property is deprecated", DeprecationWarning, stacklevel=2 + ) + return self._loop + + @property + def transport(self) -> Optional[asyncio.Transport]: + if self._protocol is None: + return None + return self._protocol.transport + + @property + def protocol(self) -> Optional[ResponseHandler]: + return self._protocol + + def add_callback(self, callback: Callable[[], None]) -> None: + if callback is not None: + self._callbacks.append(callback) + + def _notify_release(self) -> None: + callbacks, self._callbacks = self._callbacks[:], [] + + for cb in callbacks: + with suppress(Exception): + cb() + + def close(self) -> None: + self._notify_release() + + if self._protocol is not None: + self._connector._release(self._key, self._protocol, should_close=True) + self._protocol = None + + def release(self) -> None: + self._notify_release() + + if self._protocol is not None: + self._connector._release( + self._key, self._protocol, should_close=self._protocol.should_close + ) + self._protocol = None + + @property + def closed(self) -> bool: + return self._protocol is None or not self._protocol.is_connected() + + +class _TransportPlaceholder: + """placeholder for BaseConnector.connect function""" + + def close(self) -> None: + pass + + +class BaseConnector: + """Base connector class. + + keepalive_timeout - (optional) Keep-alive timeout. + force_close - Set to True to force close and do reconnect + after each request (and between redirects). + limit - The total number of simultaneous connections. + limit_per_host - Number of simultaneous connections to one host. + enable_cleanup_closed - Enables clean-up closed ssl transports. + Disabled by default. + loop - Optional event loop. + """ + + _closed = True # prevent AttributeError in __del__ if ctor was failed + _source_traceback = None + + # abort transport after 2 seconds (cleanup broken connections) + _cleanup_closed_period = 2.0 + + def __init__( + self, + *, + keepalive_timeout: Union[object, None, float] = sentinel, + force_close: bool = False, + limit: int = 100, + limit_per_host: int = 0, + enable_cleanup_closed: bool = False, + loop: Optional[asyncio.AbstractEventLoop] = None, + ) -> None: + + if force_close: + if keepalive_timeout is not None and keepalive_timeout is not sentinel: + raise ValueError( + "keepalive_timeout cannot " "be set if force_close is True" + ) + else: + if keepalive_timeout is sentinel: + keepalive_timeout = 15.0 + + loop = get_running_loop(loop) + + self._closed = False + if loop.get_debug(): + self._source_traceback = traceback.extract_stack(sys._getframe(1)) + + self._conns = ( + {} + ) # type: Dict[ConnectionKey, List[Tuple[ResponseHandler, float]]] + self._limit = limit + self._limit_per_host = limit_per_host + self._acquired = set() # type: Set[ResponseHandler] + self._acquired_per_host = defaultdict( + set + ) # type: DefaultDict[ConnectionKey, Set[ResponseHandler]] + self._keepalive_timeout = cast(float, keepalive_timeout) + self._force_close = force_close + + # {host_key: FIFO list of waiters} + self._waiters = defaultdict(deque) # type: ignore[var-annotated] + + self._loop = loop + self._factory = functools.partial(ResponseHandler, loop=loop) + + self.cookies = SimpleCookie() # type: SimpleCookie[str] + + # start keep-alive connection cleanup task + self._cleanup_handle: Optional[asyncio.TimerHandle] = None + + # start cleanup closed transports task + self._cleanup_closed_handle: Optional[asyncio.TimerHandle] = None + self._cleanup_closed_disabled = not enable_cleanup_closed + self._cleanup_closed_transports = [] # type: List[Optional[asyncio.Transport]] + self._cleanup_closed() + + def __del__(self, _warnings: Any = warnings) -> None: + if self._closed: + return + if not self._conns: + return + + conns = [repr(c) for c in self._conns.values()] + + self._close() + + if PY_36: + kwargs = {"source": self} + else: + kwargs = {} + _warnings.warn(f"Unclosed connector {self!r}", ResourceWarning, **kwargs) + context = { + "connector": self, + "connections": conns, + "message": "Unclosed connector", + } + if self._source_traceback is not None: + context["source_traceback"] = self._source_traceback + self._loop.call_exception_handler(context) + + def __enter__(self) -> "BaseConnector": + warnings.warn( + '"witn Connector():" is deprecated, ' + 'use "async with Connector():" instead', + DeprecationWarning, + ) + return self + + def __exit__(self, *exc: Any) -> None: + self.close() + + async def __aenter__(self) -> "BaseConnector": + return self + + async def __aexit__( + self, + exc_type: Optional[Type[BaseException]] = None, + exc_value: Optional[BaseException] = None, + exc_traceback: Optional[TracebackType] = None, + ) -> None: + await self.close() + + @property + def force_close(self) -> bool: + """Ultimately close connection on releasing if True.""" + return self._force_close + + @property + def limit(self) -> int: + """The total number for simultaneous connections. + + If limit is 0 the connector has no limit. + The default limit size is 100. + """ + return self._limit + + @property + def limit_per_host(self) -> int: + """The limit for simultaneous connections to the same endpoint. + + Endpoints are the same if they are have equal + (host, port, is_ssl) triple. + """ + return self._limit_per_host + + def _cleanup(self) -> None: + """Cleanup unused transports.""" + if self._cleanup_handle: + self._cleanup_handle.cancel() + # _cleanup_handle should be unset, otherwise _release() will not + # recreate it ever! + self._cleanup_handle = None + + now = self._loop.time() + timeout = self._keepalive_timeout + + if self._conns: + connections = {} + deadline = now - timeout + for key, conns in self._conns.items(): + alive = [] + for proto, use_time in conns: + if proto.is_connected(): + if use_time - deadline < 0: + transport = proto.transport + proto.close() + if key.is_ssl and not self._cleanup_closed_disabled: + self._cleanup_closed_transports.append(transport) + else: + alive.append((proto, use_time)) + else: + transport = proto.transport + proto.close() + if key.is_ssl and not self._cleanup_closed_disabled: + self._cleanup_closed_transports.append(transport) + + if alive: + connections[key] = alive + + self._conns = connections + + if self._conns: + self._cleanup_handle = helpers.weakref_handle( + self, "_cleanup", timeout, self._loop + ) + + def _drop_acquired_per_host( + self, key: "ConnectionKey", val: ResponseHandler + ) -> None: + acquired_per_host = self._acquired_per_host + if key not in acquired_per_host: + return + conns = acquired_per_host[key] + conns.remove(val) + if not conns: + del self._acquired_per_host[key] + + def _cleanup_closed(self) -> None: + """Double confirmation for transport close. + + Some broken ssl servers may leave socket open without proper close. + """ + if self._cleanup_closed_handle: + self._cleanup_closed_handle.cancel() + + for transport in self._cleanup_closed_transports: + if transport is not None: + transport.abort() + + self._cleanup_closed_transports = [] + + if not self._cleanup_closed_disabled: + self._cleanup_closed_handle = helpers.weakref_handle( + self, "_cleanup_closed", self._cleanup_closed_period, self._loop + ) + + def close(self) -> Awaitable[None]: + """Close all opened transports.""" + self._close() + return _DeprecationWaiter(noop()) + + def _close(self) -> None: + if self._closed: + return + + self._closed = True + + try: + if self._loop.is_closed(): + return + + # cancel cleanup task + if self._cleanup_handle: + self._cleanup_handle.cancel() + + # cancel cleanup close task + if self._cleanup_closed_handle: + self._cleanup_closed_handle.cancel() + + for data in self._conns.values(): + for proto, t0 in data: + proto.close() + + for proto in self._acquired: + proto.close() + + for transport in self._cleanup_closed_transports: + if transport is not None: + transport.abort() + + finally: + self._conns.clear() + self._acquired.clear() + self._waiters.clear() + self._cleanup_handle = None + self._cleanup_closed_transports.clear() + self._cleanup_closed_handle = None + + @property + def closed(self) -> bool: + """Is connector closed. + + A readonly property. + """ + return self._closed + + def _available_connections(self, key: "ConnectionKey") -> int: + """ + Return number of available connections. + + The limit, limit_per_host and the connection key are taken into account. + + If it returns less than 1 means that there are no connections + available. + """ + if self._limit: + # total calc available connections + available = self._limit - len(self._acquired) + + # check limit per host + if ( + self._limit_per_host + and available > 0 + and key in self._acquired_per_host + ): + acquired = self._acquired_per_host.get(key) + assert acquired is not None + available = self._limit_per_host - len(acquired) + + elif self._limit_per_host and key in self._acquired_per_host: + # check limit per host + acquired = self._acquired_per_host.get(key) + assert acquired is not None + available = self._limit_per_host - len(acquired) + else: + available = 1 + + return available + + async def connect( + self, req: "ClientRequest", traces: List["Trace"], timeout: "ClientTimeout" + ) -> Connection: + """Get from pool or create new connection.""" + key = req.connection_key + available = self._available_connections(key) + + # Wait if there are no available connections or if there are/were + # waiters (i.e. don't steal connection from a waiter about to wake up) + if available <= 0 or key in self._waiters: + fut = self._loop.create_future() + + # This connection will now count towards the limit. + self._waiters[key].append(fut) + + if traces: + for trace in traces: + await trace.send_connection_queued_start() + + try: + await fut + except BaseException as e: + if key in self._waiters: + # remove a waiter even if it was cancelled, normally it's + # removed when it's notified + try: + self._waiters[key].remove(fut) + except ValueError: # fut may no longer be in list + pass + + raise e + finally: + if key in self._waiters and not self._waiters[key]: + del self._waiters[key] + + if traces: + for trace in traces: + await trace.send_connection_queued_end() + + proto = self._get(key) + if proto is None: + placeholder = cast(ResponseHandler, _TransportPlaceholder()) + self._acquired.add(placeholder) + self._acquired_per_host[key].add(placeholder) + + if traces: + for trace in traces: + await trace.send_connection_create_start() + + try: + proto = await self._create_connection(req, traces, timeout) + if self._closed: + proto.close() + raise ClientConnectionError("Connector is closed.") + except BaseException: + if not self._closed: + self._acquired.remove(placeholder) + self._drop_acquired_per_host(key, placeholder) + self._release_waiter() + raise + else: + if not self._closed: + self._acquired.remove(placeholder) + self._drop_acquired_per_host(key, placeholder) + + if traces: + for trace in traces: + await trace.send_connection_create_end() + else: + if traces: + # Acquire the connection to prevent race conditions with limits + placeholder = cast(ResponseHandler, _TransportPlaceholder()) + self._acquired.add(placeholder) + self._acquired_per_host[key].add(placeholder) + for trace in traces: + await trace.send_connection_reuseconn() + self._acquired.remove(placeholder) + self._drop_acquired_per_host(key, placeholder) + + self._acquired.add(proto) + self._acquired_per_host[key].add(proto) + return Connection(self, key, proto, self._loop) + + def _get(self, key: "ConnectionKey") -> Optional[ResponseHandler]: + try: + conns = self._conns[key] + except KeyError: + return None + + t1 = self._loop.time() + while conns: + proto, t0 = conns.pop() + if proto.is_connected(): + if t1 - t0 > self._keepalive_timeout: + transport = proto.transport + proto.close() + # only for SSL transports + if key.is_ssl and not self._cleanup_closed_disabled: + self._cleanup_closed_transports.append(transport) + else: + if not conns: + # The very last connection was reclaimed: drop the key + del self._conns[key] + return proto + else: + transport = proto.transport + proto.close() + if key.is_ssl and not self._cleanup_closed_disabled: + self._cleanup_closed_transports.append(transport) + + # No more connections: drop the key + del self._conns[key] + return None + + def _release_waiter(self) -> None: + """ + Iterates over all waiters until one to be released is found. + + The one to be released is not finsihed and + belongs to a host that has available connections. + """ + if not self._waiters: + return + + # Having the dict keys ordered this avoids to iterate + # at the same order at each call. + queues = list(self._waiters.keys()) + random.shuffle(queues) + + for key in queues: + if self._available_connections(key) < 1: + continue + + waiters = self._waiters[key] + while waiters: + waiter = waiters.popleft() + if not waiter.done(): + waiter.set_result(None) + return + + def _release_acquired(self, key: "ConnectionKey", proto: ResponseHandler) -> None: + if self._closed: + # acquired connection is already released on connector closing + return + + try: + self._acquired.remove(proto) + self._drop_acquired_per_host(key, proto) + except KeyError: # pragma: no cover + # this may be result of undetermenistic order of objects + # finalization due garbage collection. + pass + else: + self._release_waiter() + + def _release( + self, + key: "ConnectionKey", + protocol: ResponseHandler, + *, + should_close: bool = False, + ) -> None: + if self._closed: + # acquired connection is already released on connector closing + return + + self._release_acquired(key, protocol) + + if self._force_close: + should_close = True + + if should_close or protocol.should_close: + transport = protocol.transport + protocol.close() + + if key.is_ssl and not self._cleanup_closed_disabled: + self._cleanup_closed_transports.append(transport) + else: + conns = self._conns.get(key) + if conns is None: + conns = self._conns[key] = [] + conns.append((protocol, self._loop.time())) + + if self._cleanup_handle is None: + self._cleanup_handle = helpers.weakref_handle( + self, "_cleanup", self._keepalive_timeout, self._loop + ) + + async def _create_connection( + self, req: "ClientRequest", traces: List["Trace"], timeout: "ClientTimeout" + ) -> ResponseHandler: + raise NotImplementedError() + + +class _DNSCacheTable: + def __init__(self, ttl: Optional[float] = None) -> None: + self._addrs_rr = ( + {} + ) # type: Dict[Tuple[str, int], Tuple[Iterator[Dict[str, Any]], int]] + self._timestamps = {} # type: Dict[Tuple[str, int], float] + self._ttl = ttl + + def __contains__(self, host: object) -> bool: + return host in self._addrs_rr + + def add(self, key: Tuple[str, int], addrs: List[Dict[str, Any]]) -> None: + self._addrs_rr[key] = (cycle(addrs), len(addrs)) + + if self._ttl: + self._timestamps[key] = monotonic() + + def remove(self, key: Tuple[str, int]) -> None: + self._addrs_rr.pop(key, None) + + if self._ttl: + self._timestamps.pop(key, None) + + def clear(self) -> None: + self._addrs_rr.clear() + self._timestamps.clear() + + def next_addrs(self, key: Tuple[str, int]) -> List[Dict[str, Any]]: + loop, length = self._addrs_rr[key] + addrs = list(islice(loop, length)) + # Consume one more element to shift internal state of `cycle` + next(loop) + return addrs + + def expired(self, key: Tuple[str, int]) -> bool: + if self._ttl is None: + return False + + return self._timestamps[key] + self._ttl < monotonic() + + +class TCPConnector(BaseConnector): + """TCP connector. + + verify_ssl - Set to True to check ssl certifications. + fingerprint - Pass the binary sha256 + digest of the expected certificate in DER format to verify + that the certificate the server presents matches. See also + https://en.wikipedia.org/wiki/Transport_Layer_Security#Certificate_pinning + resolver - Enable DNS lookups and use this + resolver + use_dns_cache - Use memory cache for DNS lookups. + ttl_dns_cache - Max seconds having cached a DNS entry, None forever. + family - socket address family + local_addr - local tuple of (host, port) to bind socket to + + keepalive_timeout - (optional) Keep-alive timeout. + force_close - Set to True to force close and do reconnect + after each request (and between redirects). + limit - The total number of simultaneous connections. + limit_per_host - Number of simultaneous connections to one host. + enable_cleanup_closed - Enables clean-up closed ssl transports. + Disabled by default. + loop - Optional event loop. + """ + + def __init__( + self, + *, + verify_ssl: bool = True, + fingerprint: Optional[bytes] = None, + use_dns_cache: bool = True, + ttl_dns_cache: Optional[int] = 10, + family: int = 0, + ssl_context: Optional[SSLContext] = None, + ssl: Union[None, bool, Fingerprint, SSLContext] = None, + local_addr: Optional[Tuple[str, int]] = None, + resolver: Optional[AbstractResolver] = None, + keepalive_timeout: Union[None, float, object] = sentinel, + force_close: bool = False, + limit: int = 100, + limit_per_host: int = 0, + enable_cleanup_closed: bool = False, + loop: Optional[asyncio.AbstractEventLoop] = None, + ): + super().__init__( + keepalive_timeout=keepalive_timeout, + force_close=force_close, + limit=limit, + limit_per_host=limit_per_host, + enable_cleanup_closed=enable_cleanup_closed, + loop=loop, + ) + + self._ssl = _merge_ssl_params(ssl, verify_ssl, ssl_context, fingerprint) + if resolver is None: + resolver = DefaultResolver(loop=self._loop) + self._resolver = resolver + + self._use_dns_cache = use_dns_cache + self._cached_hosts = _DNSCacheTable(ttl=ttl_dns_cache) + self._throttle_dns_events = ( + {} + ) # type: Dict[Tuple[str, int], EventResultOrError] + self._family = family + self._local_addr = local_addr + + def close(self) -> Awaitable[None]: + """Close all ongoing DNS calls.""" + for ev in self._throttle_dns_events.values(): + ev.cancel() + + return super().close() + + @property + def family(self) -> int: + """Socket family like AF_INET.""" + return self._family + + @property + def use_dns_cache(self) -> bool: + """True if local DNS caching is enabled.""" + return self._use_dns_cache + + def clear_dns_cache( + self, host: Optional[str] = None, port: Optional[int] = None + ) -> None: + """Remove specified host/port or clear all dns local cache.""" + if host is not None and port is not None: + self._cached_hosts.remove((host, port)) + elif host is not None or port is not None: + raise ValueError("either both host and port " "or none of them are allowed") + else: + self._cached_hosts.clear() + + async def _resolve_host( + self, host: str, port: int, traces: Optional[List["Trace"]] = None + ) -> List[Dict[str, Any]]: + if is_ip_address(host): + return [ + { + "hostname": host, + "host": host, + "port": port, + "family": self._family, + "proto": 0, + "flags": 0, + } + ] + + if not self._use_dns_cache: + + if traces: + for trace in traces: + await trace.send_dns_resolvehost_start(host) + + res = await self._resolver.resolve(host, port, family=self._family) + + if traces: + for trace in traces: + await trace.send_dns_resolvehost_end(host) + + return res + + key = (host, port) + + if (key in self._cached_hosts) and (not self._cached_hosts.expired(key)): + # get result early, before any await (#4014) + result = self._cached_hosts.next_addrs(key) + + if traces: + for trace in traces: + await trace.send_dns_cache_hit(host) + return result + + if key in self._throttle_dns_events: + # get event early, before any await (#4014) + event = self._throttle_dns_events[key] + if traces: + for trace in traces: + await trace.send_dns_cache_hit(host) + await event.wait() + else: + # update dict early, before any await (#4014) + self._throttle_dns_events[key] = EventResultOrError(self._loop) + if traces: + for trace in traces: + await trace.send_dns_cache_miss(host) + try: + + if traces: + for trace in traces: + await trace.send_dns_resolvehost_start(host) + + addrs = await self._resolver.resolve(host, port, family=self._family) + if traces: + for trace in traces: + await trace.send_dns_resolvehost_end(host) + + self._cached_hosts.add(key, addrs) + self._throttle_dns_events[key].set() + except BaseException as e: + # any DNS exception, independently of the implementation + # is set for the waiters to raise the same exception. + self._throttle_dns_events[key].set(exc=e) + raise + finally: + self._throttle_dns_events.pop(key) + + return self._cached_hosts.next_addrs(key) + + async def _create_connection( + self, req: "ClientRequest", traces: List["Trace"], timeout: "ClientTimeout" + ) -> ResponseHandler: + """Create connection. + + Has same keyword arguments as BaseEventLoop.create_connection. + """ + if req.proxy: + _, proto = await self._create_proxy_connection(req, traces, timeout) + else: + _, proto = await self._create_direct_connection(req, traces, timeout) + + return proto + + @staticmethod + @functools.lru_cache(None) + def _make_ssl_context(verified: bool) -> SSLContext: + if verified: + return ssl.create_default_context() + else: + sslcontext = ssl.SSLContext(ssl.PROTOCOL_TLS_CLIENT) + sslcontext.options |= ssl.OP_NO_SSLv2 + sslcontext.options |= ssl.OP_NO_SSLv3 + sslcontext.check_hostname = False + sslcontext.verify_mode = ssl.CERT_NONE + try: + sslcontext.options |= ssl.OP_NO_COMPRESSION + except AttributeError as attr_err: + warnings.warn( + "{!s}: The Python interpreter is compiled " + "against OpenSSL < 1.0.0. Ref: " + "https://docs.python.org/3/library/ssl.html" + "#ssl.OP_NO_COMPRESSION".format(attr_err), + ) + sslcontext.set_default_verify_paths() + return sslcontext + + def _get_ssl_context(self, req: "ClientRequest") -> Optional[SSLContext]: + """Logic to get the correct SSL context + + 0. if req.ssl is false, return None + + 1. if ssl_context is specified in req, use it + 2. if _ssl_context is specified in self, use it + 3. otherwise: + 1. if verify_ssl is not specified in req, use self.ssl_context + (will generate a default context according to self.verify_ssl) + 2. if verify_ssl is True in req, generate a default SSL context + 3. if verify_ssl is False in req, generate a SSL context that + won't verify + """ + if req.is_ssl(): + if ssl is None: # pragma: no cover + raise RuntimeError("SSL is not supported.") + sslcontext = req.ssl + if isinstance(sslcontext, ssl.SSLContext): + return sslcontext + if sslcontext is not None: + # not verified or fingerprinted + return self._make_ssl_context(False) + sslcontext = self._ssl + if isinstance(sslcontext, ssl.SSLContext): + return sslcontext + if sslcontext is not None: + # not verified or fingerprinted + return self._make_ssl_context(False) + return self._make_ssl_context(True) + else: + return None + + def _get_fingerprint(self, req: "ClientRequest") -> Optional["Fingerprint"]: + ret = req.ssl + if isinstance(ret, Fingerprint): + return ret + ret = self._ssl + if isinstance(ret, Fingerprint): + return ret + return None + + async def _wrap_create_connection( + self, + *args: Any, + req: "ClientRequest", + timeout: "ClientTimeout", + client_error: Type[Exception] = ClientConnectorError, + **kwargs: Any, + ) -> Tuple[asyncio.Transport, ResponseHandler]: + try: + async with ceil_timeout(timeout.sock_connect): + return await self._loop.create_connection(*args, **kwargs) # type: ignore[return-value] # noqa + except cert_errors as exc: + raise ClientConnectorCertificateError(req.connection_key, exc) from exc + except ssl_errors as exc: + raise ClientConnectorSSLError(req.connection_key, exc) from exc + except OSError as exc: + raise client_error(req.connection_key, exc) from exc + + def _fail_on_no_start_tls(self, req: "ClientRequest") -> None: + """Raise a :py:exc:`RuntimeError` on missing ``start_tls()``. + + One case is that :py:meth:`asyncio.loop.start_tls` is not yet + implemented under Python 3.6. It is necessary for TLS-in-TLS so + that it is possible to send HTTPS queries through HTTPS proxies. + + This doesn't affect regular HTTP requests, though. + """ + if not req.is_ssl(): + return + + proxy_url = req.proxy + assert proxy_url is not None + if proxy_url.scheme != "https": + return + + self._check_loop_for_start_tls() + + def _check_loop_for_start_tls(self) -> None: + try: + self._loop.start_tls + except AttributeError as attr_exc: + raise RuntimeError( + "An HTTPS request is being sent through an HTTPS proxy. " + "This needs support for TLS in TLS but it is not implemented " + "in your runtime for the stdlib asyncio.\n\n" + "Please upgrade to Python 3.7 or higher. For more details, " + "please see:\n" + "* https://bugs.python.org/issue37179\n" + "* https://github.com/python/cpython/pull/28073\n" + "* https://docs.aiohttp.org/en/stable/" + "client_advanced.html#proxy-support\n" + "* https://github.com/aio-libs/aiohttp/discussions/6044\n", + ) from attr_exc + + def _loop_supports_start_tls(self) -> bool: + try: + self._check_loop_for_start_tls() + except RuntimeError: + return False + else: + return True + + def _warn_about_tls_in_tls( + self, + underlying_transport: asyncio.Transport, + req: "ClientRequest", + ) -> None: + """Issue a warning if the requested URL has HTTPS scheme.""" + if req.request_info.url.scheme != "https": + return + + asyncio_supports_tls_in_tls = getattr( + underlying_transport, + "_start_tls_compatible", + False, + ) + + if asyncio_supports_tls_in_tls: + return + + warnings.warn( + "An HTTPS request is being sent through an HTTPS proxy. " + "This support for TLS in TLS is known to be disabled " + "in the stdlib asyncio. This is why you'll probably see " + "an error in the log below.\n\n" + "It is possible to enable it via monkeypatching under " + "Python 3.7 or higher. For more details, see:\n" + "* https://bugs.python.org/issue37179\n" + "* https://github.com/python/cpython/pull/28073\n\n" + "You can temporarily patch this as follows:\n" + "* https://docs.aiohttp.org/en/stable/client_advanced.html#proxy-support\n" + "* https://github.com/aio-libs/aiohttp/discussions/6044\n", + RuntimeWarning, + source=self, + # Why `4`? At least 3 of the calls in the stack originate + # from the methods in this class. + stacklevel=3, + ) + + async def _start_tls_connection( + self, + underlying_transport: asyncio.Transport, + req: "ClientRequest", + timeout: "ClientTimeout", + client_error: Type[Exception] = ClientConnectorError, + ) -> Tuple[asyncio.BaseTransport, ResponseHandler]: + """Wrap the raw TCP transport with TLS.""" + tls_proto = self._factory() # Create a brand new proto for TLS + + # Safety of the `cast()` call here is based on the fact that + # internally `_get_ssl_context()` only returns `None` when + # `req.is_ssl()` evaluates to `False` which is never gonna happen + # in this code path. Of course, it's rather fragile + # maintainability-wise but this is to be solved separately. + sslcontext = cast(ssl.SSLContext, self._get_ssl_context(req)) + + try: + async with ceil_timeout(timeout.sock_connect): + try: + tls_transport = await self._loop.start_tls( + underlying_transport, + tls_proto, + sslcontext, + server_hostname=req.host, + ssl_handshake_timeout=timeout.total, + ) + except BaseException: + # We need to close the underlying transport since + # `start_tls()` probably failed before it had a + # chance to do this: + underlying_transport.close() + raise + except cert_errors as exc: + raise ClientConnectorCertificateError(req.connection_key, exc) from exc + except ssl_errors as exc: + raise ClientConnectorSSLError(req.connection_key, exc) from exc + except OSError as exc: + raise client_error(req.connection_key, exc) from exc + except TypeError as type_err: + # Example cause looks like this: + # TypeError: transport <asyncio.sslproto._SSLProtocolTransport + # object at 0x7f760615e460> is not supported by start_tls() + + raise ClientConnectionError( + "Cannot initialize a TLS-in-TLS connection to host " + f"{req.host!s}:{req.port:d} through an underlying connection " + f"to an HTTPS proxy {req.proxy!s} ssl:{req.ssl or 'default'} " + f"[{type_err!s}]" + ) from type_err + else: + tls_proto.connection_made( + tls_transport + ) # Kick the state machine of the new TLS protocol + + return tls_transport, tls_proto + + async def _create_direct_connection( + self, + req: "ClientRequest", + traces: List["Trace"], + timeout: "ClientTimeout", + *, + client_error: Type[Exception] = ClientConnectorError, + ) -> Tuple[asyncio.Transport, ResponseHandler]: + sslcontext = self._get_ssl_context(req) + fingerprint = self._get_fingerprint(req) + + host = req.url.raw_host + assert host is not None + port = req.port + assert port is not None + host_resolved = asyncio.ensure_future( + self._resolve_host(host, port, traces=traces), loop=self._loop + ) + try: + # Cancelling this lookup should not cancel the underlying lookup + # or else the cancel event will get broadcast to all the waiters + # across all connections. + hosts = await asyncio.shield(host_resolved) + except asyncio.CancelledError: + + def drop_exception(fut: "asyncio.Future[List[Dict[str, Any]]]") -> None: + with suppress(Exception, asyncio.CancelledError): + fut.result() + + host_resolved.add_done_callback(drop_exception) + raise + except OSError as exc: + # in case of proxy it is not ClientProxyConnectionError + # it is problem of resolving proxy ip itself + raise ClientConnectorError(req.connection_key, exc) from exc + + last_exc = None # type: Optional[Exception] + + for hinfo in hosts: + host = hinfo["host"] + port = hinfo["port"] + + try: + transp, proto = await self._wrap_create_connection( + self._factory, + host, + port, + timeout=timeout, + ssl=sslcontext, + family=hinfo["family"], + proto=hinfo["proto"], + flags=hinfo["flags"], + server_hostname=hinfo["hostname"] if sslcontext else None, + local_addr=self._local_addr, + req=req, + client_error=client_error, + ) + except ClientConnectorError as exc: + last_exc = exc + continue + + if req.is_ssl() and fingerprint: + try: + fingerprint.check(transp) + except ServerFingerprintMismatch as exc: + transp.close() + if not self._cleanup_closed_disabled: + self._cleanup_closed_transports.append(transp) + last_exc = exc + continue + + return transp, proto + else: + assert last_exc is not None + raise last_exc + + async def _create_proxy_connection( + self, req: "ClientRequest", traces: List["Trace"], timeout: "ClientTimeout" + ) -> Tuple[asyncio.BaseTransport, ResponseHandler]: + self._fail_on_no_start_tls(req) + runtime_has_start_tls = self._loop_supports_start_tls() + if req.proxy.scheme != "https": + runtime_has_start_tls = False + + headers = {} # type: Dict[str, str] + if req.proxy_headers is not None: + headers = req.proxy_headers # type: ignore[assignment] + headers[hdrs.HOST] = req.headers[hdrs.HOST] + + url = req.proxy + assert url is not None + proxy_req = ClientRequest( + hdrs.METH_GET, + url, + headers=headers, + auth=req.proxy_auth, + loop=self._loop, + ssl=req.ssl, + ) + + # create connection to proxy server + transport, proto = await self._create_direct_connection( + proxy_req, [], timeout, client_error=ClientProxyConnectionError + ) + + # Many HTTP proxies has buggy keepalive support. Let's not + # reuse connection but close it after processing every + # response. + proto.force_close() + + auth = proxy_req.headers.pop(hdrs.AUTHORIZATION, None) + if auth is not None: + if not req.is_ssl(): + req.headers[hdrs.PROXY_AUTHORIZATION] = auth + else: + proxy_req.headers[hdrs.PROXY_AUTHORIZATION] = auth + + if req.is_ssl(): + if runtime_has_start_tls: + self._warn_about_tls_in_tls(transport, req) + + # For HTTPS requests over HTTP proxy + # we must notify proxy to tunnel connection + # so we send CONNECT command: + # CONNECT www.python.org:443 HTTP/1.1 + # Host: www.python.org + # + # next we must do TLS handshake and so on + # to do this we must wrap raw socket into secure one + # asyncio handles this perfectly + proxy_req.method = hdrs.METH_CONNECT + proxy_req.url = req.url + key = attr.evolve( + req.connection_key, proxy=None, proxy_auth=None, proxy_headers_hash=None + ) + conn = Connection(self, key, proto, self._loop) + proxy_resp = await proxy_req.send(conn) + try: + protocol = conn._protocol + assert protocol is not None + + # read_until_eof=True will ensure the connection isn't closed + # once the response is received and processed allowing + # START_TLS to work on the connection below. + protocol.set_response_params(read_until_eof=runtime_has_start_tls) + resp = await proxy_resp.start(conn) + except BaseException: + proxy_resp.close() + conn.close() + raise + else: + conn._protocol = None + conn._transport = None + try: + if resp.status != 200: + message = resp.reason + if message is None: + message = RESPONSES[resp.status][0] + raise ClientHttpProxyError( + proxy_resp.request_info, + resp.history, + status=resp.status, + message=message, + headers=resp.headers, + ) + if not runtime_has_start_tls: + rawsock = transport.get_extra_info("socket", default=None) + if rawsock is None: + raise RuntimeError( + "Transport does not expose socket instance" + ) + # Duplicate the socket, so now we can close proxy transport + rawsock = rawsock.dup() + except BaseException: + # It shouldn't be closed in `finally` because it's fed to + # `loop.start_tls()` and the docs say not to touch it after + # passing there. + transport.close() + raise + finally: + if not runtime_has_start_tls: + transport.close() + + if not runtime_has_start_tls: + # HTTP proxy with support for upgrade to HTTPS + sslcontext = self._get_ssl_context(req) + return await self._wrap_create_connection( + self._factory, + timeout=timeout, + ssl=sslcontext, + sock=rawsock, + server_hostname=req.host, + req=req, + ) + + return await self._start_tls_connection( + # Access the old transport for the last time before it's + # closed and forgotten forever: + transport, + req=req, + timeout=timeout, + ) + finally: + proxy_resp.close() + + return transport, proto + + +class UnixConnector(BaseConnector): + """Unix socket connector. + + path - Unix socket path. + keepalive_timeout - (optional) Keep-alive timeout. + force_close - Set to True to force close and do reconnect + after each request (and between redirects). + limit - The total number of simultaneous connections. + limit_per_host - Number of simultaneous connections to one host. + loop - Optional event loop. + """ + + def __init__( + self, + path: str, + force_close: bool = False, + keepalive_timeout: Union[object, float, None] = sentinel, + limit: int = 100, + limit_per_host: int = 0, + loop: Optional[asyncio.AbstractEventLoop] = None, + ) -> None: + super().__init__( + force_close=force_close, + keepalive_timeout=keepalive_timeout, + limit=limit, + limit_per_host=limit_per_host, + loop=loop, + ) + self._path = path + + @property + def path(self) -> str: + """Path to unix socket.""" + return self._path + + async def _create_connection( + self, req: "ClientRequest", traces: List["Trace"], timeout: "ClientTimeout" + ) -> ResponseHandler: + try: + async with ceil_timeout(timeout.sock_connect): + _, proto = await self._loop.create_unix_connection( + self._factory, self._path + ) + except OSError as exc: + raise UnixClientConnectorError(self.path, req.connection_key, exc) from exc + + return cast(ResponseHandler, proto) + + +class NamedPipeConnector(BaseConnector): + """Named pipe connector. + + Only supported by the proactor event loop. + See also: https://docs.python.org/3.7/library/asyncio-eventloop.html + + path - Windows named pipe path. + keepalive_timeout - (optional) Keep-alive timeout. + force_close - Set to True to force close and do reconnect + after each request (and between redirects). + limit - The total number of simultaneous connections. + limit_per_host - Number of simultaneous connections to one host. + loop - Optional event loop. + """ + + def __init__( + self, + path: str, + force_close: bool = False, + keepalive_timeout: Union[object, float, None] = sentinel, + limit: int = 100, + limit_per_host: int = 0, + loop: Optional[asyncio.AbstractEventLoop] = None, + ) -> None: + super().__init__( + force_close=force_close, + keepalive_timeout=keepalive_timeout, + limit=limit, + limit_per_host=limit_per_host, + loop=loop, + ) + if not isinstance( + self._loop, asyncio.ProactorEventLoop # type: ignore[attr-defined] + ): + raise RuntimeError( + "Named Pipes only available in proactor " "loop under windows" + ) + self._path = path + + @property + def path(self) -> str: + """Path to the named pipe.""" + return self._path + + async def _create_connection( + self, req: "ClientRequest", traces: List["Trace"], timeout: "ClientTimeout" + ) -> ResponseHandler: + try: + async with ceil_timeout(timeout.sock_connect): + _, proto = await self._loop.create_pipe_connection( # type: ignore[attr-defined] # noqa: E501 + self._factory, self._path + ) + # the drain is required so that the connection_made is called + # and transport is set otherwise it is not set before the + # `assert conn.transport is not None` + # in client.py's _request method + await asyncio.sleep(0) + # other option is to manually set transport like + # `proto.transport = trans` + except OSError as exc: + raise ClientConnectorError(req.connection_key, exc) from exc + + return cast(ResponseHandler, proto) diff --git a/contrib/python/aiohttp/aiohttp/cookiejar.py b/contrib/python/aiohttp/aiohttp/cookiejar.py new file mode 100644 index 0000000000..0a2656634d --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/cookiejar.py @@ -0,0 +1,413 @@ +import asyncio +import contextlib +import datetime +import os # noqa +import pathlib +import pickle +import re +from collections import defaultdict +from http.cookies import BaseCookie, Morsel, SimpleCookie +from typing import ( # noqa + DefaultDict, + Dict, + Iterable, + Iterator, + List, + Mapping, + Optional, + Set, + Tuple, + Union, + cast, +) + +from yarl import URL + +from .abc import AbstractCookieJar, ClearCookiePredicate +from .helpers import is_ip_address, next_whole_second +from .typedefs import LooseCookies, PathLike, StrOrURL + +__all__ = ("CookieJar", "DummyCookieJar") + + +CookieItem = Union[str, "Morsel[str]"] + + +class CookieJar(AbstractCookieJar): + """Implements cookie storage adhering to RFC 6265.""" + + DATE_TOKENS_RE = re.compile( + r"[\x09\x20-\x2F\x3B-\x40\x5B-\x60\x7B-\x7E]*" + r"(?P<token>[\x00-\x08\x0A-\x1F\d:a-zA-Z\x7F-\xFF]+)" + ) + + DATE_HMS_TIME_RE = re.compile(r"(\d{1,2}):(\d{1,2}):(\d{1,2})") + + DATE_DAY_OF_MONTH_RE = re.compile(r"(\d{1,2})") + + DATE_MONTH_RE = re.compile( + "(jan)|(feb)|(mar)|(apr)|(may)|(jun)|(jul)|" "(aug)|(sep)|(oct)|(nov)|(dec)", + re.I, + ) + + DATE_YEAR_RE = re.compile(r"(\d{2,4})") + + MAX_TIME = datetime.datetime.max.replace(tzinfo=datetime.timezone.utc) + + MAX_32BIT_TIME = datetime.datetime.utcfromtimestamp(2 ** 31 - 1) + + def __init__( + self, + *, + unsafe: bool = False, + quote_cookie: bool = True, + treat_as_secure_origin: Union[StrOrURL, List[StrOrURL], None] = None, + loop: Optional[asyncio.AbstractEventLoop] = None, + ) -> None: + super().__init__(loop=loop) + self._cookies = defaultdict( + SimpleCookie + ) # type: DefaultDict[str, SimpleCookie[str]] + self._host_only_cookies = set() # type: Set[Tuple[str, str]] + self._unsafe = unsafe + self._quote_cookie = quote_cookie + if treat_as_secure_origin is None: + treat_as_secure_origin = [] + elif isinstance(treat_as_secure_origin, URL): + treat_as_secure_origin = [treat_as_secure_origin.origin()] + elif isinstance(treat_as_secure_origin, str): + treat_as_secure_origin = [URL(treat_as_secure_origin).origin()] + else: + treat_as_secure_origin = [ + URL(url).origin() if isinstance(url, str) else url.origin() + for url in treat_as_secure_origin + ] + self._treat_as_secure_origin = treat_as_secure_origin + self._next_expiration = next_whole_second() + self._expirations = {} # type: Dict[Tuple[str, str], datetime.datetime] + # #4515: datetime.max may not be representable on 32-bit platforms + self._max_time = self.MAX_TIME + try: + self._max_time.timestamp() + except OverflowError: + self._max_time = self.MAX_32BIT_TIME + + def save(self, file_path: PathLike) -> None: + file_path = pathlib.Path(file_path) + with file_path.open(mode="wb") as f: + pickle.dump(self._cookies, f, pickle.HIGHEST_PROTOCOL) + + def load(self, file_path: PathLike) -> None: + file_path = pathlib.Path(file_path) + with file_path.open(mode="rb") as f: + self._cookies = pickle.load(f) + + def clear(self, predicate: Optional[ClearCookiePredicate] = None) -> None: + if predicate is None: + self._next_expiration = next_whole_second() + self._cookies.clear() + self._host_only_cookies.clear() + self._expirations.clear() + return + + to_del = [] + now = datetime.datetime.now(datetime.timezone.utc) + for domain, cookie in self._cookies.items(): + for name, morsel in cookie.items(): + key = (domain, name) + if ( + key in self._expirations and self._expirations[key] <= now + ) or predicate(morsel): + to_del.append(key) + + for domain, name in to_del: + key = (domain, name) + self._host_only_cookies.discard(key) + if key in self._expirations: + del self._expirations[(domain, name)] + self._cookies[domain].pop(name, None) + + next_expiration = min(self._expirations.values(), default=self._max_time) + try: + self._next_expiration = next_expiration.replace( + microsecond=0 + ) + datetime.timedelta(seconds=1) + except OverflowError: + self._next_expiration = self._max_time + + def clear_domain(self, domain: str) -> None: + self.clear(lambda x: self._is_domain_match(domain, x["domain"])) + + def __iter__(self) -> "Iterator[Morsel[str]]": + self._do_expiration() + for val in self._cookies.values(): + yield from val.values() + + def __len__(self) -> int: + return sum(1 for i in self) + + def _do_expiration(self) -> None: + self.clear(lambda x: False) + + def _expire_cookie(self, when: datetime.datetime, domain: str, name: str) -> None: + self._next_expiration = min(self._next_expiration, when) + self._expirations[(domain, name)] = when + + def update_cookies(self, cookies: LooseCookies, response_url: URL = URL()) -> None: + """Update cookies.""" + hostname = response_url.raw_host + + if not self._unsafe and is_ip_address(hostname): + # Don't accept cookies from IPs + return + + if isinstance(cookies, Mapping): + cookies = cookies.items() + + for name, cookie in cookies: + if not isinstance(cookie, Morsel): + tmp = SimpleCookie() # type: SimpleCookie[str] + tmp[name] = cookie # type: ignore[assignment] + cookie = tmp[name] + + domain = cookie["domain"] + + # ignore domains with trailing dots + if domain.endswith("."): + domain = "" + del cookie["domain"] + + if not domain and hostname is not None: + # Set the cookie's domain to the response hostname + # and set its host-only-flag + self._host_only_cookies.add((hostname, name)) + domain = cookie["domain"] = hostname + + if domain.startswith("."): + # Remove leading dot + domain = domain[1:] + cookie["domain"] = domain + + if hostname and not self._is_domain_match(domain, hostname): + # Setting cookies for different domains is not allowed + continue + + path = cookie["path"] + if not path or not path.startswith("/"): + # Set the cookie's path to the response path + path = response_url.path + if not path.startswith("/"): + path = "/" + else: + # Cut everything from the last slash to the end + path = "/" + path[1 : path.rfind("/")] + cookie["path"] = path + + max_age = cookie["max-age"] + if max_age: + try: + delta_seconds = int(max_age) + try: + max_age_expiration = datetime.datetime.now( + datetime.timezone.utc + ) + datetime.timedelta(seconds=delta_seconds) + except OverflowError: + max_age_expiration = self._max_time + self._expire_cookie(max_age_expiration, domain, name) + except ValueError: + cookie["max-age"] = "" + + else: + expires = cookie["expires"] + if expires: + expire_time = self._parse_date(expires) + if expire_time: + self._expire_cookie(expire_time, domain, name) + else: + cookie["expires"] = "" + + self._cookies[domain][name] = cookie + + self._do_expiration() + + def filter_cookies( + self, request_url: URL = URL() + ) -> Union["BaseCookie[str]", "SimpleCookie[str]"]: + """Returns this jar's cookies filtered by their attributes.""" + self._do_expiration() + request_url = URL(request_url) + filtered: Union["SimpleCookie[str]", "BaseCookie[str]"] = ( + SimpleCookie() if self._quote_cookie else BaseCookie() + ) + hostname = request_url.raw_host or "" + request_origin = URL() + with contextlib.suppress(ValueError): + request_origin = request_url.origin() + + is_not_secure = ( + request_url.scheme not in ("https", "wss") + and request_origin not in self._treat_as_secure_origin + ) + + for cookie in self: + name = cookie.key + domain = cookie["domain"] + + # Send shared cookies + if not domain: + filtered[name] = cookie.value + continue + + if not self._unsafe and is_ip_address(hostname): + continue + + if (domain, name) in self._host_only_cookies: + if domain != hostname: + continue + elif not self._is_domain_match(domain, hostname): + continue + + if not self._is_path_match(request_url.path, cookie["path"]): + continue + + if is_not_secure and cookie["secure"]: + continue + + # It's critical we use the Morsel so the coded_value + # (based on cookie version) is preserved + mrsl_val = cast("Morsel[str]", cookie.get(cookie.key, Morsel())) + mrsl_val.set(cookie.key, cookie.value, cookie.coded_value) + filtered[name] = mrsl_val + + return filtered + + @staticmethod + def _is_domain_match(domain: str, hostname: str) -> bool: + """Implements domain matching adhering to RFC 6265.""" + if hostname == domain: + return True + + if not hostname.endswith(domain): + return False + + non_matching = hostname[: -len(domain)] + + if not non_matching.endswith("."): + return False + + return not is_ip_address(hostname) + + @staticmethod + def _is_path_match(req_path: str, cookie_path: str) -> bool: + """Implements path matching adhering to RFC 6265.""" + if not req_path.startswith("/"): + req_path = "/" + + if req_path == cookie_path: + return True + + if not req_path.startswith(cookie_path): + return False + + if cookie_path.endswith("/"): + return True + + non_matching = req_path[len(cookie_path) :] + + return non_matching.startswith("/") + + @classmethod + def _parse_date(cls, date_str: str) -> Optional[datetime.datetime]: + """Implements date string parsing adhering to RFC 6265.""" + if not date_str: + return None + + found_time = False + found_day = False + found_month = False + found_year = False + + hour = minute = second = 0 + day = 0 + month = 0 + year = 0 + + for token_match in cls.DATE_TOKENS_RE.finditer(date_str): + + token = token_match.group("token") + + if not found_time: + time_match = cls.DATE_HMS_TIME_RE.match(token) + if time_match: + found_time = True + hour, minute, second = (int(s) for s in time_match.groups()) + continue + + if not found_day: + day_match = cls.DATE_DAY_OF_MONTH_RE.match(token) + if day_match: + found_day = True + day = int(day_match.group()) + continue + + if not found_month: + month_match = cls.DATE_MONTH_RE.match(token) + if month_match: + found_month = True + assert month_match.lastindex is not None + month = month_match.lastindex + continue + + if not found_year: + year_match = cls.DATE_YEAR_RE.match(token) + if year_match: + found_year = True + year = int(year_match.group()) + + if 70 <= year <= 99: + year += 1900 + elif 0 <= year <= 69: + year += 2000 + + if False in (found_day, found_month, found_year, found_time): + return None + + if not 1 <= day <= 31: + return None + + if year < 1601 or hour > 23 or minute > 59 or second > 59: + return None + + return datetime.datetime( + year, month, day, hour, minute, second, tzinfo=datetime.timezone.utc + ) + + +class DummyCookieJar(AbstractCookieJar): + """Implements a dummy cookie storage. + + It can be used with the ClientSession when no cookie processing is needed. + + """ + + def __init__(self, *, loop: Optional[asyncio.AbstractEventLoop] = None) -> None: + super().__init__(loop=loop) + + def __iter__(self) -> "Iterator[Morsel[str]]": + while False: + yield None + + def __len__(self) -> int: + return 0 + + def clear(self, predicate: Optional[ClearCookiePredicate] = None) -> None: + pass + + def clear_domain(self, domain: str) -> None: + pass + + def update_cookies(self, cookies: LooseCookies, response_url: URL = URL()) -> None: + pass + + def filter_cookies(self, request_url: URL) -> "BaseCookie[str]": + return SimpleCookie() diff --git a/contrib/python/aiohttp/aiohttp/formdata.py b/contrib/python/aiohttp/aiohttp/formdata.py new file mode 100644 index 0000000000..4857c89856 --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/formdata.py @@ -0,0 +1,172 @@ +import io +from typing import Any, Iterable, List, Optional +from urllib.parse import urlencode + +from multidict import MultiDict, MultiDictProxy + +from . import hdrs, multipart, payload +from .helpers import guess_filename +from .payload import Payload + +__all__ = ("FormData",) + + +class FormData: + """Helper class for form body generation. + + Supports multipart/form-data and application/x-www-form-urlencoded. + """ + + def __init__( + self, + fields: Iterable[Any] = (), + quote_fields: bool = True, + charset: Optional[str] = None, + ) -> None: + self._writer = multipart.MultipartWriter("form-data") + self._fields = [] # type: List[Any] + self._is_multipart = False + self._is_processed = False + self._quote_fields = quote_fields + self._charset = charset + + if isinstance(fields, dict): + fields = list(fields.items()) + elif not isinstance(fields, (list, tuple)): + fields = (fields,) + self.add_fields(*fields) + + @property + def is_multipart(self) -> bool: + return self._is_multipart + + def add_field( + self, + name: str, + value: Any, + *, + content_type: Optional[str] = None, + filename: Optional[str] = None, + content_transfer_encoding: Optional[str] = None, + ) -> None: + + if isinstance(value, io.IOBase): + self._is_multipart = True + elif isinstance(value, (bytes, bytearray, memoryview)): + if filename is None and content_transfer_encoding is None: + filename = name + + type_options = MultiDict({"name": name}) # type: MultiDict[str] + if filename is not None and not isinstance(filename, str): + raise TypeError( + "filename must be an instance of str. " "Got: %s" % filename + ) + if filename is None and isinstance(value, io.IOBase): + filename = guess_filename(value, name) + if filename is not None: + type_options["filename"] = filename + self._is_multipart = True + + headers = {} + if content_type is not None: + if not isinstance(content_type, str): + raise TypeError( + "content_type must be an instance of str. " "Got: %s" % content_type + ) + headers[hdrs.CONTENT_TYPE] = content_type + self._is_multipart = True + if content_transfer_encoding is not None: + if not isinstance(content_transfer_encoding, str): + raise TypeError( + "content_transfer_encoding must be an instance" + " of str. Got: %s" % content_transfer_encoding + ) + headers[hdrs.CONTENT_TRANSFER_ENCODING] = content_transfer_encoding + self._is_multipart = True + + self._fields.append((type_options, headers, value)) + + def add_fields(self, *fields: Any) -> None: + to_add = list(fields) + + while to_add: + rec = to_add.pop(0) + + if isinstance(rec, io.IOBase): + k = guess_filename(rec, "unknown") + self.add_field(k, rec) # type: ignore[arg-type] + + elif isinstance(rec, (MultiDictProxy, MultiDict)): + to_add.extend(rec.items()) + + elif isinstance(rec, (list, tuple)) and len(rec) == 2: + k, fp = rec + self.add_field(k, fp) # type: ignore[arg-type] + + else: + raise TypeError( + "Only io.IOBase, multidict and (name, file) " + "pairs allowed, use .add_field() for passing " + "more complex parameters, got {!r}".format(rec) + ) + + def _gen_form_urlencoded(self) -> payload.BytesPayload: + # form data (x-www-form-urlencoded) + data = [] + for type_options, _, value in self._fields: + data.append((type_options["name"], value)) + + charset = self._charset if self._charset is not None else "utf-8" + + if charset == "utf-8": + content_type = "application/x-www-form-urlencoded" + else: + content_type = "application/x-www-form-urlencoded; " "charset=%s" % charset + + return payload.BytesPayload( + urlencode(data, doseq=True, encoding=charset).encode(), + content_type=content_type, + ) + + def _gen_form_data(self) -> multipart.MultipartWriter: + """Encode a list of fields using the multipart/form-data MIME format""" + if self._is_processed: + raise RuntimeError("Form data has been processed already") + for dispparams, headers, value in self._fields: + try: + if hdrs.CONTENT_TYPE in headers: + part = payload.get_payload( + value, + content_type=headers[hdrs.CONTENT_TYPE], + headers=headers, + encoding=self._charset, + ) + else: + part = payload.get_payload( + value, headers=headers, encoding=self._charset + ) + except Exception as exc: + raise TypeError( + "Can not serialize value type: %r\n " + "headers: %r\n value: %r" % (type(value), headers, value) + ) from exc + + if dispparams: + part.set_content_disposition( + "form-data", quote_fields=self._quote_fields, **dispparams + ) + # FIXME cgi.FieldStorage doesn't likes body parts with + # Content-Length which were sent via chunked transfer encoding + assert part.headers is not None + part.headers.popall(hdrs.CONTENT_LENGTH, None) + + self._writer.append_payload(part) + + self._is_processed = True + return self._writer + + def __call__(self) -> Payload: + if self._is_multipart: + return self._gen_form_data() + else: + return self._gen_form_urlencoded() diff --git a/contrib/python/aiohttp/aiohttp/hdrs.py b/contrib/python/aiohttp/aiohttp/hdrs.py new file mode 100644 index 0000000000..a619f2543e --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/hdrs.py @@ -0,0 +1,114 @@ +"""HTTP Headers constants.""" + +# After changing the file content call ./tools/gen.py +# to regenerate the headers parser +import sys +from typing import Set + +from multidict import istr + +if sys.version_info >= (3, 8): + from typing import Final +else: + from typing_extensions import Final + +METH_ANY: Final[str] = "*" +METH_CONNECT: Final[str] = "CONNECT" +METH_HEAD: Final[str] = "HEAD" +METH_GET: Final[str] = "GET" +METH_DELETE: Final[str] = "DELETE" +METH_OPTIONS: Final[str] = "OPTIONS" +METH_PATCH: Final[str] = "PATCH" +METH_POST: Final[str] = "POST" +METH_PUT: Final[str] = "PUT" +METH_TRACE: Final[str] = "TRACE" + +METH_ALL: Final[Set[str]] = { + METH_CONNECT, + METH_HEAD, + METH_GET, + METH_DELETE, + METH_OPTIONS, + METH_PATCH, + METH_POST, + METH_PUT, + METH_TRACE, +} + +ACCEPT: Final[istr] = istr("Accept") +ACCEPT_CHARSET: Final[istr] = istr("Accept-Charset") +ACCEPT_ENCODING: Final[istr] = istr("Accept-Encoding") +ACCEPT_LANGUAGE: Final[istr] = istr("Accept-Language") +ACCEPT_RANGES: Final[istr] = istr("Accept-Ranges") +ACCESS_CONTROL_MAX_AGE: Final[istr] = istr("Access-Control-Max-Age") +ACCESS_CONTROL_ALLOW_CREDENTIALS: Final[istr] = istr("Access-Control-Allow-Credentials") +ACCESS_CONTROL_ALLOW_HEADERS: Final[istr] = istr("Access-Control-Allow-Headers") +ACCESS_CONTROL_ALLOW_METHODS: Final[istr] = istr("Access-Control-Allow-Methods") +ACCESS_CONTROL_ALLOW_ORIGIN: Final[istr] = istr("Access-Control-Allow-Origin") +ACCESS_CONTROL_EXPOSE_HEADERS: Final[istr] = istr("Access-Control-Expose-Headers") +ACCESS_CONTROL_REQUEST_HEADERS: Final[istr] = istr("Access-Control-Request-Headers") +ACCESS_CONTROL_REQUEST_METHOD: Final[istr] = istr("Access-Control-Request-Method") +AGE: Final[istr] = istr("Age") +ALLOW: Final[istr] = istr("Allow") +AUTHORIZATION: Final[istr] = istr("Authorization") +CACHE_CONTROL: Final[istr] = istr("Cache-Control") +CONNECTION: Final[istr] = istr("Connection") +CONTENT_DISPOSITION: Final[istr] = istr("Content-Disposition") +CONTENT_ENCODING: Final[istr] = istr("Content-Encoding") +CONTENT_LANGUAGE: Final[istr] = istr("Content-Language") +CONTENT_LENGTH: Final[istr] = istr("Content-Length") +CONTENT_LOCATION: Final[istr] = istr("Content-Location") +CONTENT_MD5: Final[istr] = istr("Content-MD5") +CONTENT_RANGE: Final[istr] = istr("Content-Range") +CONTENT_TRANSFER_ENCODING: Final[istr] = istr("Content-Transfer-Encoding") +CONTENT_TYPE: Final[istr] = istr("Content-Type") +COOKIE: Final[istr] = istr("Cookie") +DATE: Final[istr] = istr("Date") +DESTINATION: Final[istr] = istr("Destination") +DIGEST: Final[istr] = istr("Digest") +ETAG: Final[istr] = istr("Etag") +EXPECT: Final[istr] = istr("Expect") +EXPIRES: Final[istr] = istr("Expires") +FORWARDED: Final[istr] = istr("Forwarded") +FROM: Final[istr] = istr("From") +HOST: Final[istr] = istr("Host") +IF_MATCH: Final[istr] = istr("If-Match") +IF_MODIFIED_SINCE: Final[istr] = istr("If-Modified-Since") +IF_NONE_MATCH: Final[istr] = istr("If-None-Match") +IF_RANGE: Final[istr] = istr("If-Range") +IF_UNMODIFIED_SINCE: Final[istr] = istr("If-Unmodified-Since") +KEEP_ALIVE: Final[istr] = istr("Keep-Alive") +LAST_EVENT_ID: Final[istr] = istr("Last-Event-ID") +LAST_MODIFIED: Final[istr] = istr("Last-Modified") +LINK: Final[istr] = istr("Link") +LOCATION: Final[istr] = istr("Location") +MAX_FORWARDS: Final[istr] = istr("Max-Forwards") +ORIGIN: Final[istr] = istr("Origin") +PRAGMA: Final[istr] = istr("Pragma") +PROXY_AUTHENTICATE: Final[istr] = istr("Proxy-Authenticate") +PROXY_AUTHORIZATION: Final[istr] = istr("Proxy-Authorization") +RANGE: Final[istr] = istr("Range") +REFERER: Final[istr] = istr("Referer") +RETRY_AFTER: Final[istr] = istr("Retry-After") +SEC_WEBSOCKET_ACCEPT: Final[istr] = istr("Sec-WebSocket-Accept") +SEC_WEBSOCKET_VERSION: Final[istr] = istr("Sec-WebSocket-Version") +SEC_WEBSOCKET_PROTOCOL: Final[istr] = istr("Sec-WebSocket-Protocol") +SEC_WEBSOCKET_EXTENSIONS: Final[istr] = istr("Sec-WebSocket-Extensions") +SEC_WEBSOCKET_KEY: Final[istr] = istr("Sec-WebSocket-Key") +SEC_WEBSOCKET_KEY1: Final[istr] = istr("Sec-WebSocket-Key1") +SERVER: Final[istr] = istr("Server") +SET_COOKIE: Final[istr] = istr("Set-Cookie") +TE: Final[istr] = istr("TE") +TRAILER: Final[istr] = istr("Trailer") +TRANSFER_ENCODING: Final[istr] = istr("Transfer-Encoding") +UPGRADE: Final[istr] = istr("Upgrade") +URI: Final[istr] = istr("URI") +USER_AGENT: Final[istr] = istr("User-Agent") +VARY: Final[istr] = istr("Vary") +VIA: Final[istr] = istr("Via") +WANT_DIGEST: Final[istr] = istr("Want-Digest") +WARNING: Final[istr] = istr("Warning") +WWW_AUTHENTICATE: Final[istr] = istr("WWW-Authenticate") +X_FORWARDED_FOR: Final[istr] = istr("X-Forwarded-For") +X_FORWARDED_HOST: Final[istr] = istr("X-Forwarded-Host") +X_FORWARDED_PROTO: Final[istr] = istr("X-Forwarded-Proto") diff --git a/contrib/python/aiohttp/aiohttp/helpers.py b/contrib/python/aiohttp/aiohttp/helpers.py new file mode 100644 index 0000000000..f30f76ba41 --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/helpers.py @@ -0,0 +1,878 @@ +"""Various helper functions""" + +import asyncio +import base64 +import binascii +import datetime +import functools +import inspect +import netrc +import os +import platform +import re +import sys +import time +import warnings +import weakref +from collections import namedtuple +from contextlib import suppress +from email.parser import HeaderParser +from email.utils import parsedate +from math import ceil +from pathlib import Path +from types import TracebackType +from typing import ( + Any, + Callable, + ContextManager, + Dict, + Generator, + Generic, + Iterable, + Iterator, + List, + Mapping, + Optional, + Pattern, + Set, + Tuple, + Type, + TypeVar, + Union, + cast, +) +from urllib.parse import quote +from urllib.request import getproxies, proxy_bypass + +import async_timeout +import attr +from multidict import MultiDict, MultiDictProxy +from yarl import URL + +from . import hdrs +from .log import client_logger, internal_logger +from .typedefs import PathLike, Protocol # noqa + +__all__ = ("BasicAuth", "ChainMapProxy", "ETag") + +IS_MACOS = platform.system() == "Darwin" +IS_WINDOWS = platform.system() == "Windows" + +PY_36 = sys.version_info >= (3, 6) +PY_37 = sys.version_info >= (3, 7) +PY_38 = sys.version_info >= (3, 8) +PY_310 = sys.version_info >= (3, 10) + +if sys.version_info < (3, 7): + import idna_ssl + + idna_ssl.patch_match_hostname() + + def all_tasks( + loop: Optional[asyncio.AbstractEventLoop] = None, + ) -> Set["asyncio.Task[Any]"]: + tasks = list(asyncio.Task.all_tasks(loop)) + return {t for t in tasks if not t.done()} + + +else: + all_tasks = asyncio.all_tasks + + +_T = TypeVar("_T") +_S = TypeVar("_S") + + +sentinel = object() # type: Any +NO_EXTENSIONS = bool(os.environ.get("AIOHTTP_NO_EXTENSIONS")) # type: bool + +# N.B. sys.flags.dev_mode is available on Python 3.7+, use getattr +# for compatibility with older versions +DEBUG = getattr(sys.flags, "dev_mode", False) or ( + not sys.flags.ignore_environment and bool(os.environ.get("PYTHONASYNCIODEBUG")) +) # type: bool + + +CHAR = {chr(i) for i in range(0, 128)} +CTL = {chr(i) for i in range(0, 32)} | { + chr(127), +} +SEPARATORS = { + "(", + ")", + "<", + ">", + "@", + ",", + ";", + ":", + "\\", + '"', + "/", + "[", + "]", + "?", + "=", + "{", + "}", + " ", + chr(9), +} +TOKEN = CHAR ^ CTL ^ SEPARATORS + + +class noop: + def __await__(self) -> Generator[None, None, None]: + yield + + +class BasicAuth(namedtuple("BasicAuth", ["login", "password", "encoding"])): + """Http basic authentication helper.""" + + def __new__( + cls, login: str, password: str = "", encoding: str = "latin1" + ) -> "BasicAuth": + if login is None: + raise ValueError("None is not allowed as login value") + + if password is None: + raise ValueError("None is not allowed as password value") + + if ":" in login: + raise ValueError('A ":" is not allowed in login (RFC 1945#section-11.1)') + + return super().__new__(cls, login, password, encoding) + + @classmethod + def decode(cls, auth_header: str, encoding: str = "latin1") -> "BasicAuth": + """Create a BasicAuth object from an Authorization HTTP header.""" + try: + auth_type, encoded_credentials = auth_header.split(" ", 1) + except ValueError: + raise ValueError("Could not parse authorization header.") + + if auth_type.lower() != "basic": + raise ValueError("Unknown authorization method %s" % auth_type) + + try: + decoded = base64.b64decode( + encoded_credentials.encode("ascii"), validate=True + ).decode(encoding) + except binascii.Error: + raise ValueError("Invalid base64 encoding.") + + try: + # RFC 2617 HTTP Authentication + # https://www.ietf.org/rfc/rfc2617.txt + # the colon must be present, but the username and password may be + # otherwise blank. + username, password = decoded.split(":", 1) + except ValueError: + raise ValueError("Invalid credentials.") + + return cls(username, password, encoding=encoding) + + @classmethod + def from_url(cls, url: URL, *, encoding: str = "latin1") -> Optional["BasicAuth"]: + """Create BasicAuth from url.""" + if not isinstance(url, URL): + raise TypeError("url should be yarl.URL instance") + if url.user is None: + return None + return cls(url.user, url.password or "", encoding=encoding) + + def encode(self) -> str: + """Encode credentials.""" + creds = (f"{self.login}:{self.password}").encode(self.encoding) + return "Basic %s" % base64.b64encode(creds).decode(self.encoding) + + +def strip_auth_from_url(url: URL) -> Tuple[URL, Optional[BasicAuth]]: + auth = BasicAuth.from_url(url) + if auth is None: + return url, None + else: + return url.with_user(None), auth + + +def netrc_from_env() -> Optional[netrc.netrc]: + """Load netrc from file. + + Attempt to load it from the path specified by the env-var + NETRC or in the default location in the user's home directory. + + Returns None if it couldn't be found or fails to parse. + """ + netrc_env = os.environ.get("NETRC") + + if netrc_env is not None: + netrc_path = Path(netrc_env) + else: + try: + home_dir = Path.home() + except RuntimeError as e: # pragma: no cover + # if pathlib can't resolve home, it may raise a RuntimeError + client_logger.debug( + "Could not resolve home directory when " + "trying to look for .netrc file: %s", + e, + ) + return None + + netrc_path = home_dir / ("_netrc" if IS_WINDOWS else ".netrc") + + try: + return netrc.netrc(str(netrc_path)) + except netrc.NetrcParseError as e: + client_logger.warning("Could not parse .netrc file: %s", e) + except OSError as e: + # we couldn't read the file (doesn't exist, permissions, etc.) + if netrc_env or netrc_path.is_file(): + # only warn if the environment wanted us to load it, + # or it appears like the default file does actually exist + client_logger.warning("Could not read .netrc file: %s", e) + + return None + + +@attr.s(auto_attribs=True, frozen=True, slots=True) +class ProxyInfo: + proxy: URL + proxy_auth: Optional[BasicAuth] + + +def proxies_from_env() -> Dict[str, ProxyInfo]: + proxy_urls = { + k: URL(v) + for k, v in getproxies().items() + if k in ("http", "https", "ws", "wss") + } + netrc_obj = netrc_from_env() + stripped = {k: strip_auth_from_url(v) for k, v in proxy_urls.items()} + ret = {} + for proto, val in stripped.items(): + proxy, auth = val + if proxy.scheme in ("https", "wss"): + client_logger.warning( + "%s proxies %s are not supported, ignoring", proxy.scheme.upper(), proxy + ) + continue + if netrc_obj and auth is None: + auth_from_netrc = None + if proxy.host is not None: + auth_from_netrc = netrc_obj.authenticators(proxy.host) + if auth_from_netrc is not None: + # auth_from_netrc is a (`user`, `account`, `password`) tuple, + # `user` and `account` both can be username, + # if `user` is None, use `account` + *logins, password = auth_from_netrc + login = logins[0] if logins[0] else logins[-1] + auth = BasicAuth(cast(str, login), cast(str, password)) + ret[proto] = ProxyInfo(proxy, auth) + return ret + + +def current_task( + loop: Optional[asyncio.AbstractEventLoop] = None, +) -> "Optional[asyncio.Task[Any]]": + if sys.version_info >= (3, 7): + return asyncio.current_task(loop=loop) + else: + return asyncio.Task.current_task(loop=loop) + + +def get_running_loop( + loop: Optional[asyncio.AbstractEventLoop] = None, +) -> asyncio.AbstractEventLoop: + if loop is None: + loop = asyncio.get_event_loop() + if not loop.is_running(): + warnings.warn( + "The object should be created within an async function", + DeprecationWarning, + stacklevel=3, + ) + if loop.get_debug(): + internal_logger.warning( + "The object should be created within an async function", stack_info=True + ) + return loop + + +def isasyncgenfunction(obj: Any) -> bool: + func = getattr(inspect, "isasyncgenfunction", None) + if func is not None: + return func(obj) # type: ignore[no-any-return] + else: + return False + + +def get_env_proxy_for_url(url: URL) -> Tuple[URL, Optional[BasicAuth]]: + """Get a permitted proxy for the given URL from the env.""" + if url.host is not None and proxy_bypass(url.host): + raise LookupError(f"Proxying is disallowed for `{url.host!r}`") + + proxies_in_env = proxies_from_env() + try: + proxy_info = proxies_in_env[url.scheme] + except KeyError: + raise LookupError(f"No proxies found for `{url!s}` in the env") + else: + return proxy_info.proxy, proxy_info.proxy_auth + + +@attr.s(auto_attribs=True, frozen=True, slots=True) +class MimeType: + type: str + subtype: str + suffix: str + parameters: "MultiDictProxy[str]" + + +@functools.lru_cache(maxsize=56) +def parse_mimetype(mimetype: str) -> MimeType: + """Parses a MIME type into its components. + + mimetype is a MIME type string. + + Returns a MimeType object. + + Example: + + >>> parse_mimetype('text/html; charset=utf-8') + MimeType(type='text', subtype='html', suffix='', + parameters={'charset': 'utf-8'}) + + """ + if not mimetype: + return MimeType( + type="", subtype="", suffix="", parameters=MultiDictProxy(MultiDict()) + ) + + parts = mimetype.split(";") + params = MultiDict() # type: MultiDict[str] + for item in parts[1:]: + if not item: + continue + key, value = cast( + Tuple[str, str], item.split("=", 1) if "=" in item else (item, "") + ) + params.add(key.lower().strip(), value.strip(' "')) + + fulltype = parts[0].strip().lower() + if fulltype == "*": + fulltype = "*/*" + + mtype, stype = ( + cast(Tuple[str, str], fulltype.split("/", 1)) + if "/" in fulltype + else (fulltype, "") + ) + stype, suffix = ( + cast(Tuple[str, str], stype.split("+", 1)) if "+" in stype else (stype, "") + ) + + return MimeType( + type=mtype, subtype=stype, suffix=suffix, parameters=MultiDictProxy(params) + ) + + +def guess_filename(obj: Any, default: Optional[str] = None) -> Optional[str]: + name = getattr(obj, "name", None) + if name and isinstance(name, str) and name[0] != "<" and name[-1] != ">": + return Path(name).name + return default + + +not_qtext_re = re.compile(r"[^\041\043-\133\135-\176]") +QCONTENT = {chr(i) for i in range(0x20, 0x7F)} | {"\t"} + + +def quoted_string(content: str) -> str: + """Return 7-bit content as quoted-string. + + Format content into a quoted-string as defined in RFC5322 for + Internet Message Format. Notice that this is not the 8-bit HTTP + format, but the 7-bit email format. Content must be in usascii or + a ValueError is raised. + """ + if not (QCONTENT > set(content)): + raise ValueError(f"bad content for quoted-string {content!r}") + return not_qtext_re.sub(lambda x: "\\" + x.group(0), content) + + +def content_disposition_header( + disptype: str, quote_fields: bool = True, _charset: str = "utf-8", **params: str +) -> str: + """Sets ``Content-Disposition`` header for MIME. + + This is the MIME payload Content-Disposition header from RFC 2183 + and RFC 7579 section 4.2, not the HTTP Content-Disposition from + RFC 6266. + + disptype is a disposition type: inline, attachment, form-data. + Should be valid extension token (see RFC 2183) + + quote_fields performs value quoting to 7-bit MIME headers + according to RFC 7578. Set to quote_fields to False if recipient + can take 8-bit file names and field values. + + _charset specifies the charset to use when quote_fields is True. + + params is a dict with disposition params. + """ + if not disptype or not (TOKEN > set(disptype)): + raise ValueError("bad content disposition type {!r}" "".format(disptype)) + + value = disptype + if params: + lparams = [] + for key, val in params.items(): + if not key or not (TOKEN > set(key)): + raise ValueError( + "bad content disposition parameter" " {!r}={!r}".format(key, val) + ) + if quote_fields: + if key.lower() == "filename": + qval = quote(val, "", encoding=_charset) + lparams.append((key, '"%s"' % qval)) + else: + try: + qval = quoted_string(val) + except ValueError: + qval = "".join( + (_charset, "''", quote(val, "", encoding=_charset)) + ) + lparams.append((key + "*", qval)) + else: + lparams.append((key, '"%s"' % qval)) + else: + qval = val.replace("\\", "\\\\").replace('"', '\\"') + lparams.append((key, '"%s"' % qval)) + sparams = "; ".join("=".join(pair) for pair in lparams) + value = "; ".join((value, sparams)) + return value + + +class _TSelf(Protocol, Generic[_T]): + _cache: Dict[str, _T] + + +class reify(Generic[_T]): + """Use as a class method decorator. + + It operates almost exactly like + the Python `@property` decorator, but it puts the result of the + method it decorates into the instance dict after the first call, + effectively replacing the function it decorates with an instance + variable. It is, in Python parlance, a data descriptor. + """ + + def __init__(self, wrapped: Callable[..., _T]) -> None: + self.wrapped = wrapped + self.__doc__ = wrapped.__doc__ + self.name = wrapped.__name__ + + def __get__(self, inst: _TSelf[_T], owner: Optional[Type[Any]] = None) -> _T: + try: + try: + return inst._cache[self.name] + except KeyError: + val = self.wrapped(inst) + inst._cache[self.name] = val + return val + except AttributeError: + if inst is None: + return self + raise + + def __set__(self, inst: _TSelf[_T], value: _T) -> None: + raise AttributeError("reified property is read-only") + + +reify_py = reify + +try: + from ._helpers import reify as reify_c + + if not NO_EXTENSIONS: + reify = reify_c # type: ignore[misc,assignment] +except ImportError: + pass + +_ipv4_pattern = ( + r"^(?:(?:25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)\.){3}" + r"(?:25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)$" +) +_ipv6_pattern = ( + r"^(?:(?:(?:[A-F0-9]{1,4}:){6}|(?=(?:[A-F0-9]{0,4}:){0,6}" + r"(?:[0-9]{1,3}\.){3}[0-9]{1,3}$)(([0-9A-F]{1,4}:){0,5}|:)" + r"((:[0-9A-F]{1,4}){1,5}:|:)|::(?:[A-F0-9]{1,4}:){5})" + r"(?:(?:25[0-5]|2[0-4][0-9]|1[0-9][0-9]|[1-9]?[0-9])\.){3}" + r"(?:25[0-5]|2[0-4][0-9]|1[0-9][0-9]|[1-9]?[0-9])|(?:[A-F0-9]{1,4}:){7}" + r"[A-F0-9]{1,4}|(?=(?:[A-F0-9]{0,4}:){0,7}[A-F0-9]{0,4}$)" + r"(([0-9A-F]{1,4}:){1,7}|:)((:[0-9A-F]{1,4}){1,7}|:)|(?:[A-F0-9]{1,4}:){7}" + r":|:(:[A-F0-9]{1,4}){7})$" +) +_ipv4_regex = re.compile(_ipv4_pattern) +_ipv6_regex = re.compile(_ipv6_pattern, flags=re.IGNORECASE) +_ipv4_regexb = re.compile(_ipv4_pattern.encode("ascii")) +_ipv6_regexb = re.compile(_ipv6_pattern.encode("ascii"), flags=re.IGNORECASE) + + +def _is_ip_address( + regex: Pattern[str], regexb: Pattern[bytes], host: Optional[Union[str, bytes]] +) -> bool: + if host is None: + return False + if isinstance(host, str): + return bool(regex.match(host)) + elif isinstance(host, (bytes, bytearray, memoryview)): + return bool(regexb.match(host)) + else: + raise TypeError(f"{host} [{type(host)}] is not a str or bytes") + + +is_ipv4_address = functools.partial(_is_ip_address, _ipv4_regex, _ipv4_regexb) +is_ipv6_address = functools.partial(_is_ip_address, _ipv6_regex, _ipv6_regexb) + + +def is_ip_address(host: Optional[Union[str, bytes, bytearray, memoryview]]) -> bool: + return is_ipv4_address(host) or is_ipv6_address(host) + + +def next_whole_second() -> datetime.datetime: + """Return current time rounded up to the next whole second.""" + return datetime.datetime.now(datetime.timezone.utc).replace( + microsecond=0 + ) + datetime.timedelta(seconds=0) + + +_cached_current_datetime = None # type: Optional[int] +_cached_formatted_datetime = "" + + +def rfc822_formatted_time() -> str: + global _cached_current_datetime + global _cached_formatted_datetime + + now = int(time.time()) + if now != _cached_current_datetime: + # Weekday and month names for HTTP date/time formatting; + # always English! + # Tuples are constants stored in codeobject! + _weekdayname = ("Mon", "Tue", "Wed", "Thu", "Fri", "Sat", "Sun") + _monthname = ( + "", # Dummy so we can use 1-based month numbers + "Jan", + "Feb", + "Mar", + "Apr", + "May", + "Jun", + "Jul", + "Aug", + "Sep", + "Oct", + "Nov", + "Dec", + ) + + year, month, day, hh, mm, ss, wd, *tail = time.gmtime(now) + _cached_formatted_datetime = "%s, %02d %3s %4d %02d:%02d:%02d GMT" % ( + _weekdayname[wd], + day, + _monthname[month], + year, + hh, + mm, + ss, + ) + _cached_current_datetime = now + return _cached_formatted_datetime + + +def _weakref_handle(info: "Tuple[weakref.ref[object], str]") -> None: + ref, name = info + ob = ref() + if ob is not None: + with suppress(Exception): + getattr(ob, name)() + + +def weakref_handle( + ob: object, name: str, timeout: float, loop: asyncio.AbstractEventLoop +) -> Optional[asyncio.TimerHandle]: + if timeout is not None and timeout > 0: + when = loop.time() + timeout + if timeout >= 5: + when = ceil(when) + + return loop.call_at(when, _weakref_handle, (weakref.ref(ob), name)) + return None + + +def call_later( + cb: Callable[[], Any], timeout: float, loop: asyncio.AbstractEventLoop +) -> Optional[asyncio.TimerHandle]: + if timeout is not None and timeout > 0: + when = loop.time() + timeout + if timeout > 5: + when = ceil(when) + return loop.call_at(when, cb) + return None + + +class TimeoutHandle: + """Timeout handle""" + + def __init__( + self, loop: asyncio.AbstractEventLoop, timeout: Optional[float] + ) -> None: + self._timeout = timeout + self._loop = loop + self._callbacks = ( + [] + ) # type: List[Tuple[Callable[..., None], Tuple[Any, ...], Dict[str, Any]]] + + def register( + self, callback: Callable[..., None], *args: Any, **kwargs: Any + ) -> None: + self._callbacks.append((callback, args, kwargs)) + + def close(self) -> None: + self._callbacks.clear() + + def start(self) -> Optional[asyncio.Handle]: + timeout = self._timeout + if timeout is not None and timeout > 0: + when = self._loop.time() + timeout + if timeout >= 5: + when = ceil(when) + return self._loop.call_at(when, self.__call__) + else: + return None + + def timer(self) -> "BaseTimerContext": + if self._timeout is not None and self._timeout > 0: + timer = TimerContext(self._loop) + self.register(timer.timeout) + return timer + else: + return TimerNoop() + + def __call__(self) -> None: + for cb, args, kwargs in self._callbacks: + with suppress(Exception): + cb(*args, **kwargs) + + self._callbacks.clear() + + +class BaseTimerContext(ContextManager["BaseTimerContext"]): + pass + + +class TimerNoop(BaseTimerContext): + def __enter__(self) -> BaseTimerContext: + return self + + def __exit__( + self, + exc_type: Optional[Type[BaseException]], + exc_val: Optional[BaseException], + exc_tb: Optional[TracebackType], + ) -> None: + return + + +class TimerContext(BaseTimerContext): + """Low resolution timeout context manager""" + + def __init__(self, loop: asyncio.AbstractEventLoop) -> None: + self._loop = loop + self._tasks = [] # type: List[asyncio.Task[Any]] + self._cancelled = False + + def __enter__(self) -> BaseTimerContext: + task = current_task(loop=self._loop) + + if task is None: + raise RuntimeError( + "Timeout context manager should be used " "inside a task" + ) + + if self._cancelled: + raise asyncio.TimeoutError from None + + self._tasks.append(task) + return self + + def __exit__( + self, + exc_type: Optional[Type[BaseException]], + exc_val: Optional[BaseException], + exc_tb: Optional[TracebackType], + ) -> Optional[bool]: + if self._tasks: + self._tasks.pop() + + if exc_type is asyncio.CancelledError and self._cancelled: + raise asyncio.TimeoutError from None + return None + + def timeout(self) -> None: + if not self._cancelled: + for task in set(self._tasks): + task.cancel() + + self._cancelled = True + + +def ceil_timeout(delay: Optional[float]) -> async_timeout.Timeout: + if delay is None or delay <= 0: + return async_timeout.timeout(None) + + loop = get_running_loop() + now = loop.time() + when = now + delay + if delay > 5: + when = ceil(when) + return async_timeout.timeout_at(when) + + +class HeadersMixin: + + ATTRS = frozenset(["_content_type", "_content_dict", "_stored_content_type"]) + + _content_type = None # type: Optional[str] + _content_dict = None # type: Optional[Dict[str, str]] + _stored_content_type = sentinel + + def _parse_content_type(self, raw: str) -> None: + self._stored_content_type = raw + if raw is None: + # default value according to RFC 2616 + self._content_type = "application/octet-stream" + self._content_dict = {} + else: + msg = HeaderParser().parsestr("Content-Type: " + raw) + self._content_type = msg.get_content_type() + params = msg.get_params() + self._content_dict = dict(params[1:]) # First element is content type again + + @property + def content_type(self) -> str: + """The value of content part for Content-Type HTTP header.""" + raw = self._headers.get(hdrs.CONTENT_TYPE) # type: ignore[attr-defined] + if self._stored_content_type != raw: + self._parse_content_type(raw) + return self._content_type # type: ignore[return-value] + + @property + def charset(self) -> Optional[str]: + """The value of charset part for Content-Type HTTP header.""" + raw = self._headers.get(hdrs.CONTENT_TYPE) # type: ignore[attr-defined] + if self._stored_content_type != raw: + self._parse_content_type(raw) + return self._content_dict.get("charset") # type: ignore[union-attr] + + @property + def content_length(self) -> Optional[int]: + """The value of Content-Length HTTP header.""" + content_length = self._headers.get( # type: ignore[attr-defined] + hdrs.CONTENT_LENGTH + ) + + if content_length is not None: + return int(content_length) + else: + return None + + +def set_result(fut: "asyncio.Future[_T]", result: _T) -> None: + if not fut.done(): + fut.set_result(result) + + +def set_exception(fut: "asyncio.Future[_T]", exc: BaseException) -> None: + if not fut.done(): + fut.set_exception(exc) + + +class ChainMapProxy(Mapping[str, Any]): + __slots__ = ("_maps",) + + def __init__(self, maps: Iterable[Mapping[str, Any]]) -> None: + self._maps = tuple(maps) + + def __init_subclass__(cls) -> None: + raise TypeError( + "Inheritance class {} from ChainMapProxy " + "is forbidden".format(cls.__name__) + ) + + def __getitem__(self, key: str) -> Any: + for mapping in self._maps: + try: + return mapping[key] + except KeyError: + pass + raise KeyError(key) + + def get(self, key: str, default: Any = None) -> Any: + return self[key] if key in self else default + + def __len__(self) -> int: + # reuses stored hash values if possible + return len(set().union(*self._maps)) # type: ignore[arg-type] + + def __iter__(self) -> Iterator[str]: + d = {} # type: Dict[str, Any] + for mapping in reversed(self._maps): + # reuses stored hash values if possible + d.update(mapping) + return iter(d) + + def __contains__(self, key: object) -> bool: + return any(key in m for m in self._maps) + + def __bool__(self) -> bool: + return any(self._maps) + + def __repr__(self) -> str: + content = ", ".join(map(repr, self._maps)) + return f"ChainMapProxy({content})" + + +# https://tools.ietf.org/html/rfc7232#section-2.3 +_ETAGC = r"[!#-}\x80-\xff]+" +_ETAGC_RE = re.compile(_ETAGC) +_QUOTED_ETAG = fr'(W/)?"({_ETAGC})"' +QUOTED_ETAG_RE = re.compile(_QUOTED_ETAG) +LIST_QUOTED_ETAG_RE = re.compile(fr"({_QUOTED_ETAG})(?:\s*,\s*|$)|(.)") + +ETAG_ANY = "*" + + +@attr.s(auto_attribs=True, frozen=True, slots=True) +class ETag: + value: str + is_weak: bool = False + + +def validate_etag_value(value: str) -> None: + if value != ETAG_ANY and not _ETAGC_RE.fullmatch(value): + raise ValueError( + f"Value {value!r} is not a valid etag. Maybe it contains '\"'?" + ) + + +def parse_http_date(date_str: Optional[str]) -> Optional[datetime.datetime]: + """Process a date string, return a datetime object""" + if date_str is not None: + timetuple = parsedate(date_str) + if timetuple is not None: + with suppress(ValueError): + return datetime.datetime(*timetuple[:6], tzinfo=datetime.timezone.utc) + return None diff --git a/contrib/python/aiohttp/aiohttp/http.py b/contrib/python/aiohttp/aiohttp/http.py new file mode 100644 index 0000000000..415ffbf563 --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/http.py @@ -0,0 +1,72 @@ +import http.server +import sys +from typing import Mapping, Tuple + +from . import __version__ +from .http_exceptions import HttpProcessingError as HttpProcessingError +from .http_parser import ( + HeadersParser as HeadersParser, + HttpParser as HttpParser, + HttpRequestParser as HttpRequestParser, + HttpResponseParser as HttpResponseParser, + RawRequestMessage as RawRequestMessage, + RawResponseMessage as RawResponseMessage, +) +from .http_websocket import ( + WS_CLOSED_MESSAGE as WS_CLOSED_MESSAGE, + WS_CLOSING_MESSAGE as WS_CLOSING_MESSAGE, + WS_KEY as WS_KEY, + WebSocketError as WebSocketError, + WebSocketReader as WebSocketReader, + WebSocketWriter as WebSocketWriter, + WSCloseCode as WSCloseCode, + WSMessage as WSMessage, + WSMsgType as WSMsgType, + ws_ext_gen as ws_ext_gen, + ws_ext_parse as ws_ext_parse, +) +from .http_writer import ( + HttpVersion as HttpVersion, + HttpVersion10 as HttpVersion10, + HttpVersion11 as HttpVersion11, + StreamWriter as StreamWriter, +) + +__all__ = ( + "HttpProcessingError", + "RESPONSES", + "SERVER_SOFTWARE", + # .http_writer + "StreamWriter", + "HttpVersion", + "HttpVersion10", + "HttpVersion11", + # .http_parser + "HeadersParser", + "HttpParser", + "HttpRequestParser", + "HttpResponseParser", + "RawRequestMessage", + "RawResponseMessage", + # .http_websocket + "WS_CLOSED_MESSAGE", + "WS_CLOSING_MESSAGE", + "WS_KEY", + "WebSocketReader", + "WebSocketWriter", + "ws_ext_gen", + "ws_ext_parse", + "WSMessage", + "WebSocketError", + "WSMsgType", + "WSCloseCode", +) + + +SERVER_SOFTWARE = "Python/{0[0]}.{0[1]} aiohttp/{1}".format( + sys.version_info, __version__ +) # type: str + +RESPONSES = ( + http.server.BaseHTTPRequestHandler.responses +) # type: Mapping[int, Tuple[str, str]] diff --git a/contrib/python/aiohttp/aiohttp/http_exceptions.py b/contrib/python/aiohttp/aiohttp/http_exceptions.py new file mode 100644 index 0000000000..c885f80f32 --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/http_exceptions.py @@ -0,0 +1,105 @@ +"""Low-level http related exceptions.""" + + +from typing import Optional, Union + +from .typedefs import _CIMultiDict + +__all__ = ("HttpProcessingError",) + + +class HttpProcessingError(Exception): + """HTTP error. + + Shortcut for raising HTTP errors with custom code, message and headers. + + code: HTTP Error code. + message: (optional) Error message. + headers: (optional) Headers to be sent in response, a list of pairs + """ + + code = 0 + message = "" + headers = None + + def __init__( + self, + *, + code: Optional[int] = None, + message: str = "", + headers: Optional[_CIMultiDict] = None, + ) -> None: + if code is not None: + self.code = code + self.headers = headers + self.message = message + + def __str__(self) -> str: + return f"{self.code}, message={self.message!r}" + + def __repr__(self) -> str: + return f"<{self.__class__.__name__}: {self}>" + + +class BadHttpMessage(HttpProcessingError): + + code = 400 + message = "Bad Request" + + def __init__(self, message: str, *, headers: Optional[_CIMultiDict] = None) -> None: + super().__init__(message=message, headers=headers) + self.args = (message,) + + +class HttpBadRequest(BadHttpMessage): + + code = 400 + message = "Bad Request" + + +class PayloadEncodingError(BadHttpMessage): + """Base class for payload errors""" + + +class ContentEncodingError(PayloadEncodingError): + """Content encoding error.""" + + +class TransferEncodingError(PayloadEncodingError): + """transfer encoding error.""" + + +class ContentLengthError(PayloadEncodingError): + """Not enough data for satisfy content length header.""" + + +class LineTooLong(BadHttpMessage): + def __init__( + self, line: str, limit: str = "Unknown", actual_size: str = "Unknown" + ) -> None: + super().__init__( + f"Got more than {limit} bytes ({actual_size}) when reading {line}." + ) + self.args = (line, limit, actual_size) + + +class InvalidHeader(BadHttpMessage): + def __init__(self, hdr: Union[bytes, str]) -> None: + if isinstance(hdr, bytes): + hdr = hdr.decode("utf-8", "surrogateescape") + super().__init__(f"Invalid HTTP Header: {hdr}") + self.hdr = hdr + self.args = (hdr,) + + +class BadStatusLine(BadHttpMessage): + def __init__(self, line: str = "") -> None: + if not isinstance(line, str): + line = repr(line) + super().__init__(f"Bad status line {line!r}") + self.args = (line,) + self.line = line + + +class InvalidURLError(BadHttpMessage): + pass diff --git a/contrib/python/aiohttp/aiohttp/http_parser.py b/contrib/python/aiohttp/aiohttp/http_parser.py new file mode 100644 index 0000000000..2dc9482f4f --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/http_parser.py @@ -0,0 +1,956 @@ +import abc +import asyncio +import collections +import re +import string +import zlib +from contextlib import suppress +from enum import IntEnum +from typing import ( + Any, + Generic, + List, + NamedTuple, + Optional, + Pattern, + Set, + Tuple, + Type, + TypeVar, + Union, + cast, +) + +from multidict import CIMultiDict, CIMultiDictProxy, istr +from yarl import URL + +from . import hdrs +from .base_protocol import BaseProtocol +from .helpers import NO_EXTENSIONS, BaseTimerContext +from .http_exceptions import ( + BadHttpMessage, + BadStatusLine, + ContentEncodingError, + ContentLengthError, + InvalidHeader, + LineTooLong, + TransferEncodingError, +) +from .http_writer import HttpVersion, HttpVersion10 +from .log import internal_logger +from .streams import EMPTY_PAYLOAD, StreamReader +from .typedefs import Final, RawHeaders + +try: + import brotli + + HAS_BROTLI = True +except ImportError: # pragma: no cover + HAS_BROTLI = False + + +__all__ = ( + "HeadersParser", + "HttpParser", + "HttpRequestParser", + "HttpResponseParser", + "RawRequestMessage", + "RawResponseMessage", +) + +ASCIISET: Final[Set[str]] = set(string.printable) + +# See https://tools.ietf.org/html/rfc7230#section-3.1.1 +# and https://tools.ietf.org/html/rfc7230#appendix-B +# +# method = token +# tchar = "!" / "#" / "$" / "%" / "&" / "'" / "*" / "+" / "-" / "." / +# "^" / "_" / "`" / "|" / "~" / DIGIT / ALPHA +# token = 1*tchar +METHRE: Final[Pattern[str]] = re.compile(r"[!#$%&'*+\-.^_`|~0-9A-Za-z]+") +VERSRE: Final[Pattern[str]] = re.compile(r"HTTP/(\d+).(\d+)") +HDRRE: Final[Pattern[bytes]] = re.compile(rb"[\x00-\x1F\x7F()<>@,;:\[\]={} \t\\\\\"]") + + +class RawRequestMessage(NamedTuple): + method: str + path: str + version: HttpVersion + headers: "CIMultiDictProxy[str]" + raw_headers: RawHeaders + should_close: bool + compression: Optional[str] + upgrade: bool + chunked: bool + url: URL + + +RawResponseMessage = collections.namedtuple( + "RawResponseMessage", + [ + "version", + "code", + "reason", + "headers", + "raw_headers", + "should_close", + "compression", + "upgrade", + "chunked", + ], +) + + +_MsgT = TypeVar("_MsgT", RawRequestMessage, RawResponseMessage) + + +class ParseState(IntEnum): + + PARSE_NONE = 0 + PARSE_LENGTH = 1 + PARSE_CHUNKED = 2 + PARSE_UNTIL_EOF = 3 + + +class ChunkState(IntEnum): + PARSE_CHUNKED_SIZE = 0 + PARSE_CHUNKED_CHUNK = 1 + PARSE_CHUNKED_CHUNK_EOF = 2 + PARSE_MAYBE_TRAILERS = 3 + PARSE_TRAILERS = 4 + + +class HeadersParser: + def __init__( + self, + max_line_size: int = 8190, + max_headers: int = 32768, + max_field_size: int = 8190, + ) -> None: + self.max_line_size = max_line_size + self.max_headers = max_headers + self.max_field_size = max_field_size + + def parse_headers( + self, lines: List[bytes] + ) -> Tuple["CIMultiDictProxy[str]", RawHeaders]: + headers = CIMultiDict() # type: CIMultiDict[str] + raw_headers = [] + + lines_idx = 1 + line = lines[1] + line_count = len(lines) + + while line: + # Parse initial header name : value pair. + try: + bname, bvalue = line.split(b":", 1) + except ValueError: + raise InvalidHeader(line) from None + + bname = bname.strip(b" \t") + bvalue = bvalue.lstrip() + if HDRRE.search(bname): + raise InvalidHeader(bname) + if len(bname) > self.max_field_size: + raise LineTooLong( + "request header name {}".format( + bname.decode("utf8", "xmlcharrefreplace") + ), + str(self.max_field_size), + str(len(bname)), + ) + + header_length = len(bvalue) + + # next line + lines_idx += 1 + line = lines[lines_idx] + + # consume continuation lines + continuation = line and line[0] in (32, 9) # (' ', '\t') + + if continuation: + bvalue_lst = [bvalue] + while continuation: + header_length += len(line) + if header_length > self.max_field_size: + raise LineTooLong( + "request header field {}".format( + bname.decode("utf8", "xmlcharrefreplace") + ), + str(self.max_field_size), + str(header_length), + ) + bvalue_lst.append(line) + + # next line + lines_idx += 1 + if lines_idx < line_count: + line = lines[lines_idx] + if line: + continuation = line[0] in (32, 9) # (' ', '\t') + else: + line = b"" + break + bvalue = b"".join(bvalue_lst) + else: + if header_length > self.max_field_size: + raise LineTooLong( + "request header field {}".format( + bname.decode("utf8", "xmlcharrefreplace") + ), + str(self.max_field_size), + str(header_length), + ) + + bvalue = bvalue.strip() + name = bname.decode("utf-8", "surrogateescape") + value = bvalue.decode("utf-8", "surrogateescape") + + headers.add(name, value) + raw_headers.append((bname, bvalue)) + + return (CIMultiDictProxy(headers), tuple(raw_headers)) + + +class HttpParser(abc.ABC, Generic[_MsgT]): + def __init__( + self, + protocol: Optional[BaseProtocol] = None, + loop: Optional[asyncio.AbstractEventLoop] = None, + limit: int = 2 ** 16, + max_line_size: int = 8190, + max_headers: int = 32768, + max_field_size: int = 8190, + timer: Optional[BaseTimerContext] = None, + code: Optional[int] = None, + method: Optional[str] = None, + readall: bool = False, + payload_exception: Optional[Type[BaseException]] = None, + response_with_body: bool = True, + read_until_eof: bool = False, + auto_decompress: bool = True, + ) -> None: + self.protocol = protocol + self.loop = loop + self.max_line_size = max_line_size + self.max_headers = max_headers + self.max_field_size = max_field_size + self.timer = timer + self.code = code + self.method = method + self.readall = readall + self.payload_exception = payload_exception + self.response_with_body = response_with_body + self.read_until_eof = read_until_eof + + self._lines = [] # type: List[bytes] + self._tail = b"" + self._upgraded = False + self._payload = None + self._payload_parser = None # type: Optional[HttpPayloadParser] + self._auto_decompress = auto_decompress + self._limit = limit + self._headers_parser = HeadersParser(max_line_size, max_headers, max_field_size) + + @abc.abstractmethod + def parse_message(self, lines: List[bytes]) -> _MsgT: + pass + + def feed_eof(self) -> Optional[_MsgT]: + if self._payload_parser is not None: + self._payload_parser.feed_eof() + self._payload_parser = None + else: + # try to extract partial message + if self._tail: + self._lines.append(self._tail) + + if self._lines: + if self._lines[-1] != "\r\n": + self._lines.append(b"") + with suppress(Exception): + return self.parse_message(self._lines) + return None + + def feed_data( + self, + data: bytes, + SEP: bytes = b"\r\n", + EMPTY: bytes = b"", + CONTENT_LENGTH: istr = hdrs.CONTENT_LENGTH, + METH_CONNECT: str = hdrs.METH_CONNECT, + SEC_WEBSOCKET_KEY1: istr = hdrs.SEC_WEBSOCKET_KEY1, + ) -> Tuple[List[Tuple[_MsgT, StreamReader]], bool, bytes]: + + messages = [] + + if self._tail: + data, self._tail = self._tail + data, b"" + + data_len = len(data) + start_pos = 0 + loop = self.loop + + while start_pos < data_len: + + # read HTTP message (request/response line + headers), \r\n\r\n + # and split by lines + if self._payload_parser is None and not self._upgraded: + pos = data.find(SEP, start_pos) + # consume \r\n + if pos == start_pos and not self._lines: + start_pos = pos + 2 + continue + + if pos >= start_pos: + # line found + self._lines.append(data[start_pos:pos]) + start_pos = pos + 2 + + # \r\n\r\n found + if self._lines[-1] == EMPTY: + try: + msg: _MsgT = self.parse_message(self._lines) + finally: + self._lines.clear() + + def get_content_length() -> Optional[int]: + # payload length + length_hdr = msg.headers.get(CONTENT_LENGTH) + if length_hdr is None: + return None + + try: + length = int(length_hdr) + except ValueError: + raise InvalidHeader(CONTENT_LENGTH) + + if length < 0: + raise InvalidHeader(CONTENT_LENGTH) + + return length + + length = get_content_length() + # do not support old websocket spec + if SEC_WEBSOCKET_KEY1 in msg.headers: + raise InvalidHeader(SEC_WEBSOCKET_KEY1) + + self._upgraded = msg.upgrade + + method = getattr(msg, "method", self.method) + + assert self.protocol is not None + # calculate payload + if ( + (length is not None and length > 0) + or msg.chunked + and not msg.upgrade + ): + payload = StreamReader( + self.protocol, + timer=self.timer, + loop=loop, + limit=self._limit, + ) + payload_parser = HttpPayloadParser( + payload, + length=length, + chunked=msg.chunked, + method=method, + compression=msg.compression, + code=self.code, + readall=self.readall, + response_with_body=self.response_with_body, + auto_decompress=self._auto_decompress, + ) + if not payload_parser.done: + self._payload_parser = payload_parser + elif method == METH_CONNECT: + assert isinstance(msg, RawRequestMessage) + payload = StreamReader( + self.protocol, + timer=self.timer, + loop=loop, + limit=self._limit, + ) + self._upgraded = True + self._payload_parser = HttpPayloadParser( + payload, + method=msg.method, + compression=msg.compression, + readall=True, + auto_decompress=self._auto_decompress, + ) + else: + if ( + getattr(msg, "code", 100) >= 199 + and length is None + and self.read_until_eof + ): + payload = StreamReader( + self.protocol, + timer=self.timer, + loop=loop, + limit=self._limit, + ) + payload_parser = HttpPayloadParser( + payload, + length=length, + chunked=msg.chunked, + method=method, + compression=msg.compression, + code=self.code, + readall=True, + response_with_body=self.response_with_body, + auto_decompress=self._auto_decompress, + ) + if not payload_parser.done: + self._payload_parser = payload_parser + else: + payload = EMPTY_PAYLOAD + + messages.append((msg, payload)) + else: + self._tail = data[start_pos:] + data = EMPTY + break + + # no parser, just store + elif self._payload_parser is None and self._upgraded: + assert not self._lines + break + + # feed payload + elif data and start_pos < data_len: + assert not self._lines + assert self._payload_parser is not None + try: + eof, data = self._payload_parser.feed_data(data[start_pos:]) + except BaseException as exc: + if self.payload_exception is not None: + self._payload_parser.payload.set_exception( + self.payload_exception(str(exc)) + ) + else: + self._payload_parser.payload.set_exception(exc) + + eof = True + data = b"" + + if eof: + start_pos = 0 + data_len = len(data) + self._payload_parser = None + continue + else: + break + + if data and start_pos < data_len: + data = data[start_pos:] + else: + data = EMPTY + + return messages, self._upgraded, data + + def parse_headers( + self, lines: List[bytes] + ) -> Tuple[ + "CIMultiDictProxy[str]", RawHeaders, Optional[bool], Optional[str], bool, bool + ]: + """Parses RFC 5322 headers from a stream. + + Line continuations are supported. Returns list of header name + and value pairs. Header name is in upper case. + """ + headers, raw_headers = self._headers_parser.parse_headers(lines) + close_conn = None + encoding = None + upgrade = False + chunked = False + + # keep-alive + conn = headers.get(hdrs.CONNECTION) + if conn: + v = conn.lower() + if v == "close": + close_conn = True + elif v == "keep-alive": + close_conn = False + elif v == "upgrade": + upgrade = True + + # encoding + enc = headers.get(hdrs.CONTENT_ENCODING) + if enc: + enc = enc.lower() + if enc in ("gzip", "deflate", "br"): + encoding = enc + + # chunking + te = headers.get(hdrs.TRANSFER_ENCODING) + if te is not None: + if "chunked" == te.lower(): + chunked = True + else: + raise BadHttpMessage("Request has invalid `Transfer-Encoding`") + + if hdrs.CONTENT_LENGTH in headers: + raise BadHttpMessage( + "Content-Length can't be present with Transfer-Encoding", + ) + + return (headers, raw_headers, close_conn, encoding, upgrade, chunked) + + def set_upgraded(self, val: bool) -> None: + """Set connection upgraded (to websocket) mode. + + :param bool val: new state. + """ + self._upgraded = val + + +class HttpRequestParser(HttpParser[RawRequestMessage]): + """Read request status line. + + Exception .http_exceptions.BadStatusLine + could be raised in case of any errors in status line. + Returns RawRequestMessage. + """ + + def parse_message(self, lines: List[bytes]) -> RawRequestMessage: + # request line + line = lines[0].decode("utf-8", "surrogateescape") + try: + method, path, version = line.split(None, 2) + except ValueError: + raise BadStatusLine(line) from None + + if len(path) > self.max_line_size: + raise LineTooLong( + "Status line is too long", str(self.max_line_size), str(len(path)) + ) + + path_part, _hash_separator, url_fragment = path.partition("#") + path_part, _question_mark_separator, qs_part = path_part.partition("?") + + # method + if not METHRE.match(method): + raise BadStatusLine(method) + + # version + try: + if version.startswith("HTTP/"): + n1, n2 = version[5:].split(".", 1) + version_o = HttpVersion(int(n1), int(n2)) + else: + raise BadStatusLine(version) + except Exception: + raise BadStatusLine(version) + + # read headers + ( + headers, + raw_headers, + close, + compression, + upgrade, + chunked, + ) = self.parse_headers(lines) + + if close is None: # then the headers weren't set in the request + if version_o <= HttpVersion10: # HTTP 1.0 must asks to not close + close = True + else: # HTTP 1.1 must ask to close. + close = False + + return RawRequestMessage( + method, + path, + version_o, + headers, + raw_headers, + close, + compression, + upgrade, + chunked, + # NOTE: `yarl.URL.build()` is used to mimic what the Cython-based + # NOTE: parser does, otherwise it results into the same + # NOTE: HTTP Request-Line input producing different + # NOTE: `yarl.URL()` objects + URL.build( + path=path_part, + query_string=qs_part, + fragment=url_fragment, + encoded=True, + ), + ) + + +class HttpResponseParser(HttpParser[RawResponseMessage]): + """Read response status line and headers. + + BadStatusLine could be raised in case of any errors in status line. + Returns RawResponseMessage. + """ + + def parse_message(self, lines: List[bytes]) -> RawResponseMessage: + line = lines[0].decode("utf-8", "surrogateescape") + try: + version, status = line.split(None, 1) + except ValueError: + raise BadStatusLine(line) from None + + try: + status, reason = status.split(None, 1) + except ValueError: + reason = "" + + if len(reason) > self.max_line_size: + raise LineTooLong( + "Status line is too long", str(self.max_line_size), str(len(reason)) + ) + + # version + match = VERSRE.match(version) + if match is None: + raise BadStatusLine(line) + version_o = HttpVersion(int(match.group(1)), int(match.group(2))) + + # The status code is a three-digit number + try: + status_i = int(status) + except ValueError: + raise BadStatusLine(line) from None + + if status_i > 999: + raise BadStatusLine(line) + + # read headers + ( + headers, + raw_headers, + close, + compression, + upgrade, + chunked, + ) = self.parse_headers(lines) + + if close is None: + close = version_o <= HttpVersion10 + + return RawResponseMessage( + version_o, + status_i, + reason.strip(), + headers, + raw_headers, + close, + compression, + upgrade, + chunked, + ) + + +class HttpPayloadParser: + def __init__( + self, + payload: StreamReader, + length: Optional[int] = None, + chunked: bool = False, + compression: Optional[str] = None, + code: Optional[int] = None, + method: Optional[str] = None, + readall: bool = False, + response_with_body: bool = True, + auto_decompress: bool = True, + ) -> None: + self._length = 0 + self._type = ParseState.PARSE_NONE + self._chunk = ChunkState.PARSE_CHUNKED_SIZE + self._chunk_size = 0 + self._chunk_tail = b"" + self._auto_decompress = auto_decompress + self.done = False + + # payload decompression wrapper + if response_with_body and compression and self._auto_decompress: + real_payload = DeflateBuffer( + payload, compression + ) # type: Union[StreamReader, DeflateBuffer] + else: + real_payload = payload + + # payload parser + if not response_with_body: + # don't parse payload if it's not expected to be received + self._type = ParseState.PARSE_NONE + real_payload.feed_eof() + self.done = True + + elif chunked: + self._type = ParseState.PARSE_CHUNKED + elif length is not None: + self._type = ParseState.PARSE_LENGTH + self._length = length + if self._length == 0: + real_payload.feed_eof() + self.done = True + else: + if readall and code != 204: + self._type = ParseState.PARSE_UNTIL_EOF + elif method in ("PUT", "POST"): + internal_logger.warning( # pragma: no cover + "Content-Length or Transfer-Encoding header is required" + ) + self._type = ParseState.PARSE_NONE + real_payload.feed_eof() + self.done = True + + self.payload = real_payload + + def feed_eof(self) -> None: + if self._type == ParseState.PARSE_UNTIL_EOF: + self.payload.feed_eof() + elif self._type == ParseState.PARSE_LENGTH: + raise ContentLengthError( + "Not enough data for satisfy content length header." + ) + elif self._type == ParseState.PARSE_CHUNKED: + raise TransferEncodingError( + "Not enough data for satisfy transfer length header." + ) + + def feed_data( + self, chunk: bytes, SEP: bytes = b"\r\n", CHUNK_EXT: bytes = b";" + ) -> Tuple[bool, bytes]: + # Read specified amount of bytes + if self._type == ParseState.PARSE_LENGTH: + required = self._length + chunk_len = len(chunk) + + if required >= chunk_len: + self._length = required - chunk_len + self.payload.feed_data(chunk, chunk_len) + if self._length == 0: + self.payload.feed_eof() + return True, b"" + else: + self._length = 0 + self.payload.feed_data(chunk[:required], required) + self.payload.feed_eof() + return True, chunk[required:] + + # Chunked transfer encoding parser + elif self._type == ParseState.PARSE_CHUNKED: + if self._chunk_tail: + chunk = self._chunk_tail + chunk + self._chunk_tail = b"" + + while chunk: + + # read next chunk size + if self._chunk == ChunkState.PARSE_CHUNKED_SIZE: + pos = chunk.find(SEP) + if pos >= 0: + i = chunk.find(CHUNK_EXT, 0, pos) + if i >= 0: + size_b = chunk[:i] # strip chunk-extensions + else: + size_b = chunk[:pos] + + try: + size = int(bytes(size_b), 16) + except ValueError: + exc = TransferEncodingError( + chunk[:pos].decode("ascii", "surrogateescape") + ) + self.payload.set_exception(exc) + raise exc from None + + chunk = chunk[pos + 2 :] + if size == 0: # eof marker + self._chunk = ChunkState.PARSE_MAYBE_TRAILERS + else: + self._chunk = ChunkState.PARSE_CHUNKED_CHUNK + self._chunk_size = size + self.payload.begin_http_chunk_receiving() + else: + self._chunk_tail = chunk + return False, b"" + + # read chunk and feed buffer + if self._chunk == ChunkState.PARSE_CHUNKED_CHUNK: + required = self._chunk_size + chunk_len = len(chunk) + + if required > chunk_len: + self._chunk_size = required - chunk_len + self.payload.feed_data(chunk, chunk_len) + return False, b"" + else: + self._chunk_size = 0 + self.payload.feed_data(chunk[:required], required) + chunk = chunk[required:] + self._chunk = ChunkState.PARSE_CHUNKED_CHUNK_EOF + self.payload.end_http_chunk_receiving() + + # toss the CRLF at the end of the chunk + if self._chunk == ChunkState.PARSE_CHUNKED_CHUNK_EOF: + if chunk[:2] == SEP: + chunk = chunk[2:] + self._chunk = ChunkState.PARSE_CHUNKED_SIZE + else: + self._chunk_tail = chunk + return False, b"" + + # if stream does not contain trailer, after 0\r\n + # we should get another \r\n otherwise + # trailers needs to be skiped until \r\n\r\n + if self._chunk == ChunkState.PARSE_MAYBE_TRAILERS: + head = chunk[:2] + if head == SEP: + # end of stream + self.payload.feed_eof() + return True, chunk[2:] + # Both CR and LF, or only LF may not be received yet. It is + # expected that CRLF or LF will be shown at the very first + # byte next time, otherwise trailers should come. The last + # CRLF which marks the end of response might not be + # contained in the same TCP segment which delivered the + # size indicator. + if not head: + return False, b"" + if head == SEP[:1]: + self._chunk_tail = head + return False, b"" + self._chunk = ChunkState.PARSE_TRAILERS + + # read and discard trailer up to the CRLF terminator + if self._chunk == ChunkState.PARSE_TRAILERS: + pos = chunk.find(SEP) + if pos >= 0: + chunk = chunk[pos + 2 :] + self._chunk = ChunkState.PARSE_MAYBE_TRAILERS + else: + self._chunk_tail = chunk + return False, b"" + + # Read all bytes until eof + elif self._type == ParseState.PARSE_UNTIL_EOF: + self.payload.feed_data(chunk, len(chunk)) + + return False, b"" + + +class DeflateBuffer: + """DeflateStream decompress stream and feed data into specified stream.""" + + decompressor: Any + + def __init__(self, out: StreamReader, encoding: Optional[str]) -> None: + self.out = out + self.size = 0 + self.encoding = encoding + self._started_decoding = False + + if encoding == "br": + if not HAS_BROTLI: # pragma: no cover + raise ContentEncodingError( + "Can not decode content-encoding: brotli (br). " + "Please install `Brotli`" + ) + + class BrotliDecoder: + # Supports both 'brotlipy' and 'Brotli' packages + # since they share an import name. The top branches + # are for 'brotlipy' and bottom branches for 'Brotli' + def __init__(self) -> None: + self._obj = brotli.Decompressor() + + def decompress(self, data: bytes) -> bytes: + if hasattr(self._obj, "decompress"): + return cast(bytes, self._obj.decompress(data)) + return cast(bytes, self._obj.process(data)) + + def flush(self) -> bytes: + if hasattr(self._obj, "flush"): + return cast(bytes, self._obj.flush()) + return b"" + + self.decompressor = BrotliDecoder() + else: + zlib_mode = 16 + zlib.MAX_WBITS if encoding == "gzip" else zlib.MAX_WBITS + self.decompressor = zlib.decompressobj(wbits=zlib_mode) + + def set_exception(self, exc: BaseException) -> None: + self.out.set_exception(exc) + + def feed_data(self, chunk: bytes, size: int) -> None: + if not size: + return + + self.size += size + + # RFC1950 + # bits 0..3 = CM = 0b1000 = 8 = "deflate" + # bits 4..7 = CINFO = 1..7 = windows size. + if ( + not self._started_decoding + and self.encoding == "deflate" + and chunk[0] & 0xF != 8 + ): + # Change the decoder to decompress incorrectly compressed data + # Actually we should issue a warning about non-RFC-compliant data. + self.decompressor = zlib.decompressobj(wbits=-zlib.MAX_WBITS) + + try: + chunk = self.decompressor.decompress(chunk) + except Exception: + raise ContentEncodingError( + "Can not decode content-encoding: %s" % self.encoding + ) + + self._started_decoding = True + + if chunk: + self.out.feed_data(chunk, len(chunk)) + + def feed_eof(self) -> None: + chunk = self.decompressor.flush() + + if chunk or self.size > 0: + self.out.feed_data(chunk, len(chunk)) + if self.encoding == "deflate" and not self.decompressor.eof: + raise ContentEncodingError("deflate") + + self.out.feed_eof() + + def begin_http_chunk_receiving(self) -> None: + self.out.begin_http_chunk_receiving() + + def end_http_chunk_receiving(self) -> None: + self.out.end_http_chunk_receiving() + + +HttpRequestParserPy = HttpRequestParser +HttpResponseParserPy = HttpResponseParser +RawRequestMessagePy = RawRequestMessage +RawResponseMessagePy = RawResponseMessage + +try: + if not NO_EXTENSIONS: + from ._http_parser import ( # type: ignore[import,no-redef] + HttpRequestParser, + HttpResponseParser, + RawRequestMessage, + RawResponseMessage, + ) + + HttpRequestParserC = HttpRequestParser + HttpResponseParserC = HttpResponseParser + RawRequestMessageC = RawRequestMessage + RawResponseMessageC = RawResponseMessage +except ImportError: # pragma: no cover + pass diff --git a/contrib/python/aiohttp/aiohttp/http_websocket.py b/contrib/python/aiohttp/aiohttp/http_websocket.py new file mode 100644 index 0000000000..991a149d09 --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/http_websocket.py @@ -0,0 +1,701 @@ +"""WebSocket protocol versions 13 and 8.""" + +import asyncio +import collections +import json +import random +import re +import sys +import zlib +from enum import IntEnum +from struct import Struct +from typing import Any, Callable, List, Optional, Pattern, Set, Tuple, Union, cast + +from .base_protocol import BaseProtocol +from .helpers import NO_EXTENSIONS +from .streams import DataQueue +from .typedefs import Final + +__all__ = ( + "WS_CLOSED_MESSAGE", + "WS_CLOSING_MESSAGE", + "WS_KEY", + "WebSocketReader", + "WebSocketWriter", + "WSMessage", + "WebSocketError", + "WSMsgType", + "WSCloseCode", +) + + +class WSCloseCode(IntEnum): + OK = 1000 + GOING_AWAY = 1001 + PROTOCOL_ERROR = 1002 + UNSUPPORTED_DATA = 1003 + ABNORMAL_CLOSURE = 1006 + INVALID_TEXT = 1007 + POLICY_VIOLATION = 1008 + MESSAGE_TOO_BIG = 1009 + MANDATORY_EXTENSION = 1010 + INTERNAL_ERROR = 1011 + SERVICE_RESTART = 1012 + TRY_AGAIN_LATER = 1013 + BAD_GATEWAY = 1014 + + +ALLOWED_CLOSE_CODES: Final[Set[int]] = {int(i) for i in WSCloseCode} + + +class WSMsgType(IntEnum): + # websocket spec types + CONTINUATION = 0x0 + TEXT = 0x1 + BINARY = 0x2 + PING = 0x9 + PONG = 0xA + CLOSE = 0x8 + + # aiohttp specific types + CLOSING = 0x100 + CLOSED = 0x101 + ERROR = 0x102 + + text = TEXT + binary = BINARY + ping = PING + pong = PONG + close = CLOSE + closing = CLOSING + closed = CLOSED + error = ERROR + + +WS_KEY: Final[bytes] = b"258EAFA5-E914-47DA-95CA-C5AB0DC85B11" + + +UNPACK_LEN2 = Struct("!H").unpack_from +UNPACK_LEN3 = Struct("!Q").unpack_from +UNPACK_CLOSE_CODE = Struct("!H").unpack +PACK_LEN1 = Struct("!BB").pack +PACK_LEN2 = Struct("!BBH").pack +PACK_LEN3 = Struct("!BBQ").pack +PACK_CLOSE_CODE = Struct("!H").pack +MSG_SIZE: Final[int] = 2 ** 14 +DEFAULT_LIMIT: Final[int] = 2 ** 16 + + +_WSMessageBase = collections.namedtuple("_WSMessageBase", ["type", "data", "extra"]) + + +class WSMessage(_WSMessageBase): + def json(self, *, loads: Callable[[Any], Any] = json.loads) -> Any: + """Return parsed JSON data. + + .. versionadded:: 0.22 + """ + return loads(self.data) + + +WS_CLOSED_MESSAGE = WSMessage(WSMsgType.CLOSED, None, None) +WS_CLOSING_MESSAGE = WSMessage(WSMsgType.CLOSING, None, None) + + +class WebSocketError(Exception): + """WebSocket protocol parser error.""" + + def __init__(self, code: int, message: str) -> None: + self.code = code + super().__init__(code, message) + + def __str__(self) -> str: + return cast(str, self.args[1]) + + +class WSHandshakeError(Exception): + """WebSocket protocol handshake error.""" + + +native_byteorder: Final[str] = sys.byteorder + + +# Used by _websocket_mask_python +_XOR_TABLE: Final[List[bytes]] = [bytes(a ^ b for a in range(256)) for b in range(256)] + + +def _websocket_mask_python(mask: bytes, data: bytearray) -> None: + """Websocket masking function. + + `mask` is a `bytes` object of length 4; `data` is a `bytearray` + object of any length. The contents of `data` are masked with `mask`, + as specified in section 5.3 of RFC 6455. + + Note that this function mutates the `data` argument. + + This pure-python implementation may be replaced by an optimized + version when available. + + """ + assert isinstance(data, bytearray), data + assert len(mask) == 4, mask + + if data: + a, b, c, d = (_XOR_TABLE[n] for n in mask) + data[::4] = data[::4].translate(a) + data[1::4] = data[1::4].translate(b) + data[2::4] = data[2::4].translate(c) + data[3::4] = data[3::4].translate(d) + + +if NO_EXTENSIONS: # pragma: no cover + _websocket_mask = _websocket_mask_python +else: + try: + from ._websocket import _websocket_mask_cython # type: ignore[import] + + _websocket_mask = _websocket_mask_cython + except ImportError: # pragma: no cover + _websocket_mask = _websocket_mask_python + +_WS_DEFLATE_TRAILING: Final[bytes] = bytes([0x00, 0x00, 0xFF, 0xFF]) + + +_WS_EXT_RE: Final[Pattern[str]] = re.compile( + r"^(?:;\s*(?:" + r"(server_no_context_takeover)|" + r"(client_no_context_takeover)|" + r"(server_max_window_bits(?:=(\d+))?)|" + r"(client_max_window_bits(?:=(\d+))?)))*$" +) + +_WS_EXT_RE_SPLIT: Final[Pattern[str]] = re.compile(r"permessage-deflate([^,]+)?") + + +def ws_ext_parse(extstr: Optional[str], isserver: bool = False) -> Tuple[int, bool]: + if not extstr: + return 0, False + + compress = 0 + notakeover = False + for ext in _WS_EXT_RE_SPLIT.finditer(extstr): + defext = ext.group(1) + # Return compress = 15 when get `permessage-deflate` + if not defext: + compress = 15 + break + match = _WS_EXT_RE.match(defext) + if match: + compress = 15 + if isserver: + # Server never fail to detect compress handshake. + # Server does not need to send max wbit to client + if match.group(4): + compress = int(match.group(4)) + # Group3 must match if group4 matches + # Compress wbit 8 does not support in zlib + # If compress level not support, + # CONTINUE to next extension + if compress > 15 or compress < 9: + compress = 0 + continue + if match.group(1): + notakeover = True + # Ignore regex group 5 & 6 for client_max_window_bits + break + else: + if match.group(6): + compress = int(match.group(6)) + # Group5 must match if group6 matches + # Compress wbit 8 does not support in zlib + # If compress level not support, + # FAIL the parse progress + if compress > 15 or compress < 9: + raise WSHandshakeError("Invalid window size") + if match.group(2): + notakeover = True + # Ignore regex group 5 & 6 for client_max_window_bits + break + # Return Fail if client side and not match + elif not isserver: + raise WSHandshakeError("Extension for deflate not supported" + ext.group(1)) + + return compress, notakeover + + +def ws_ext_gen( + compress: int = 15, isserver: bool = False, server_notakeover: bool = False +) -> str: + # client_notakeover=False not used for server + # compress wbit 8 does not support in zlib + if compress < 9 or compress > 15: + raise ValueError( + "Compress wbits must between 9 and 15, " "zlib does not support wbits=8" + ) + enabledext = ["permessage-deflate"] + if not isserver: + enabledext.append("client_max_window_bits") + + if compress < 15: + enabledext.append("server_max_window_bits=" + str(compress)) + if server_notakeover: + enabledext.append("server_no_context_takeover") + # if client_notakeover: + # enabledext.append('client_no_context_takeover') + return "; ".join(enabledext) + + +class WSParserState(IntEnum): + READ_HEADER = 1 + READ_PAYLOAD_LENGTH = 2 + READ_PAYLOAD_MASK = 3 + READ_PAYLOAD = 4 + + +class WebSocketReader: + def __init__( + self, queue: DataQueue[WSMessage], max_msg_size: int, compress: bool = True + ) -> None: + self.queue = queue + self._max_msg_size = max_msg_size + + self._exc = None # type: Optional[BaseException] + self._partial = bytearray() + self._state = WSParserState.READ_HEADER + + self._opcode = None # type: Optional[int] + self._frame_fin = False + self._frame_opcode = None # type: Optional[int] + self._frame_payload = bytearray() + + self._tail = b"" + self._has_mask = False + self._frame_mask = None # type: Optional[bytes] + self._payload_length = 0 + self._payload_length_flag = 0 + self._compressed = None # type: Optional[bool] + self._decompressobj = None # type: Any # zlib.decompressobj actually + self._compress = compress + + def feed_eof(self) -> None: + self.queue.feed_eof() + + def feed_data(self, data: bytes) -> Tuple[bool, bytes]: + if self._exc: + return True, data + + try: + return self._feed_data(data) + except Exception as exc: + self._exc = exc + self.queue.set_exception(exc) + return True, b"" + + def _feed_data(self, data: bytes) -> Tuple[bool, bytes]: + for fin, opcode, payload, compressed in self.parse_frame(data): + if compressed and not self._decompressobj: + self._decompressobj = zlib.decompressobj(wbits=-zlib.MAX_WBITS) + if opcode == WSMsgType.CLOSE: + if len(payload) >= 2: + close_code = UNPACK_CLOSE_CODE(payload[:2])[0] + if close_code < 3000 and close_code not in ALLOWED_CLOSE_CODES: + raise WebSocketError( + WSCloseCode.PROTOCOL_ERROR, + f"Invalid close code: {close_code}", + ) + try: + close_message = payload[2:].decode("utf-8") + except UnicodeDecodeError as exc: + raise WebSocketError( + WSCloseCode.INVALID_TEXT, "Invalid UTF-8 text message" + ) from exc + msg = WSMessage(WSMsgType.CLOSE, close_code, close_message) + elif payload: + raise WebSocketError( + WSCloseCode.PROTOCOL_ERROR, + f"Invalid close frame: {fin} {opcode} {payload!r}", + ) + else: + msg = WSMessage(WSMsgType.CLOSE, 0, "") + + self.queue.feed_data(msg, 0) + + elif opcode == WSMsgType.PING: + self.queue.feed_data( + WSMessage(WSMsgType.PING, payload, ""), len(payload) + ) + + elif opcode == WSMsgType.PONG: + self.queue.feed_data( + WSMessage(WSMsgType.PONG, payload, ""), len(payload) + ) + + elif ( + opcode not in (WSMsgType.TEXT, WSMsgType.BINARY) + and self._opcode is None + ): + raise WebSocketError( + WSCloseCode.PROTOCOL_ERROR, f"Unexpected opcode={opcode!r}" + ) + else: + # load text/binary + if not fin: + # got partial frame payload + if opcode != WSMsgType.CONTINUATION: + self._opcode = opcode + self._partial.extend(payload) + if self._max_msg_size and len(self._partial) >= self._max_msg_size: + raise WebSocketError( + WSCloseCode.MESSAGE_TOO_BIG, + "Message size {} exceeds limit {}".format( + len(self._partial), self._max_msg_size + ), + ) + else: + # previous frame was non finished + # we should get continuation opcode + if self._partial: + if opcode != WSMsgType.CONTINUATION: + raise WebSocketError( + WSCloseCode.PROTOCOL_ERROR, + "The opcode in non-fin frame is expected " + "to be zero, got {!r}".format(opcode), + ) + + if opcode == WSMsgType.CONTINUATION: + assert self._opcode is not None + opcode = self._opcode + self._opcode = None + + self._partial.extend(payload) + if self._max_msg_size and len(self._partial) >= self._max_msg_size: + raise WebSocketError( + WSCloseCode.MESSAGE_TOO_BIG, + "Message size {} exceeds limit {}".format( + len(self._partial), self._max_msg_size + ), + ) + + # Decompress process must to be done after all packets + # received. + if compressed: + self._partial.extend(_WS_DEFLATE_TRAILING) + payload_merged = self._decompressobj.decompress( + self._partial, self._max_msg_size + ) + if self._decompressobj.unconsumed_tail: + left = len(self._decompressobj.unconsumed_tail) + raise WebSocketError( + WSCloseCode.MESSAGE_TOO_BIG, + "Decompressed message size {} exceeds limit {}".format( + self._max_msg_size + left, self._max_msg_size + ), + ) + else: + payload_merged = bytes(self._partial) + + self._partial.clear() + + if opcode == WSMsgType.TEXT: + try: + text = payload_merged.decode("utf-8") + self.queue.feed_data( + WSMessage(WSMsgType.TEXT, text, ""), len(text) + ) + except UnicodeDecodeError as exc: + raise WebSocketError( + WSCloseCode.INVALID_TEXT, "Invalid UTF-8 text message" + ) from exc + else: + self.queue.feed_data( + WSMessage(WSMsgType.BINARY, payload_merged, ""), + len(payload_merged), + ) + + return False, b"" + + def parse_frame( + self, buf: bytes + ) -> List[Tuple[bool, Optional[int], bytearray, Optional[bool]]]: + """Return the next frame from the socket.""" + frames = [] + if self._tail: + buf, self._tail = self._tail + buf, b"" + + start_pos = 0 + buf_length = len(buf) + + while True: + # read header + if self._state == WSParserState.READ_HEADER: + if buf_length - start_pos >= 2: + data = buf[start_pos : start_pos + 2] + start_pos += 2 + first_byte, second_byte = data + + fin = (first_byte >> 7) & 1 + rsv1 = (first_byte >> 6) & 1 + rsv2 = (first_byte >> 5) & 1 + rsv3 = (first_byte >> 4) & 1 + opcode = first_byte & 0xF + + # frame-fin = %x0 ; more frames of this message follow + # / %x1 ; final frame of this message + # frame-rsv1 = %x0 ; + # 1 bit, MUST be 0 unless negotiated otherwise + # frame-rsv2 = %x0 ; + # 1 bit, MUST be 0 unless negotiated otherwise + # frame-rsv3 = %x0 ; + # 1 bit, MUST be 0 unless negotiated otherwise + # + # Remove rsv1 from this test for deflate development + if rsv2 or rsv3 or (rsv1 and not self._compress): + raise WebSocketError( + WSCloseCode.PROTOCOL_ERROR, + "Received frame with non-zero reserved bits", + ) + + if opcode > 0x7 and fin == 0: + raise WebSocketError( + WSCloseCode.PROTOCOL_ERROR, + "Received fragmented control frame", + ) + + has_mask = (second_byte >> 7) & 1 + length = second_byte & 0x7F + + # Control frames MUST have a payload + # length of 125 bytes or less + if opcode > 0x7 and length > 125: + raise WebSocketError( + WSCloseCode.PROTOCOL_ERROR, + "Control frame payload cannot be " "larger than 125 bytes", + ) + + # Set compress status if last package is FIN + # OR set compress status if this is first fragment + # Raise error if not first fragment with rsv1 = 0x1 + if self._frame_fin or self._compressed is None: + self._compressed = True if rsv1 else False + elif rsv1: + raise WebSocketError( + WSCloseCode.PROTOCOL_ERROR, + "Received frame with non-zero reserved bits", + ) + + self._frame_fin = bool(fin) + self._frame_opcode = opcode + self._has_mask = bool(has_mask) + self._payload_length_flag = length + self._state = WSParserState.READ_PAYLOAD_LENGTH + else: + break + + # read payload length + if self._state == WSParserState.READ_PAYLOAD_LENGTH: + length = self._payload_length_flag + if length == 126: + if buf_length - start_pos >= 2: + data = buf[start_pos : start_pos + 2] + start_pos += 2 + length = UNPACK_LEN2(data)[0] + self._payload_length = length + self._state = ( + WSParserState.READ_PAYLOAD_MASK + if self._has_mask + else WSParserState.READ_PAYLOAD + ) + else: + break + elif length > 126: + if buf_length - start_pos >= 8: + data = buf[start_pos : start_pos + 8] + start_pos += 8 + length = UNPACK_LEN3(data)[0] + self._payload_length = length + self._state = ( + WSParserState.READ_PAYLOAD_MASK + if self._has_mask + else WSParserState.READ_PAYLOAD + ) + else: + break + else: + self._payload_length = length + self._state = ( + WSParserState.READ_PAYLOAD_MASK + if self._has_mask + else WSParserState.READ_PAYLOAD + ) + + # read payload mask + if self._state == WSParserState.READ_PAYLOAD_MASK: + if buf_length - start_pos >= 4: + self._frame_mask = buf[start_pos : start_pos + 4] + start_pos += 4 + self._state = WSParserState.READ_PAYLOAD + else: + break + + if self._state == WSParserState.READ_PAYLOAD: + length = self._payload_length + payload = self._frame_payload + + chunk_len = buf_length - start_pos + if length >= chunk_len: + self._payload_length = length - chunk_len + payload.extend(buf[start_pos:]) + start_pos = buf_length + else: + self._payload_length = 0 + payload.extend(buf[start_pos : start_pos + length]) + start_pos = start_pos + length + + if self._payload_length == 0: + if self._has_mask: + assert self._frame_mask is not None + _websocket_mask(self._frame_mask, payload) + + frames.append( + (self._frame_fin, self._frame_opcode, payload, self._compressed) + ) + + self._frame_payload = bytearray() + self._state = WSParserState.READ_HEADER + else: + break + + self._tail = buf[start_pos:] + + return frames + + +class WebSocketWriter: + def __init__( + self, + protocol: BaseProtocol, + transport: asyncio.Transport, + *, + use_mask: bool = False, + limit: int = DEFAULT_LIMIT, + random: Any = random.Random(), + compress: int = 0, + notakeover: bool = False, + ) -> None: + self.protocol = protocol + self.transport = transport + self.use_mask = use_mask + self.randrange = random.randrange + self.compress = compress + self.notakeover = notakeover + self._closing = False + self._limit = limit + self._output_size = 0 + self._compressobj = None # type: Any # actually compressobj + + async def _send_frame( + self, message: bytes, opcode: int, compress: Optional[int] = None + ) -> None: + """Send a frame over the websocket with message as its payload.""" + if self._closing and not (opcode & WSMsgType.CLOSE): + raise ConnectionResetError("Cannot write to closing transport") + + rsv = 0 + + # Only compress larger packets (disabled) + # Does small packet needs to be compressed? + # if self.compress and opcode < 8 and len(message) > 124: + if (compress or self.compress) and opcode < 8: + if compress: + # Do not set self._compress if compressing is for this frame + compressobj = zlib.compressobj(level=zlib.Z_BEST_SPEED, wbits=-compress) + else: # self.compress + if not self._compressobj: + self._compressobj = zlib.compressobj( + level=zlib.Z_BEST_SPEED, wbits=-self.compress + ) + compressobj = self._compressobj + + message = compressobj.compress(message) + message = message + compressobj.flush( + zlib.Z_FULL_FLUSH if self.notakeover else zlib.Z_SYNC_FLUSH + ) + if message.endswith(_WS_DEFLATE_TRAILING): + message = message[:-4] + rsv = rsv | 0x40 + + msg_length = len(message) + + use_mask = self.use_mask + if use_mask: + mask_bit = 0x80 + else: + mask_bit = 0 + + if msg_length < 126: + header = PACK_LEN1(0x80 | rsv | opcode, msg_length | mask_bit) + elif msg_length < (1 << 16): + header = PACK_LEN2(0x80 | rsv | opcode, 126 | mask_bit, msg_length) + else: + header = PACK_LEN3(0x80 | rsv | opcode, 127 | mask_bit, msg_length) + if use_mask: + mask = self.randrange(0, 0xFFFFFFFF) + mask = mask.to_bytes(4, "big") + message = bytearray(message) + _websocket_mask(mask, message) + self._write(header + mask + message) + self._output_size += len(header) + len(mask) + len(message) + else: + if len(message) > MSG_SIZE: + self._write(header) + self._write(message) + else: + self._write(header + message) + + self._output_size += len(header) + len(message) + + if self._output_size > self._limit: + self._output_size = 0 + await self.protocol._drain_helper() + + def _write(self, data: bytes) -> None: + if self.transport is None or self.transport.is_closing(): + raise ConnectionResetError("Cannot write to closing transport") + self.transport.write(data) + + async def pong(self, message: bytes = b"") -> None: + """Send pong message.""" + if isinstance(message, str): + message = message.encode("utf-8") + await self._send_frame(message, WSMsgType.PONG) + + async def ping(self, message: bytes = b"") -> None: + """Send ping message.""" + if isinstance(message, str): + message = message.encode("utf-8") + await self._send_frame(message, WSMsgType.PING) + + async def send( + self, + message: Union[str, bytes], + binary: bool = False, + compress: Optional[int] = None, + ) -> None: + """Send a frame over the websocket with message as its payload.""" + if isinstance(message, str): + message = message.encode("utf-8") + if binary: + await self._send_frame(message, WSMsgType.BINARY, compress) + else: + await self._send_frame(message, WSMsgType.TEXT, compress) + + async def close(self, code: int = 1000, message: bytes = b"") -> None: + """Close the websocket, sending the specified code and message.""" + if isinstance(message, str): + message = message.encode("utf-8") + try: + await self._send_frame( + PACK_CLOSE_CODE(code) + message, opcode=WSMsgType.CLOSE + ) + finally: + self._closing = True diff --git a/contrib/python/aiohttp/aiohttp/http_writer.py b/contrib/python/aiohttp/aiohttp/http_writer.py new file mode 100644 index 0000000000..e09144736c --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/http_writer.py @@ -0,0 +1,200 @@ +"""Http related parsers and protocol.""" + +import asyncio +import zlib +from typing import Any, Awaitable, Callable, NamedTuple, Optional, Union # noqa + +from multidict import CIMultiDict + +from .abc import AbstractStreamWriter +from .base_protocol import BaseProtocol +from .helpers import NO_EXTENSIONS + +__all__ = ("StreamWriter", "HttpVersion", "HttpVersion10", "HttpVersion11") + + +class HttpVersion(NamedTuple): + major: int + minor: int + + +HttpVersion10 = HttpVersion(1, 0) +HttpVersion11 = HttpVersion(1, 1) + + +_T_OnChunkSent = Optional[Callable[[bytes], Awaitable[None]]] +_T_OnHeadersSent = Optional[Callable[["CIMultiDict[str]"], Awaitable[None]]] + + +class StreamWriter(AbstractStreamWriter): + def __init__( + self, + protocol: BaseProtocol, + loop: asyncio.AbstractEventLoop, + on_chunk_sent: _T_OnChunkSent = None, + on_headers_sent: _T_OnHeadersSent = None, + ) -> None: + self._protocol = protocol + self._transport = protocol.transport + + self.loop = loop + self.length = None + self.chunked = False + self.buffer_size = 0 + self.output_size = 0 + + self._eof = False + self._compress = None # type: Any + self._drain_waiter = None + + self._on_chunk_sent = on_chunk_sent # type: _T_OnChunkSent + self._on_headers_sent = on_headers_sent # type: _T_OnHeadersSent + + @property + def transport(self) -> Optional[asyncio.Transport]: + return self._transport + + @property + def protocol(self) -> BaseProtocol: + return self._protocol + + def enable_chunking(self) -> None: + self.chunked = True + + def enable_compression( + self, encoding: str = "deflate", strategy: int = zlib.Z_DEFAULT_STRATEGY + ) -> None: + zlib_mode = 16 + zlib.MAX_WBITS if encoding == "gzip" else zlib.MAX_WBITS + self._compress = zlib.compressobj(wbits=zlib_mode, strategy=strategy) + + def _write(self, chunk: bytes) -> None: + size = len(chunk) + self.buffer_size += size + self.output_size += size + + if self._transport is None or self._transport.is_closing(): + raise ConnectionResetError("Cannot write to closing transport") + self._transport.write(chunk) + + async def write( + self, chunk: bytes, *, drain: bool = True, LIMIT: int = 0x10000 + ) -> None: + """Writes chunk of data to a stream. + + write_eof() indicates end of stream. + writer can't be used after write_eof() method being called. + write() return drain future. + """ + if self._on_chunk_sent is not None: + await self._on_chunk_sent(chunk) + + if isinstance(chunk, memoryview): + if chunk.nbytes != len(chunk): + # just reshape it + chunk = chunk.cast("c") + + if self._compress is not None: + chunk = self._compress.compress(chunk) + if not chunk: + return + + if self.length is not None: + chunk_len = len(chunk) + if self.length >= chunk_len: + self.length = self.length - chunk_len + else: + chunk = chunk[: self.length] + self.length = 0 + if not chunk: + return + + if chunk: + if self.chunked: + chunk_len_pre = ("%x\r\n" % len(chunk)).encode("ascii") + chunk = chunk_len_pre + chunk + b"\r\n" + + self._write(chunk) + + if self.buffer_size > LIMIT and drain: + self.buffer_size = 0 + await self.drain() + + async def write_headers( + self, status_line: str, headers: "CIMultiDict[str]" + ) -> None: + """Write request/response status and headers.""" + if self._on_headers_sent is not None: + await self._on_headers_sent(headers) + + # status + headers + buf = _serialize_headers(status_line, headers) + self._write(buf) + + async def write_eof(self, chunk: bytes = b"") -> None: + if self._eof: + return + + if chunk and self._on_chunk_sent is not None: + await self._on_chunk_sent(chunk) + + if self._compress: + if chunk: + chunk = self._compress.compress(chunk) + + chunk = chunk + self._compress.flush() + if chunk and self.chunked: + chunk_len = ("%x\r\n" % len(chunk)).encode("ascii") + chunk = chunk_len + chunk + b"\r\n0\r\n\r\n" + else: + if self.chunked: + if chunk: + chunk_len = ("%x\r\n" % len(chunk)).encode("ascii") + chunk = chunk_len + chunk + b"\r\n0\r\n\r\n" + else: + chunk = b"0\r\n\r\n" + + if chunk: + self._write(chunk) + + await self.drain() + + self._eof = True + self._transport = None + + async def drain(self) -> None: + """Flush the write buffer. + + The intended use is to write + + await w.write(data) + await w.drain() + """ + if self._protocol.transport is not None: + await self._protocol._drain_helper() + + +def _safe_header(string: str) -> str: + if "\r" in string or "\n" in string: + raise ValueError( + "Newline or carriage return detected in headers. " + "Potential header injection attack." + ) + return string + + +def _py_serialize_headers(status_line: str, headers: "CIMultiDict[str]") -> bytes: + headers_gen = (_safe_header(k) + ": " + _safe_header(v) for k, v in headers.items()) + line = status_line + "\r\n" + "\r\n".join(headers_gen) + "\r\n\r\n" + return line.encode("utf-8") + + +_serialize_headers = _py_serialize_headers + +try: + import aiohttp._http_writer as _http_writer # type: ignore[import] + + _c_serialize_headers = _http_writer._serialize_headers + if not NO_EXTENSIONS: + _serialize_headers = _c_serialize_headers +except ImportError: + pass diff --git a/contrib/python/aiohttp/aiohttp/locks.py b/contrib/python/aiohttp/aiohttp/locks.py new file mode 100644 index 0000000000..df65e3e47d --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/locks.py @@ -0,0 +1,41 @@ +import asyncio +import collections +from typing import Any, Deque, Optional + + +class EventResultOrError: + """Event asyncio lock helper class. + + Wraps the Event asyncio lock allowing either to awake the + locked Tasks without any error or raising an exception. + + thanks to @vorpalsmith for the simple design. + """ + + def __init__(self, loop: asyncio.AbstractEventLoop) -> None: + self._loop = loop + self._exc = None # type: Optional[BaseException] + self._event = asyncio.Event() + self._waiters = collections.deque() # type: Deque[asyncio.Future[Any]] + + def set(self, exc: Optional[BaseException] = None) -> None: + self._exc = exc + self._event.set() + + async def wait(self) -> Any: + waiter = self._loop.create_task(self._event.wait()) + self._waiters.append(waiter) + try: + val = await waiter + finally: + self._waiters.remove(waiter) + + if self._exc is not None: + raise self._exc + + return val + + def cancel(self) -> None: + """Cancel all waiters""" + for waiter in self._waiters: + waiter.cancel() diff --git a/contrib/python/aiohttp/aiohttp/log.py b/contrib/python/aiohttp/aiohttp/log.py new file mode 100644 index 0000000000..3cecea2bac --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/log.py @@ -0,0 +1,8 @@ +import logging + +access_logger = logging.getLogger("aiohttp.access") +client_logger = logging.getLogger("aiohttp.client") +internal_logger = logging.getLogger("aiohttp.internal") +server_logger = logging.getLogger("aiohttp.server") +web_logger = logging.getLogger("aiohttp.web") +ws_logger = logging.getLogger("aiohttp.websocket") diff --git a/contrib/python/aiohttp/aiohttp/multipart.py b/contrib/python/aiohttp/aiohttp/multipart.py new file mode 100644 index 0000000000..c84e20044f --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/multipart.py @@ -0,0 +1,963 @@ +import base64 +import binascii +import json +import re +import uuid +import warnings +import zlib +from collections import deque +from types import TracebackType +from typing import ( + TYPE_CHECKING, + Any, + AsyncIterator, + Deque, + Dict, + Iterator, + List, + Mapping, + Optional, + Sequence, + Tuple, + Type, + Union, + cast, +) +from urllib.parse import parse_qsl, unquote, urlencode + +from multidict import CIMultiDict, CIMultiDictProxy, MultiMapping + +from .hdrs import ( + CONTENT_DISPOSITION, + CONTENT_ENCODING, + CONTENT_LENGTH, + CONTENT_TRANSFER_ENCODING, + CONTENT_TYPE, +) +from .helpers import CHAR, TOKEN, parse_mimetype, reify +from .http import HeadersParser +from .payload import ( + JsonPayload, + LookupError, + Order, + Payload, + StringPayload, + get_payload, + payload_type, +) +from .streams import StreamReader + +__all__ = ( + "MultipartReader", + "MultipartWriter", + "BodyPartReader", + "BadContentDispositionHeader", + "BadContentDispositionParam", + "parse_content_disposition", + "content_disposition_filename", +) + + +if TYPE_CHECKING: # pragma: no cover + from .client_reqrep import ClientResponse + + +class BadContentDispositionHeader(RuntimeWarning): + pass + + +class BadContentDispositionParam(RuntimeWarning): + pass + + +def parse_content_disposition( + header: Optional[str], +) -> Tuple[Optional[str], Dict[str, str]]: + def is_token(string: str) -> bool: + return bool(string) and TOKEN >= set(string) + + def is_quoted(string: str) -> bool: + return string[0] == string[-1] == '"' + + def is_rfc5987(string: str) -> bool: + return is_token(string) and string.count("'") == 2 + + def is_extended_param(string: str) -> bool: + return string.endswith("*") + + def is_continuous_param(string: str) -> bool: + pos = string.find("*") + 1 + if not pos: + return False + substring = string[pos:-1] if string.endswith("*") else string[pos:] + return substring.isdigit() + + def unescape(text: str, *, chars: str = "".join(map(re.escape, CHAR))) -> str: + return re.sub(f"\\\\([{chars}])", "\\1", text) + + if not header: + return None, {} + + disptype, *parts = header.split(";") + if not is_token(disptype): + warnings.warn(BadContentDispositionHeader(header)) + return None, {} + + params = {} # type: Dict[str, str] + while parts: + item = parts.pop(0) + + if "=" not in item: + warnings.warn(BadContentDispositionHeader(header)) + return None, {} + + key, value = item.split("=", 1) + key = key.lower().strip() + value = value.lstrip() + + if key in params: + warnings.warn(BadContentDispositionHeader(header)) + return None, {} + + if not is_token(key): + warnings.warn(BadContentDispositionParam(item)) + continue + + elif is_continuous_param(key): + if is_quoted(value): + value = unescape(value[1:-1]) + elif not is_token(value): + warnings.warn(BadContentDispositionParam(item)) + continue + + elif is_extended_param(key): + if is_rfc5987(value): + encoding, _, value = value.split("'", 2) + encoding = encoding or "utf-8" + else: + warnings.warn(BadContentDispositionParam(item)) + continue + + try: + value = unquote(value, encoding, "strict") + except UnicodeDecodeError: # pragma: nocover + warnings.warn(BadContentDispositionParam(item)) + continue + + else: + failed = True + if is_quoted(value): + failed = False + value = unescape(value[1:-1].lstrip("\\/")) + elif is_token(value): + failed = False + elif parts: + # maybe just ; in filename, in any case this is just + # one case fix, for proper fix we need to redesign parser + _value = f"{value};{parts[0]}" + if is_quoted(_value): + parts.pop(0) + value = unescape(_value[1:-1].lstrip("\\/")) + failed = False + + if failed: + warnings.warn(BadContentDispositionHeader(header)) + return None, {} + + params[key] = value + + return disptype.lower(), params + + +def content_disposition_filename( + params: Mapping[str, str], name: str = "filename" +) -> Optional[str]: + name_suf = "%s*" % name + if not params: + return None + elif name_suf in params: + return params[name_suf] + elif name in params: + return params[name] + else: + parts = [] + fnparams = sorted( + (key, value) for key, value in params.items() if key.startswith(name_suf) + ) + for num, (key, value) in enumerate(fnparams): + _, tail = key.split("*", 1) + if tail.endswith("*"): + tail = tail[:-1] + if tail == str(num): + parts.append(value) + else: + break + if not parts: + return None + value = "".join(parts) + if "'" in value: + encoding, _, value = value.split("'", 2) + encoding = encoding or "utf-8" + return unquote(value, encoding, "strict") + return value + + +class MultipartResponseWrapper: + """Wrapper around the MultipartReader. + + It takes care about + underlying connection and close it when it needs in. + """ + + def __init__( + self, + resp: "ClientResponse", + stream: "MultipartReader", + ) -> None: + self.resp = resp + self.stream = stream + + def __aiter__(self) -> "MultipartResponseWrapper": + return self + + async def __anext__( + self, + ) -> Union["MultipartReader", "BodyPartReader"]: + part = await self.next() + if part is None: + raise StopAsyncIteration + return part + + def at_eof(self) -> bool: + """Returns True when all response data had been read.""" + return self.resp.content.at_eof() + + async def next( + self, + ) -> Optional[Union["MultipartReader", "BodyPartReader"]]: + """Emits next multipart reader object.""" + item = await self.stream.next() + if self.stream.at_eof(): + await self.release() + return item + + async def release(self) -> None: + """Release the connection gracefully. + + All remaining content is read to the void. + """ + await self.resp.release() + + +class BodyPartReader: + """Multipart reader for single body part.""" + + chunk_size = 8192 + + def __init__( + self, boundary: bytes, headers: "CIMultiDictProxy[str]", content: StreamReader + ) -> None: + self.headers = headers + self._boundary = boundary + self._content = content + self._at_eof = False + length = self.headers.get(CONTENT_LENGTH, None) + self._length = int(length) if length is not None else None + self._read_bytes = 0 + # TODO: typeing.Deque is not supported by Python 3.5 + self._unread: Deque[bytes] = deque() + self._prev_chunk = None # type: Optional[bytes] + self._content_eof = 0 + self._cache = {} # type: Dict[str, Any] + + def __aiter__(self) -> AsyncIterator["BodyPartReader"]: + return self # type: ignore[return-value] + + async def __anext__(self) -> bytes: + part = await self.next() + if part is None: + raise StopAsyncIteration + return part + + async def next(self) -> Optional[bytes]: + item = await self.read() + if not item: + return None + return item + + async def read(self, *, decode: bool = False) -> bytes: + """Reads body part data. + + decode: Decodes data following by encoding + method from Content-Encoding header. If it missed + data remains untouched + """ + if self._at_eof: + return b"" + data = bytearray() + while not self._at_eof: + data.extend(await self.read_chunk(self.chunk_size)) + if decode: + return self.decode(data) + return data + + async def read_chunk(self, size: int = chunk_size) -> bytes: + """Reads body part content chunk of the specified size. + + size: chunk size + """ + if self._at_eof: + return b"" + if self._length: + chunk = await self._read_chunk_from_length(size) + else: + chunk = await self._read_chunk_from_stream(size) + + self._read_bytes += len(chunk) + if self._read_bytes == self._length: + self._at_eof = True + if self._at_eof: + clrf = await self._content.readline() + assert ( + b"\r\n" == clrf + ), "reader did not read all the data or it is malformed" + return chunk + + async def _read_chunk_from_length(self, size: int) -> bytes: + # Reads body part content chunk of the specified size. + # The body part must has Content-Length header with proper value. + assert self._length is not None, "Content-Length required for chunked read" + chunk_size = min(size, self._length - self._read_bytes) + chunk = await self._content.read(chunk_size) + return chunk + + async def _read_chunk_from_stream(self, size: int) -> bytes: + # Reads content chunk of body part with unknown length. + # The Content-Length header for body part is not necessary. + assert ( + size >= len(self._boundary) + 2 + ), "Chunk size must be greater or equal than boundary length + 2" + first_chunk = self._prev_chunk is None + if first_chunk: + self._prev_chunk = await self._content.read(size) + + chunk = await self._content.read(size) + self._content_eof += int(self._content.at_eof()) + assert self._content_eof < 3, "Reading after EOF" + assert self._prev_chunk is not None + window = self._prev_chunk + chunk + sub = b"\r\n" + self._boundary + if first_chunk: + idx = window.find(sub) + else: + idx = window.find(sub, max(0, len(self._prev_chunk) - len(sub))) + if idx >= 0: + # pushing boundary back to content + with warnings.catch_warnings(): + warnings.filterwarnings("ignore", category=DeprecationWarning) + self._content.unread_data(window[idx:]) + if size > idx: + self._prev_chunk = self._prev_chunk[:idx] + chunk = window[len(self._prev_chunk) : idx] + if not chunk: + self._at_eof = True + result = self._prev_chunk + self._prev_chunk = chunk + return result + + async def readline(self) -> bytes: + """Reads body part by line by line.""" + if self._at_eof: + return b"" + + if self._unread: + line = self._unread.popleft() + else: + line = await self._content.readline() + + if line.startswith(self._boundary): + # the very last boundary may not come with \r\n, + # so set single rules for everyone + sline = line.rstrip(b"\r\n") + boundary = self._boundary + last_boundary = self._boundary + b"--" + # ensure that we read exactly the boundary, not something alike + if sline == boundary or sline == last_boundary: + self._at_eof = True + self._unread.append(line) + return b"" + else: + next_line = await self._content.readline() + if next_line.startswith(self._boundary): + line = line[:-2] # strip CRLF but only once + self._unread.append(next_line) + + return line + + async def release(self) -> None: + """Like read(), but reads all the data to the void.""" + if self._at_eof: + return + while not self._at_eof: + await self.read_chunk(self.chunk_size) + + async def text(self, *, encoding: Optional[str] = None) -> str: + """Like read(), but assumes that body part contains text data.""" + data = await self.read(decode=True) + # see https://www.w3.org/TR/html5/forms.html#multipart/form-data-encoding-algorithm # NOQA + # and https://dvcs.w3.org/hg/xhr/raw-file/tip/Overview.html#dom-xmlhttprequest-send # NOQA + encoding = encoding or self.get_charset(default="utf-8") + return data.decode(encoding) + + async def json(self, *, encoding: Optional[str] = None) -> Optional[Dict[str, Any]]: + """Like read(), but assumes that body parts contains JSON data.""" + data = await self.read(decode=True) + if not data: + return None + encoding = encoding or self.get_charset(default="utf-8") + return cast(Dict[str, Any], json.loads(data.decode(encoding))) + + async def form(self, *, encoding: Optional[str] = None) -> List[Tuple[str, str]]: + """Like read(), but assumes that body parts contain form urlencoded data.""" + data = await self.read(decode=True) + if not data: + return [] + if encoding is not None: + real_encoding = encoding + else: + real_encoding = self.get_charset(default="utf-8") + return parse_qsl( + data.rstrip().decode(real_encoding), + keep_blank_values=True, + encoding=real_encoding, + ) + + def at_eof(self) -> bool: + """Returns True if the boundary was reached or False otherwise.""" + return self._at_eof + + def decode(self, data: bytes) -> bytes: + """Decodes data. + + Decoding is done according the specified Content-Encoding + or Content-Transfer-Encoding headers value. + """ + if CONTENT_TRANSFER_ENCODING in self.headers: + data = self._decode_content_transfer(data) + if CONTENT_ENCODING in self.headers: + return self._decode_content(data) + return data + + def _decode_content(self, data: bytes) -> bytes: + encoding = self.headers.get(CONTENT_ENCODING, "").lower() + + if encoding == "deflate": + return zlib.decompress(data, -zlib.MAX_WBITS) + elif encoding == "gzip": + return zlib.decompress(data, 16 + zlib.MAX_WBITS) + elif encoding == "identity": + return data + else: + raise RuntimeError(f"unknown content encoding: {encoding}") + + def _decode_content_transfer(self, data: bytes) -> bytes: + encoding = self.headers.get(CONTENT_TRANSFER_ENCODING, "").lower() + + if encoding == "base64": + return base64.b64decode(data) + elif encoding == "quoted-printable": + return binascii.a2b_qp(data) + elif encoding in ("binary", "8bit", "7bit"): + return data + else: + raise RuntimeError( + "unknown content transfer encoding: {}" "".format(encoding) + ) + + def get_charset(self, default: str) -> str: + """Returns charset parameter from Content-Type header or default.""" + ctype = self.headers.get(CONTENT_TYPE, "") + mimetype = parse_mimetype(ctype) + return mimetype.parameters.get("charset", default) + + @reify + def name(self) -> Optional[str]: + """Returns name specified in Content-Disposition header. + + If the header is missing or malformed, returns None. + """ + _, params = parse_content_disposition(self.headers.get(CONTENT_DISPOSITION)) + return content_disposition_filename(params, "name") + + @reify + def filename(self) -> Optional[str]: + """Returns filename specified in Content-Disposition header. + + Returns None if the header is missing or malformed. + """ + _, params = parse_content_disposition(self.headers.get(CONTENT_DISPOSITION)) + return content_disposition_filename(params, "filename") + + +@payload_type(BodyPartReader, order=Order.try_first) +class BodyPartReaderPayload(Payload): + def __init__(self, value: BodyPartReader, *args: Any, **kwargs: Any) -> None: + super().__init__(value, *args, **kwargs) + + params = {} # type: Dict[str, str] + if value.name is not None: + params["name"] = value.name + if value.filename is not None: + params["filename"] = value.filename + + if params: + self.set_content_disposition("attachment", True, **params) + + async def write(self, writer: Any) -> None: + field = self._value + chunk = await field.read_chunk(size=2 ** 16) + while chunk: + await writer.write(field.decode(chunk)) + chunk = await field.read_chunk(size=2 ** 16) + + +class MultipartReader: + """Multipart body reader.""" + + #: Response wrapper, used when multipart readers constructs from response. + response_wrapper_cls = MultipartResponseWrapper + #: Multipart reader class, used to handle multipart/* body parts. + #: None points to type(self) + multipart_reader_cls = None + #: Body part reader class for non multipart/* content types. + part_reader_cls = BodyPartReader + + def __init__(self, headers: Mapping[str, str], content: StreamReader) -> None: + self.headers = headers + self._boundary = ("--" + self._get_boundary()).encode() + self._content = content + self._last_part = ( + None + ) # type: Optional[Union['MultipartReader', BodyPartReader]] + self._at_eof = False + self._at_bof = True + self._unread = [] # type: List[bytes] + + def __aiter__( + self, + ) -> AsyncIterator["BodyPartReader"]: + return self # type: ignore[return-value] + + async def __anext__( + self, + ) -> Optional[Union["MultipartReader", BodyPartReader]]: + part = await self.next() + if part is None: + raise StopAsyncIteration + return part + + @classmethod + def from_response( + cls, + response: "ClientResponse", + ) -> MultipartResponseWrapper: + """Constructs reader instance from HTTP response. + + :param response: :class:`~aiohttp.client.ClientResponse` instance + """ + obj = cls.response_wrapper_cls( + response, cls(response.headers, response.content) + ) + return obj + + def at_eof(self) -> bool: + """Returns True if the final boundary was reached, false otherwise.""" + return self._at_eof + + async def next( + self, + ) -> Optional[Union["MultipartReader", BodyPartReader]]: + """Emits the next multipart body part.""" + # So, if we're at BOF, we need to skip till the boundary. + if self._at_eof: + return None + await self._maybe_release_last_part() + if self._at_bof: + await self._read_until_first_boundary() + self._at_bof = False + else: + await self._read_boundary() + if self._at_eof: # we just read the last boundary, nothing to do there + return None + self._last_part = await self.fetch_next_part() + return self._last_part + + async def release(self) -> None: + """Reads all the body parts to the void till the final boundary.""" + while not self._at_eof: + item = await self.next() + if item is None: + break + await item.release() + + async def fetch_next_part( + self, + ) -> Union["MultipartReader", BodyPartReader]: + """Returns the next body part reader.""" + headers = await self._read_headers() + return self._get_part_reader(headers) + + def _get_part_reader( + self, + headers: "CIMultiDictProxy[str]", + ) -> Union["MultipartReader", BodyPartReader]: + """Dispatches the response by the `Content-Type` header. + + Returns a suitable reader instance. + + :param dict headers: Response headers + """ + ctype = headers.get(CONTENT_TYPE, "") + mimetype = parse_mimetype(ctype) + + if mimetype.type == "multipart": + if self.multipart_reader_cls is None: + return type(self)(headers, self._content) + return self.multipart_reader_cls(headers, self._content) + else: + return self.part_reader_cls(self._boundary, headers, self._content) + + def _get_boundary(self) -> str: + mimetype = parse_mimetype(self.headers[CONTENT_TYPE]) + + assert mimetype.type == "multipart", "multipart/* content type expected" + + if "boundary" not in mimetype.parameters: + raise ValueError( + "boundary missed for Content-Type: %s" % self.headers[CONTENT_TYPE] + ) + + boundary = mimetype.parameters["boundary"] + if len(boundary) > 70: + raise ValueError("boundary %r is too long (70 chars max)" % boundary) + + return boundary + + async def _readline(self) -> bytes: + if self._unread: + return self._unread.pop() + return await self._content.readline() + + async def _read_until_first_boundary(self) -> None: + while True: + chunk = await self._readline() + if chunk == b"": + raise ValueError( + "Could not find starting boundary %r" % (self._boundary) + ) + chunk = chunk.rstrip() + if chunk == self._boundary: + return + elif chunk == self._boundary + b"--": + self._at_eof = True + return + + async def _read_boundary(self) -> None: + chunk = (await self._readline()).rstrip() + if chunk == self._boundary: + pass + elif chunk == self._boundary + b"--": + self._at_eof = True + epilogue = await self._readline() + next_line = await self._readline() + + # the epilogue is expected and then either the end of input or the + # parent multipart boundary, if the parent boundary is found then + # it should be marked as unread and handed to the parent for + # processing + if next_line[:2] == b"--": + self._unread.append(next_line) + # otherwise the request is likely missing an epilogue and both + # lines should be passed to the parent for processing + # (this handles the old behavior gracefully) + else: + self._unread.extend([next_line, epilogue]) + else: + raise ValueError(f"Invalid boundary {chunk!r}, expected {self._boundary!r}") + + async def _read_headers(self) -> "CIMultiDictProxy[str]": + lines = [b""] + while True: + chunk = await self._content.readline() + chunk = chunk.strip() + lines.append(chunk) + if not chunk: + break + parser = HeadersParser() + headers, raw_headers = parser.parse_headers(lines) + return headers + + async def _maybe_release_last_part(self) -> None: + """Ensures that the last read body part is read completely.""" + if self._last_part is not None: + if not self._last_part.at_eof(): + await self._last_part.release() + self._unread.extend(self._last_part._unread) + self._last_part = None + + +_Part = Tuple[Payload, str, str] + + +class MultipartWriter(Payload): + """Multipart body writer.""" + + def __init__(self, subtype: str = "mixed", boundary: Optional[str] = None) -> None: + boundary = boundary if boundary is not None else uuid.uuid4().hex + # The underlying Payload API demands a str (utf-8), not bytes, + # so we need to ensure we don't lose anything during conversion. + # As a result, require the boundary to be ASCII only. + # In both situations. + + try: + self._boundary = boundary.encode("ascii") + except UnicodeEncodeError: + raise ValueError("boundary should contain ASCII only chars") from None + ctype = f"multipart/{subtype}; boundary={self._boundary_value}" + + super().__init__(None, content_type=ctype) + + self._parts = [] # type: List[_Part] + + def __enter__(self) -> "MultipartWriter": + return self + + def __exit__( + self, + exc_type: Optional[Type[BaseException]], + exc_val: Optional[BaseException], + exc_tb: Optional[TracebackType], + ) -> None: + pass + + def __iter__(self) -> Iterator[_Part]: + return iter(self._parts) + + def __len__(self) -> int: + return len(self._parts) + + def __bool__(self) -> bool: + return True + + _valid_tchar_regex = re.compile(br"\A[!#$%&'*+\-.^_`|~\w]+\Z") + _invalid_qdtext_char_regex = re.compile(br"[\x00-\x08\x0A-\x1F\x7F]") + + @property + def _boundary_value(self) -> str: + """Wrap boundary parameter value in quotes, if necessary. + + Reads self.boundary and returns a unicode sting. + """ + # Refer to RFCs 7231, 7230, 5234. + # + # parameter = token "=" ( token / quoted-string ) + # token = 1*tchar + # quoted-string = DQUOTE *( qdtext / quoted-pair ) DQUOTE + # qdtext = HTAB / SP / %x21 / %x23-5B / %x5D-7E / obs-text + # obs-text = %x80-FF + # quoted-pair = "\" ( HTAB / SP / VCHAR / obs-text ) + # tchar = "!" / "#" / "$" / "%" / "&" / "'" / "*" + # / "+" / "-" / "." / "^" / "_" / "`" / "|" / "~" + # / DIGIT / ALPHA + # ; any VCHAR, except delimiters + # VCHAR = %x21-7E + value = self._boundary + if re.match(self._valid_tchar_regex, value): + return value.decode("ascii") # cannot fail + + if re.search(self._invalid_qdtext_char_regex, value): + raise ValueError("boundary value contains invalid characters") + + # escape %x5C and %x22 + quoted_value_content = value.replace(b"\\", b"\\\\") + quoted_value_content = quoted_value_content.replace(b'"', b'\\"') + + return '"' + quoted_value_content.decode("ascii") + '"' + + @property + def boundary(self) -> str: + return self._boundary.decode("ascii") + + def append(self, obj: Any, headers: Optional[MultiMapping[str]] = None) -> Payload: + if headers is None: + headers = CIMultiDict() + + if isinstance(obj, Payload): + obj.headers.update(headers) + return self.append_payload(obj) + else: + try: + payload = get_payload(obj, headers=headers) + except LookupError: + raise TypeError("Cannot create payload from %r" % obj) + else: + return self.append_payload(payload) + + def append_payload(self, payload: Payload) -> Payload: + """Adds a new body part to multipart writer.""" + # compression + encoding = payload.headers.get( + CONTENT_ENCODING, + "", + ).lower() # type: Optional[str] + if encoding and encoding not in ("deflate", "gzip", "identity"): + raise RuntimeError(f"unknown content encoding: {encoding}") + if encoding == "identity": + encoding = None + + # te encoding + te_encoding = payload.headers.get( + CONTENT_TRANSFER_ENCODING, + "", + ).lower() # type: Optional[str] + if te_encoding not in ("", "base64", "quoted-printable", "binary"): + raise RuntimeError( + "unknown content transfer encoding: {}" "".format(te_encoding) + ) + if te_encoding == "binary": + te_encoding = None + + # size + size = payload.size + if size is not None and not (encoding or te_encoding): + payload.headers[CONTENT_LENGTH] = str(size) + + self._parts.append((payload, encoding, te_encoding)) # type: ignore[arg-type] + return payload + + def append_json( + self, obj: Any, headers: Optional[MultiMapping[str]] = None + ) -> Payload: + """Helper to append JSON part.""" + if headers is None: + headers = CIMultiDict() + + return self.append_payload(JsonPayload(obj, headers=headers)) + + def append_form( + self, + obj: Union[Sequence[Tuple[str, str]], Mapping[str, str]], + headers: Optional[MultiMapping[str]] = None, + ) -> Payload: + """Helper to append form urlencoded part.""" + assert isinstance(obj, (Sequence, Mapping)) + + if headers is None: + headers = CIMultiDict() + + if isinstance(obj, Mapping): + obj = list(obj.items()) + data = urlencode(obj, doseq=True) + + return self.append_payload( + StringPayload( + data, headers=headers, content_type="application/x-www-form-urlencoded" + ) + ) + + @property + def size(self) -> Optional[int]: + """Size of the payload.""" + total = 0 + for part, encoding, te_encoding in self._parts: + if encoding or te_encoding or part.size is None: + return None + + total += int( + 2 + + len(self._boundary) + + 2 + + part.size # b'--'+self._boundary+b'\r\n' + + len(part._binary_headers) + + 2 # b'\r\n' + ) + + total += 2 + len(self._boundary) + 4 # b'--'+self._boundary+b'--\r\n' + return total + + async def write(self, writer: Any, close_boundary: bool = True) -> None: + """Write body.""" + for part, encoding, te_encoding in self._parts: + await writer.write(b"--" + self._boundary + b"\r\n") + await writer.write(part._binary_headers) + + if encoding or te_encoding: + w = MultipartPayloadWriter(writer) + if encoding: + w.enable_compression(encoding) + if te_encoding: + w.enable_encoding(te_encoding) + await part.write(w) # type: ignore[arg-type] + await w.write_eof() + else: + await part.write(writer) + + await writer.write(b"\r\n") + + if close_boundary: + await writer.write(b"--" + self._boundary + b"--\r\n") + + +class MultipartPayloadWriter: + def __init__(self, writer: Any) -> None: + self._writer = writer + self._encoding = None # type: Optional[str] + self._compress = None # type: Any + self._encoding_buffer = None # type: Optional[bytearray] + + def enable_encoding(self, encoding: str) -> None: + if encoding == "base64": + self._encoding = encoding + self._encoding_buffer = bytearray() + elif encoding == "quoted-printable": + self._encoding = "quoted-printable" + + def enable_compression( + self, encoding: str = "deflate", strategy: int = zlib.Z_DEFAULT_STRATEGY + ) -> None: + zlib_mode = 16 + zlib.MAX_WBITS if encoding == "gzip" else -zlib.MAX_WBITS + self._compress = zlib.compressobj(wbits=zlib_mode, strategy=strategy) + + async def write_eof(self) -> None: + if self._compress is not None: + chunk = self._compress.flush() + if chunk: + self._compress = None + await self.write(chunk) + + if self._encoding == "base64": + if self._encoding_buffer: + await self._writer.write(base64.b64encode(self._encoding_buffer)) + + async def write(self, chunk: bytes) -> None: + if self._compress is not None: + if chunk: + chunk = self._compress.compress(chunk) + if not chunk: + return + + if self._encoding == "base64": + buf = self._encoding_buffer + assert buf is not None + buf.extend(chunk) + + if buf: + div, mod = divmod(len(buf), 3) + enc_chunk, self._encoding_buffer = (buf[: div * 3], buf[div * 3 :]) + if enc_chunk: + b64chunk = base64.b64encode(enc_chunk) + await self._writer.write(b64chunk) + elif self._encoding == "quoted-printable": + await self._writer.write(binascii.b2a_qp(chunk)) + else: + await self._writer.write(chunk) diff --git a/contrib/python/aiohttp/aiohttp/payload.py b/contrib/python/aiohttp/aiohttp/payload.py new file mode 100644 index 0000000000..2ee90beea8 --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/payload.py @@ -0,0 +1,465 @@ +import asyncio +import enum +import io +import json +import mimetypes +import os +import warnings +from abc import ABC, abstractmethod +from itertools import chain +from typing import ( + IO, + TYPE_CHECKING, + Any, + ByteString, + Dict, + Iterable, + Optional, + TextIO, + Tuple, + Type, + Union, +) + +from multidict import CIMultiDict + +from . import hdrs +from .abc import AbstractStreamWriter +from .helpers import ( + PY_36, + content_disposition_header, + guess_filename, + parse_mimetype, + sentinel, +) +from .streams import StreamReader +from .typedefs import Final, JSONEncoder, _CIMultiDict + +__all__ = ( + "PAYLOAD_REGISTRY", + "get_payload", + "payload_type", + "Payload", + "BytesPayload", + "StringPayload", + "IOBasePayload", + "BytesIOPayload", + "BufferedReaderPayload", + "TextIOPayload", + "StringIOPayload", + "JsonPayload", + "AsyncIterablePayload", +) + +TOO_LARGE_BYTES_BODY: Final[int] = 2 ** 20 # 1 MB + +if TYPE_CHECKING: # pragma: no cover + from typing import List + + +class LookupError(Exception): + pass + + +class Order(str, enum.Enum): + normal = "normal" + try_first = "try_first" + try_last = "try_last" + + +def get_payload(data: Any, *args: Any, **kwargs: Any) -> "Payload": + return PAYLOAD_REGISTRY.get(data, *args, **kwargs) + + +def register_payload( + factory: Type["Payload"], type: Any, *, order: Order = Order.normal +) -> None: + PAYLOAD_REGISTRY.register(factory, type, order=order) + + +class payload_type: + def __init__(self, type: Any, *, order: Order = Order.normal) -> None: + self.type = type + self.order = order + + def __call__(self, factory: Type["Payload"]) -> Type["Payload"]: + register_payload(factory, self.type, order=self.order) + return factory + + +PayloadType = Type["Payload"] +_PayloadRegistryItem = Tuple[PayloadType, Any] + + +class PayloadRegistry: + """Payload registry. + + note: we need zope.interface for more efficient adapter search + """ + + def __init__(self) -> None: + self._first = [] # type: List[_PayloadRegistryItem] + self._normal = [] # type: List[_PayloadRegistryItem] + self._last = [] # type: List[_PayloadRegistryItem] + + def get( + self, + data: Any, + *args: Any, + _CHAIN: "Type[chain[_PayloadRegistryItem]]" = chain, + **kwargs: Any, + ) -> "Payload": + if isinstance(data, Payload): + return data + for factory, type in _CHAIN(self._first, self._normal, self._last): + if isinstance(data, type): + return factory(data, *args, **kwargs) + + raise LookupError() + + def register( + self, factory: PayloadType, type: Any, *, order: Order = Order.normal + ) -> None: + if order is Order.try_first: + self._first.append((factory, type)) + elif order is Order.normal: + self._normal.append((factory, type)) + elif order is Order.try_last: + self._last.append((factory, type)) + else: + raise ValueError(f"Unsupported order {order!r}") + + +class Payload(ABC): + + _default_content_type = "application/octet-stream" # type: str + _size = None # type: Optional[int] + + def __init__( + self, + value: Any, + headers: Optional[ + Union[_CIMultiDict, Dict[str, str], Iterable[Tuple[str, str]]] + ] = None, + content_type: Optional[str] = sentinel, + filename: Optional[str] = None, + encoding: Optional[str] = None, + **kwargs: Any, + ) -> None: + self._encoding = encoding + self._filename = filename + self._headers = CIMultiDict() # type: _CIMultiDict + self._value = value + if content_type is not sentinel and content_type is not None: + self._headers[hdrs.CONTENT_TYPE] = content_type + elif self._filename is not None: + content_type = mimetypes.guess_type(self._filename)[0] + if content_type is None: + content_type = self._default_content_type + self._headers[hdrs.CONTENT_TYPE] = content_type + else: + self._headers[hdrs.CONTENT_TYPE] = self._default_content_type + self._headers.update(headers or {}) + + @property + def size(self) -> Optional[int]: + """Size of the payload.""" + return self._size + + @property + def filename(self) -> Optional[str]: + """Filename of the payload.""" + return self._filename + + @property + def headers(self) -> _CIMultiDict: + """Custom item headers""" + return self._headers + + @property + def _binary_headers(self) -> bytes: + return ( + "".join([k + ": " + v + "\r\n" for k, v in self.headers.items()]).encode( + "utf-8" + ) + + b"\r\n" + ) + + @property + def encoding(self) -> Optional[str]: + """Payload encoding""" + return self._encoding + + @property + def content_type(self) -> str: + """Content type""" + return self._headers[hdrs.CONTENT_TYPE] + + def set_content_disposition( + self, + disptype: str, + quote_fields: bool = True, + _charset: str = "utf-8", + **params: Any, + ) -> None: + """Sets ``Content-Disposition`` header.""" + self._headers[hdrs.CONTENT_DISPOSITION] = content_disposition_header( + disptype, quote_fields=quote_fields, _charset=_charset, **params + ) + + @abstractmethod + async def write(self, writer: AbstractStreamWriter) -> None: + """Write payload. + + writer is an AbstractStreamWriter instance: + """ + + +class BytesPayload(Payload): + def __init__(self, value: ByteString, *args: Any, **kwargs: Any) -> None: + if not isinstance(value, (bytes, bytearray, memoryview)): + raise TypeError(f"value argument must be byte-ish, not {type(value)!r}") + + if "content_type" not in kwargs: + kwargs["content_type"] = "application/octet-stream" + + super().__init__(value, *args, **kwargs) + + if isinstance(value, memoryview): + self._size = value.nbytes + else: + self._size = len(value) + + if self._size > TOO_LARGE_BYTES_BODY: + if PY_36: + kwargs = {"source": self} + else: + kwargs = {} + warnings.warn( + "Sending a large body directly with raw bytes might" + " lock the event loop. You should probably pass an " + "io.BytesIO object instead", + ResourceWarning, + **kwargs, + ) + + async def write(self, writer: AbstractStreamWriter) -> None: + await writer.write(self._value) + + +class StringPayload(BytesPayload): + def __init__( + self, + value: str, + *args: Any, + encoding: Optional[str] = None, + content_type: Optional[str] = None, + **kwargs: Any, + ) -> None: + + if encoding is None: + if content_type is None: + real_encoding = "utf-8" + content_type = "text/plain; charset=utf-8" + else: + mimetype = parse_mimetype(content_type) + real_encoding = mimetype.parameters.get("charset", "utf-8") + else: + if content_type is None: + content_type = "text/plain; charset=%s" % encoding + real_encoding = encoding + + super().__init__( + value.encode(real_encoding), + encoding=real_encoding, + content_type=content_type, + *args, + **kwargs, + ) + + +class StringIOPayload(StringPayload): + def __init__(self, value: IO[str], *args: Any, **kwargs: Any) -> None: + super().__init__(value.read(), *args, **kwargs) + + +class IOBasePayload(Payload): + _value: IO[Any] + + def __init__( + self, value: IO[Any], disposition: str = "attachment", *args: Any, **kwargs: Any + ) -> None: + if "filename" not in kwargs: + kwargs["filename"] = guess_filename(value) + + super().__init__(value, *args, **kwargs) + + if self._filename is not None and disposition is not None: + if hdrs.CONTENT_DISPOSITION not in self.headers: + self.set_content_disposition(disposition, filename=self._filename) + + async def write(self, writer: AbstractStreamWriter) -> None: + loop = asyncio.get_event_loop() + try: + chunk = await loop.run_in_executor(None, self._value.read, 2 ** 16) + while chunk: + await writer.write(chunk) + chunk = await loop.run_in_executor(None, self._value.read, 2 ** 16) + finally: + await loop.run_in_executor(None, self._value.close) + + +class TextIOPayload(IOBasePayload): + _value: TextIO + + def __init__( + self, + value: TextIO, + *args: Any, + encoding: Optional[str] = None, + content_type: Optional[str] = None, + **kwargs: Any, + ) -> None: + + if encoding is None: + if content_type is None: + encoding = "utf-8" + content_type = "text/plain; charset=utf-8" + else: + mimetype = parse_mimetype(content_type) + encoding = mimetype.parameters.get("charset", "utf-8") + else: + if content_type is None: + content_type = "text/plain; charset=%s" % encoding + + super().__init__( + value, + content_type=content_type, + encoding=encoding, + *args, + **kwargs, + ) + + @property + def size(self) -> Optional[int]: + try: + return os.fstat(self._value.fileno()).st_size - self._value.tell() + except OSError: + return None + + async def write(self, writer: AbstractStreamWriter) -> None: + loop = asyncio.get_event_loop() + try: + chunk = await loop.run_in_executor(None, self._value.read, 2 ** 16) + while chunk: + data = ( + chunk.encode(encoding=self._encoding) + if self._encoding + else chunk.encode() + ) + await writer.write(data) + chunk = await loop.run_in_executor(None, self._value.read, 2 ** 16) + finally: + await loop.run_in_executor(None, self._value.close) + + +class BytesIOPayload(IOBasePayload): + @property + def size(self) -> int: + position = self._value.tell() + end = self._value.seek(0, os.SEEK_END) + self._value.seek(position) + return end - position + + +class BufferedReaderPayload(IOBasePayload): + @property + def size(self) -> Optional[int]: + try: + return os.fstat(self._value.fileno()).st_size - self._value.tell() + except OSError: + # data.fileno() is not supported, e.g. + # io.BufferedReader(io.BytesIO(b'data')) + return None + + +class JsonPayload(BytesPayload): + def __init__( + self, + value: Any, + encoding: str = "utf-8", + content_type: str = "application/json", + dumps: JSONEncoder = json.dumps, + *args: Any, + **kwargs: Any, + ) -> None: + + super().__init__( + dumps(value).encode(encoding), + content_type=content_type, + encoding=encoding, + *args, + **kwargs, + ) + + +if TYPE_CHECKING: # pragma: no cover + from typing import AsyncIterable, AsyncIterator + + _AsyncIterator = AsyncIterator[bytes] + _AsyncIterable = AsyncIterable[bytes] +else: + from collections.abc import AsyncIterable, AsyncIterator + + _AsyncIterator = AsyncIterator + _AsyncIterable = AsyncIterable + + +class AsyncIterablePayload(Payload): + + _iter = None # type: Optional[_AsyncIterator] + + def __init__(self, value: _AsyncIterable, *args: Any, **kwargs: Any) -> None: + if not isinstance(value, AsyncIterable): + raise TypeError( + "value argument must support " + "collections.abc.AsyncIterablebe interface, " + "got {!r}".format(type(value)) + ) + + if "content_type" not in kwargs: + kwargs["content_type"] = "application/octet-stream" + + super().__init__(value, *args, **kwargs) + + self._iter = value.__aiter__() + + async def write(self, writer: AbstractStreamWriter) -> None: + if self._iter: + try: + # iter is not None check prevents rare cases + # when the case iterable is used twice + while True: + chunk = await self._iter.__anext__() + await writer.write(chunk) + except StopAsyncIteration: + self._iter = None + + +class StreamReaderPayload(AsyncIterablePayload): + def __init__(self, value: StreamReader, *args: Any, **kwargs: Any) -> None: + super().__init__(value.iter_any(), *args, **kwargs) + + +PAYLOAD_REGISTRY = PayloadRegistry() +PAYLOAD_REGISTRY.register(BytesPayload, (bytes, bytearray, memoryview)) +PAYLOAD_REGISTRY.register(StringPayload, str) +PAYLOAD_REGISTRY.register(StringIOPayload, io.StringIO) +PAYLOAD_REGISTRY.register(TextIOPayload, io.TextIOBase) +PAYLOAD_REGISTRY.register(BytesIOPayload, io.BytesIO) +PAYLOAD_REGISTRY.register(BufferedReaderPayload, (io.BufferedReader, io.BufferedRandom)) +PAYLOAD_REGISTRY.register(IOBasePayload, io.IOBase) +PAYLOAD_REGISTRY.register(StreamReaderPayload, StreamReader) +# try_last for giving a chance to more specialized async interables like +# multidict.BodyPartReaderPayload override the default +PAYLOAD_REGISTRY.register(AsyncIterablePayload, AsyncIterable, order=Order.try_last) diff --git a/contrib/python/aiohttp/aiohttp/payload_streamer.py b/contrib/python/aiohttp/aiohttp/payload_streamer.py new file mode 100644 index 0000000000..9f8b8bc57c --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/payload_streamer.py @@ -0,0 +1,75 @@ +""" +Payload implemenation for coroutines as data provider. + +As a simple case, you can upload data from file:: + + @aiohttp.streamer + async def file_sender(writer, file_name=None): + with open(file_name, 'rb') as f: + chunk = f.read(2**16) + while chunk: + await writer.write(chunk) + + chunk = f.read(2**16) + +Then you can use `file_sender` like this: + + async with session.post('http://httpbin.org/post', + data=file_sender(file_name='huge_file')) as resp: + print(await resp.text()) + +..note:: Coroutine must accept `writer` as first argument + +""" + +import types +import warnings +from typing import Any, Awaitable, Callable, Dict, Tuple + +from .abc import AbstractStreamWriter +from .payload import Payload, payload_type + +__all__ = ("streamer",) + + +class _stream_wrapper: + def __init__( + self, + coro: Callable[..., Awaitable[None]], + args: Tuple[Any, ...], + kwargs: Dict[str, Any], + ) -> None: + self.coro = types.coroutine(coro) + self.args = args + self.kwargs = kwargs + + async def __call__(self, writer: AbstractStreamWriter) -> None: + await self.coro(writer, *self.args, **self.kwargs) # type: ignore[operator] + + +class streamer: + def __init__(self, coro: Callable[..., Awaitable[None]]) -> None: + warnings.warn( + "@streamer is deprecated, use async generators instead", + DeprecationWarning, + stacklevel=2, + ) + self.coro = coro + + def __call__(self, *args: Any, **kwargs: Any) -> _stream_wrapper: + return _stream_wrapper(self.coro, args, kwargs) + + +@payload_type(_stream_wrapper) +class StreamWrapperPayload(Payload): + async def write(self, writer: AbstractStreamWriter) -> None: + await self._value(writer) + + +@payload_type(streamer) +class StreamPayload(StreamWrapperPayload): + def __init__(self, value: Any, *args: Any, **kwargs: Any) -> None: + super().__init__(value(), *args, **kwargs) + + async def write(self, writer: AbstractStreamWriter) -> None: + await self._value(writer) diff --git a/contrib/python/aiohttp/aiohttp/py.typed b/contrib/python/aiohttp/aiohttp/py.typed new file mode 100644 index 0000000000..f5642f79f2 --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/py.typed @@ -0,0 +1 @@ +Marker diff --git a/contrib/python/aiohttp/aiohttp/pytest_plugin.py b/contrib/python/aiohttp/aiohttp/pytest_plugin.py new file mode 100644 index 0000000000..dd9a9f6179 --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/pytest_plugin.py @@ -0,0 +1,391 @@ +import asyncio +import contextlib +import warnings +from collections.abc import Callable +from typing import Any, Awaitable, Callable, Dict, Generator, Optional, Union + +import pytest + +from aiohttp.helpers import PY_37, isasyncgenfunction +from aiohttp.web import Application + +from .test_utils import ( + BaseTestServer, + RawTestServer, + TestClient, + TestServer, + loop_context, + setup_test_loop, + teardown_test_loop, + unused_port as _unused_port, +) + +try: + import uvloop +except ImportError: # pragma: no cover + uvloop = None + +try: + import tokio +except ImportError: # pragma: no cover + tokio = None + +AiohttpClient = Callable[[Union[Application, BaseTestServer]], Awaitable[TestClient]] + + +def pytest_addoption(parser): # type: ignore[no-untyped-def] + parser.addoption( + "--aiohttp-fast", + action="store_true", + default=False, + help="run tests faster by disabling extra checks", + ) + parser.addoption( + "--aiohttp-loop", + action="store", + default="pyloop", + help="run tests with specific loop: pyloop, uvloop, tokio or all", + ) + parser.addoption( + "--aiohttp-enable-loop-debug", + action="store_true", + default=False, + help="enable event loop debug mode", + ) + + +def pytest_fixture_setup(fixturedef): # type: ignore[no-untyped-def] + """Set up pytest fixture. + + Allow fixtures to be coroutines. Run coroutine fixtures in an event loop. + """ + func = fixturedef.func + + if isasyncgenfunction(func): + # async generator fixture + is_async_gen = True + elif asyncio.iscoroutinefunction(func): + # regular async fixture + is_async_gen = False + else: + # not an async fixture, nothing to do + return + + strip_request = False + if "request" not in fixturedef.argnames: + fixturedef.argnames += ("request",) + strip_request = True + + def wrapper(*args, **kwargs): # type: ignore[no-untyped-def] + request = kwargs["request"] + if strip_request: + del kwargs["request"] + + # if neither the fixture nor the test use the 'loop' fixture, + # 'getfixturevalue' will fail because the test is not parameterized + # (this can be removed someday if 'loop' is no longer parameterized) + if "loop" not in request.fixturenames: + raise Exception( + "Asynchronous fixtures must depend on the 'loop' fixture or " + "be used in tests depending from it." + ) + + _loop = request.getfixturevalue("loop") + + if is_async_gen: + # for async generators, we need to advance the generator once, + # then advance it again in a finalizer + gen = func(*args, **kwargs) + + def finalizer(): # type: ignore[no-untyped-def] + try: + return _loop.run_until_complete(gen.__anext__()) + except StopAsyncIteration: + pass + + request.addfinalizer(finalizer) + return _loop.run_until_complete(gen.__anext__()) + else: + return _loop.run_until_complete(func(*args, **kwargs)) + + fixturedef.func = wrapper + + +@pytest.fixture +def fast(request): # type: ignore[no-untyped-def] + """--fast config option""" + return request.config.getoption("--aiohttp-fast") + + +@pytest.fixture +def loop_debug(request): # type: ignore[no-untyped-def] + """--enable-loop-debug config option""" + return request.config.getoption("--aiohttp-enable-loop-debug") + + +@contextlib.contextmanager +def _runtime_warning_context(): # type: ignore[no-untyped-def] + """Context manager which checks for RuntimeWarnings. + + This exists specifically to + avoid "coroutine 'X' was never awaited" warnings being missed. + + If RuntimeWarnings occur in the context a RuntimeError is raised. + """ + with warnings.catch_warnings(record=True) as _warnings: + yield + rw = [ + "{w.filename}:{w.lineno}:{w.message}".format(w=w) + for w in _warnings + if w.category == RuntimeWarning + ] + if rw: + raise RuntimeError( + "{} Runtime Warning{},\n{}".format( + len(rw), "" if len(rw) == 1 else "s", "\n".join(rw) + ) + ) + + +@contextlib.contextmanager +def _passthrough_loop_context(loop, fast=False): # type: ignore[no-untyped-def] + """Passthrough loop context. + + Sets up and tears down a loop unless one is passed in via the loop + argument when it's passed straight through. + """ + if loop: + # loop already exists, pass it straight through + yield loop + else: + # this shadows loop_context's standard behavior + loop = setup_test_loop() + yield loop + teardown_test_loop(loop, fast=fast) + + +def pytest_pycollect_makeitem(collector, name, obj): # type: ignore[no-untyped-def] + """Fix pytest collecting for coroutines.""" + if collector.funcnamefilter(name) and asyncio.iscoroutinefunction(obj): + return list(collector._genfunctions(name, obj)) + + +def pytest_pyfunc_call(pyfuncitem): # type: ignore[no-untyped-def] + """Run coroutines in an event loop instead of a normal function call.""" + fast = pyfuncitem.config.getoption("--aiohttp-fast") + if asyncio.iscoroutinefunction(pyfuncitem.function): + existing_loop = pyfuncitem.funcargs.get( + "proactor_loop" + ) or pyfuncitem.funcargs.get("loop", None) + with _runtime_warning_context(): + with _passthrough_loop_context(existing_loop, fast=fast) as _loop: + testargs = { + arg: pyfuncitem.funcargs[arg] + for arg in pyfuncitem._fixtureinfo.argnames + } + _loop.run_until_complete(pyfuncitem.obj(**testargs)) + + return True + + +def pytest_generate_tests(metafunc): # type: ignore[no-untyped-def] + if "loop_factory" not in metafunc.fixturenames: + return + + loops = metafunc.config.option.aiohttp_loop + avail_factories = {"pyloop": asyncio.DefaultEventLoopPolicy} + + if uvloop is not None: # pragma: no cover + avail_factories["uvloop"] = uvloop.EventLoopPolicy + + if tokio is not None: # pragma: no cover + avail_factories["tokio"] = tokio.EventLoopPolicy + + if loops == "all": + loops = "pyloop,uvloop?,tokio?" + + factories = {} # type: ignore[var-annotated] + for name in loops.split(","): + required = not name.endswith("?") + name = name.strip(" ?") + if name not in avail_factories: # pragma: no cover + if required: + raise ValueError( + "Unknown loop '%s', available loops: %s" + % (name, list(factories.keys())) + ) + else: + continue + factories[name] = avail_factories[name] + metafunc.parametrize( + "loop_factory", list(factories.values()), ids=list(factories.keys()) + ) + + +@pytest.fixture +def loop(loop_factory, fast, loop_debug): # type: ignore[no-untyped-def] + """Return an instance of the event loop.""" + policy = loop_factory() + asyncio.set_event_loop_policy(policy) + with loop_context(fast=fast) as _loop: + if loop_debug: + _loop.set_debug(True) # pragma: no cover + asyncio.set_event_loop(_loop) + yield _loop + + +@pytest.fixture +def proactor_loop(): # type: ignore[no-untyped-def] + if not PY_37: + policy = asyncio.get_event_loop_policy() + policy._loop_factory = asyncio.ProactorEventLoop # type: ignore[attr-defined] + else: + policy = asyncio.WindowsProactorEventLoopPolicy() # type: ignore[attr-defined] + asyncio.set_event_loop_policy(policy) + + with loop_context(policy.new_event_loop) as _loop: + asyncio.set_event_loop(_loop) + yield _loop + + +@pytest.fixture +def unused_port(aiohttp_unused_port): # type: ignore[no-untyped-def] # pragma: no cover + warnings.warn( + "Deprecated, use aiohttp_unused_port fixture instead", + DeprecationWarning, + stacklevel=2, + ) + return aiohttp_unused_port + + +@pytest.fixture +def aiohttp_unused_port(): # type: ignore[no-untyped-def] + """Return a port that is unused on the current host.""" + return _unused_port + + +@pytest.fixture +def aiohttp_server(loop): # type: ignore[no-untyped-def] + """Factory to create a TestServer instance, given an app. + + aiohttp_server(app, **kwargs) + """ + servers = [] + + async def go(app, *, port=None, **kwargs): # type: ignore[no-untyped-def] + server = TestServer(app, port=port) + await server.start_server(loop=loop, **kwargs) + servers.append(server) + return server + + yield go + + async def finalize() -> None: + while servers: + await servers.pop().close() + + loop.run_until_complete(finalize()) + + +@pytest.fixture +def test_server(aiohttp_server): # type: ignore[no-untyped-def] # pragma: no cover + warnings.warn( + "Deprecated, use aiohttp_server fixture instead", + DeprecationWarning, + stacklevel=2, + ) + return aiohttp_server + + +@pytest.fixture +def aiohttp_raw_server(loop): # type: ignore[no-untyped-def] + """Factory to create a RawTestServer instance, given a web handler. + + aiohttp_raw_server(handler, **kwargs) + """ + servers = [] + + async def go(handler, *, port=None, **kwargs): # type: ignore[no-untyped-def] + server = RawTestServer(handler, port=port) + await server.start_server(loop=loop, **kwargs) + servers.append(server) + return server + + yield go + + async def finalize() -> None: + while servers: + await servers.pop().close() + + loop.run_until_complete(finalize()) + + +@pytest.fixture +def raw_test_server( # type: ignore[no-untyped-def] # pragma: no cover + aiohttp_raw_server, +): + warnings.warn( + "Deprecated, use aiohttp_raw_server fixture instead", + DeprecationWarning, + stacklevel=2, + ) + return aiohttp_raw_server + + +@pytest.fixture +def aiohttp_client( + loop: asyncio.AbstractEventLoop, +) -> Generator[AiohttpClient, None, None]: + """Factory to create a TestClient instance. + + aiohttp_client(app, **kwargs) + aiohttp_client(server, **kwargs) + aiohttp_client(raw_server, **kwargs) + """ + clients = [] + + async def go( + __param: Union[Application, BaseTestServer], + *args: Any, + server_kwargs: Optional[Dict[str, Any]] = None, + **kwargs: Any + ) -> TestClient: + + if isinstance(__param, Callable) and not isinstance( # type: ignore[arg-type] + __param, (Application, BaseTestServer) + ): + __param = __param(loop, *args, **kwargs) + kwargs = {} + else: + assert not args, "args should be empty" + + if isinstance(__param, Application): + server_kwargs = server_kwargs or {} + server = TestServer(__param, loop=loop, **server_kwargs) + client = TestClient(server, loop=loop, **kwargs) + elif isinstance(__param, BaseTestServer): + client = TestClient(__param, loop=loop, **kwargs) + else: + raise ValueError("Unknown argument type: %r" % type(__param)) + + await client.start_server() + clients.append(client) + return client + + yield go + + async def finalize() -> None: + while clients: + await clients.pop().close() + + loop.run_until_complete(finalize()) + + +@pytest.fixture +def test_client(aiohttp_client): # type: ignore[no-untyped-def] # pragma: no cover + warnings.warn( + "Deprecated, use aiohttp_client fixture instead", + DeprecationWarning, + stacklevel=2, + ) + return aiohttp_client diff --git a/contrib/python/aiohttp/aiohttp/resolver.py b/contrib/python/aiohttp/aiohttp/resolver.py new file mode 100644 index 0000000000..6668fa80ec --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/resolver.py @@ -0,0 +1,160 @@ +import asyncio +import socket +from typing import Any, Dict, List, Optional, Type, Union + +from .abc import AbstractResolver +from .helpers import get_running_loop + +__all__ = ("ThreadedResolver", "AsyncResolver", "DefaultResolver") + +try: + import aiodns + + # aiodns_default = hasattr(aiodns.DNSResolver, 'gethostbyname') +except ImportError: # pragma: no cover + aiodns = None + +aiodns_default = False + + +class ThreadedResolver(AbstractResolver): + """Threaded resolver. + + Uses an Executor for synchronous getaddrinfo() calls. + concurrent.futures.ThreadPoolExecutor is used by default. + """ + + def __init__(self, loop: Optional[asyncio.AbstractEventLoop] = None) -> None: + self._loop = get_running_loop(loop) + + async def resolve( + self, hostname: str, port: int = 0, family: int = socket.AF_INET + ) -> List[Dict[str, Any]]: + infos = await self._loop.getaddrinfo( + hostname, + port, + type=socket.SOCK_STREAM, + family=family, + # flags=socket.AI_ADDRCONFIG, + ) + + hosts = [] + for family, _, proto, _, address in infos: + if family == socket.AF_INET6: + if len(address) < 3: + # IPv6 is not supported by Python build, + # or IPv6 is not enabled in the host + continue + if address[3]: # type: ignore[misc] + # This is essential for link-local IPv6 addresses. + # LL IPv6 is a VERY rare case. Strictly speaking, we should use + # getnameinfo() unconditionally, but performance makes sense. + host, _port = socket.getnameinfo( + address, socket.NI_NUMERICHOST | socket.NI_NUMERICSERV + ) + port = int(_port) + else: + host, port = address[:2] + else: # IPv4 + assert family == socket.AF_INET + host, port = address # type: ignore[misc] + hosts.append( + { + "hostname": hostname, + "host": host, + "port": port, + "family": family, + "proto": proto, + "flags": socket.AI_NUMERICHOST | socket.AI_NUMERICSERV, + } + ) + + return hosts + + async def close(self) -> None: + pass + + +class AsyncResolver(AbstractResolver): + """Use the `aiodns` package to make asynchronous DNS lookups""" + + def __init__( + self, + loop: Optional[asyncio.AbstractEventLoop] = None, + *args: Any, + **kwargs: Any + ) -> None: + if aiodns is None: + raise RuntimeError("Resolver requires aiodns library") + + self._loop = get_running_loop(loop) + self._resolver = aiodns.DNSResolver(*args, loop=loop, **kwargs) + + if not hasattr(self._resolver, "gethostbyname"): + # aiodns 1.1 is not available, fallback to DNSResolver.query + self.resolve = self._resolve_with_query # type: ignore + + async def resolve( + self, host: str, port: int = 0, family: int = socket.AF_INET + ) -> List[Dict[str, Any]]: + try: + resp = await self._resolver.gethostbyname(host, family) + except aiodns.error.DNSError as exc: + msg = exc.args[1] if len(exc.args) >= 1 else "DNS lookup failed" + raise OSError(msg) from exc + hosts = [] + for address in resp.addresses: + hosts.append( + { + "hostname": host, + "host": address, + "port": port, + "family": family, + "proto": 0, + "flags": socket.AI_NUMERICHOST | socket.AI_NUMERICSERV, + } + ) + + if not hosts: + raise OSError("DNS lookup failed") + + return hosts + + async def _resolve_with_query( + self, host: str, port: int = 0, family: int = socket.AF_INET + ) -> List[Dict[str, Any]]: + if family == socket.AF_INET6: + qtype = "AAAA" + else: + qtype = "A" + + try: + resp = await self._resolver.query(host, qtype) + except aiodns.error.DNSError as exc: + msg = exc.args[1] if len(exc.args) >= 1 else "DNS lookup failed" + raise OSError(msg) from exc + + hosts = [] + for rr in resp: + hosts.append( + { + "hostname": host, + "host": rr.host, + "port": port, + "family": family, + "proto": 0, + "flags": socket.AI_NUMERICHOST, + } + ) + + if not hosts: + raise OSError("DNS lookup failed") + + return hosts + + async def close(self) -> None: + self._resolver.cancel() + + +_DefaultType = Type[Union[AsyncResolver, ThreadedResolver]] +DefaultResolver: _DefaultType = AsyncResolver if aiodns_default else ThreadedResolver diff --git a/contrib/python/aiohttp/aiohttp/streams.py b/contrib/python/aiohttp/aiohttp/streams.py new file mode 100644 index 0000000000..055848877e --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/streams.py @@ -0,0 +1,660 @@ +import asyncio +import collections +import warnings +from typing import Awaitable, Callable, Deque, Generic, List, Optional, Tuple, TypeVar + +from .base_protocol import BaseProtocol +from .helpers import BaseTimerContext, set_exception, set_result +from .log import internal_logger +from .typedefs import Final + +__all__ = ( + "EMPTY_PAYLOAD", + "EofStream", + "StreamReader", + "DataQueue", + "FlowControlDataQueue", +) + +_T = TypeVar("_T") + + +class EofStream(Exception): + """eof stream indication.""" + + +class AsyncStreamIterator(Generic[_T]): + def __init__(self, read_func: Callable[[], Awaitable[_T]]) -> None: + self.read_func = read_func + + def __aiter__(self) -> "AsyncStreamIterator[_T]": + return self + + async def __anext__(self) -> _T: + try: + rv = await self.read_func() + except EofStream: + raise StopAsyncIteration + if rv == b"": + raise StopAsyncIteration + return rv + + +class ChunkTupleAsyncStreamIterator: + def __init__(self, stream: "StreamReader") -> None: + self._stream = stream + + def __aiter__(self) -> "ChunkTupleAsyncStreamIterator": + return self + + async def __anext__(self) -> Tuple[bytes, bool]: + rv = await self._stream.readchunk() + if rv == (b"", False): + raise StopAsyncIteration + return rv + + +class AsyncStreamReaderMixin: + def __aiter__(self) -> AsyncStreamIterator[bytes]: + return AsyncStreamIterator(self.readline) # type: ignore[attr-defined] + + def iter_chunked(self, n: int) -> AsyncStreamIterator[bytes]: + """Returns an asynchronous iterator that yields chunks of size n. + + Python-3.5 available for Python 3.5+ only + """ + return AsyncStreamIterator( + lambda: self.read(n) # type: ignore[attr-defined,no-any-return] + ) + + def iter_any(self) -> AsyncStreamIterator[bytes]: + """Yield all available data as soon as it is received. + + Python-3.5 available for Python 3.5+ only + """ + return AsyncStreamIterator(self.readany) # type: ignore[attr-defined] + + def iter_chunks(self) -> ChunkTupleAsyncStreamIterator: + """Yield chunks of data as they are received by the server. + + The yielded objects are tuples + of (bytes, bool) as returned by the StreamReader.readchunk method. + + Python-3.5 available for Python 3.5+ only + """ + return ChunkTupleAsyncStreamIterator(self) # type: ignore[arg-type] + + +class StreamReader(AsyncStreamReaderMixin): + """An enhancement of asyncio.StreamReader. + + Supports asynchronous iteration by line, chunk or as available:: + + async for line in reader: + ... + async for chunk in reader.iter_chunked(1024): + ... + async for slice in reader.iter_any(): + ... + + """ + + total_bytes = 0 + + def __init__( + self, + protocol: BaseProtocol, + limit: int, + *, + timer: Optional[BaseTimerContext] = None, + loop: Optional[asyncio.AbstractEventLoop] = None, + ) -> None: + self._protocol = protocol + self._low_water = limit + self._high_water = limit * 2 + if loop is None: + loop = asyncio.get_event_loop() + self._loop = loop + self._size = 0 + self._cursor = 0 + self._http_chunk_splits = None # type: Optional[List[int]] + self._buffer = collections.deque() # type: Deque[bytes] + self._buffer_offset = 0 + self._eof = False + self._waiter = None # type: Optional[asyncio.Future[None]] + self._eof_waiter = None # type: Optional[asyncio.Future[None]] + self._exception = None # type: Optional[BaseException] + self._timer = timer + self._eof_callbacks = [] # type: List[Callable[[], None]] + + def __repr__(self) -> str: + info = [self.__class__.__name__] + if self._size: + info.append("%d bytes" % self._size) + if self._eof: + info.append("eof") + if self._low_water != 2 ** 16: # default limit + info.append("low=%d high=%d" % (self._low_water, self._high_water)) + if self._waiter: + info.append("w=%r" % self._waiter) + if self._exception: + info.append("e=%r" % self._exception) + return "<%s>" % " ".join(info) + + def get_read_buffer_limits(self) -> Tuple[int, int]: + return (self._low_water, self._high_water) + + def exception(self) -> Optional[BaseException]: + return self._exception + + def set_exception(self, exc: BaseException) -> None: + self._exception = exc + self._eof_callbacks.clear() + + waiter = self._waiter + if waiter is not None: + self._waiter = None + set_exception(waiter, exc) + + waiter = self._eof_waiter + if waiter is not None: + self._eof_waiter = None + set_exception(waiter, exc) + + def on_eof(self, callback: Callable[[], None]) -> None: + if self._eof: + try: + callback() + except Exception: + internal_logger.exception("Exception in eof callback") + else: + self._eof_callbacks.append(callback) + + def feed_eof(self) -> None: + self._eof = True + + waiter = self._waiter + if waiter is not None: + self._waiter = None + set_result(waiter, None) + + waiter = self._eof_waiter + if waiter is not None: + self._eof_waiter = None + set_result(waiter, None) + + for cb in self._eof_callbacks: + try: + cb() + except Exception: + internal_logger.exception("Exception in eof callback") + + self._eof_callbacks.clear() + + def is_eof(self) -> bool: + """Return True if 'feed_eof' was called.""" + return self._eof + + def at_eof(self) -> bool: + """Return True if the buffer is empty and 'feed_eof' was called.""" + return self._eof and not self._buffer + + async def wait_eof(self) -> None: + if self._eof: + return + + assert self._eof_waiter is None + self._eof_waiter = self._loop.create_future() + try: + await self._eof_waiter + finally: + self._eof_waiter = None + + def unread_data(self, data: bytes) -> None: + """rollback reading some data from stream, inserting it to buffer head.""" + warnings.warn( + "unread_data() is deprecated " + "and will be removed in future releases (#3260)", + DeprecationWarning, + stacklevel=2, + ) + if not data: + return + + if self._buffer_offset: + self._buffer[0] = self._buffer[0][self._buffer_offset :] + self._buffer_offset = 0 + self._size += len(data) + self._cursor -= len(data) + self._buffer.appendleft(data) + self._eof_counter = 0 + + # TODO: size is ignored, remove the param later + def feed_data(self, data: bytes, size: int = 0) -> None: + assert not self._eof, "feed_data after feed_eof" + + if not data: + return + + self._size += len(data) + self._buffer.append(data) + self.total_bytes += len(data) + + waiter = self._waiter + if waiter is not None: + self._waiter = None + set_result(waiter, None) + + if self._size > self._high_water and not self._protocol._reading_paused: + self._protocol.pause_reading() + + def begin_http_chunk_receiving(self) -> None: + if self._http_chunk_splits is None: + if self.total_bytes: + raise RuntimeError( + "Called begin_http_chunk_receiving when" "some data was already fed" + ) + self._http_chunk_splits = [] + + def end_http_chunk_receiving(self) -> None: + if self._http_chunk_splits is None: + raise RuntimeError( + "Called end_chunk_receiving without calling " + "begin_chunk_receiving first" + ) + + # self._http_chunk_splits contains logical byte offsets from start of + # the body transfer. Each offset is the offset of the end of a chunk. + # "Logical" means bytes, accessible for a user. + # If no chunks containig logical data were received, current position + # is difinitely zero. + pos = self._http_chunk_splits[-1] if self._http_chunk_splits else 0 + + if self.total_bytes == pos: + # We should not add empty chunks here. So we check for that. + # Note, when chunked + gzip is used, we can receive a chunk + # of compressed data, but that data may not be enough for gzip FSM + # to yield any uncompressed data. That's why current position may + # not change after receiving a chunk. + return + + self._http_chunk_splits.append(self.total_bytes) + + # wake up readchunk when end of http chunk received + waiter = self._waiter + if waiter is not None: + self._waiter = None + set_result(waiter, None) + + async def _wait(self, func_name: str) -> None: + # StreamReader uses a future to link the protocol feed_data() method + # to a read coroutine. Running two read coroutines at the same time + # would have an unexpected behaviour. It would not possible to know + # which coroutine would get the next data. + if self._waiter is not None: + raise RuntimeError( + "%s() called while another coroutine is " + "already waiting for incoming data" % func_name + ) + + waiter = self._waiter = self._loop.create_future() + try: + if self._timer: + with self._timer: + await waiter + else: + await waiter + finally: + self._waiter = None + + async def readline(self) -> bytes: + return await self.readuntil() + + async def readuntil(self, separator: bytes = b"\n") -> bytes: + seplen = len(separator) + if seplen == 0: + raise ValueError("Separator should be at least one-byte string") + + if self._exception is not None: + raise self._exception + + chunk = b"" + chunk_size = 0 + not_enough = True + + while not_enough: + while self._buffer and not_enough: + offset = self._buffer_offset + ichar = self._buffer[0].find(separator, offset) + 1 + # Read from current offset to found separator or to the end. + data = self._read_nowait_chunk(ichar - offset if ichar else -1) + chunk += data + chunk_size += len(data) + if ichar: + not_enough = False + + if chunk_size > self._high_water: + raise ValueError("Chunk too big") + + if self._eof: + break + + if not_enough: + await self._wait("readuntil") + + return chunk + + async def read(self, n: int = -1) -> bytes: + if self._exception is not None: + raise self._exception + + # migration problem; with DataQueue you have to catch + # EofStream exception, so common way is to run payload.read() inside + # infinite loop. what can cause real infinite loop with StreamReader + # lets keep this code one major release. + if __debug__: + if self._eof and not self._buffer: + self._eof_counter = getattr(self, "_eof_counter", 0) + 1 + if self._eof_counter > 5: + internal_logger.warning( + "Multiple access to StreamReader in eof state, " + "might be infinite loop.", + stack_info=True, + ) + + if not n: + return b"" + + if n < 0: + # This used to just loop creating a new waiter hoping to + # collect everything in self._buffer, but that would + # deadlock if the subprocess sends more than self.limit + # bytes. So just call self.readany() until EOF. + blocks = [] + while True: + block = await self.readany() + if not block: + break + blocks.append(block) + return b"".join(blocks) + + # TODO: should be `if` instead of `while` + # because waiter maybe triggered on chunk end, + # without feeding any data + while not self._buffer and not self._eof: + await self._wait("read") + + return self._read_nowait(n) + + async def readany(self) -> bytes: + if self._exception is not None: + raise self._exception + + # TODO: should be `if` instead of `while` + # because waiter maybe triggered on chunk end, + # without feeding any data + while not self._buffer and not self._eof: + await self._wait("readany") + + return self._read_nowait(-1) + + async def readchunk(self) -> Tuple[bytes, bool]: + """Returns a tuple of (data, end_of_http_chunk). + + When chunked transfer + encoding is used, end_of_http_chunk is a boolean indicating if the end + of the data corresponds to the end of a HTTP chunk , otherwise it is + always False. + """ + while True: + if self._exception is not None: + raise self._exception + + while self._http_chunk_splits: + pos = self._http_chunk_splits.pop(0) + if pos == self._cursor: + return (b"", True) + if pos > self._cursor: + return (self._read_nowait(pos - self._cursor), True) + internal_logger.warning( + "Skipping HTTP chunk end due to data " + "consumption beyond chunk boundary" + ) + + if self._buffer: + return (self._read_nowait_chunk(-1), False) + # return (self._read_nowait(-1), False) + + if self._eof: + # Special case for signifying EOF. + # (b'', True) is not a final return value actually. + return (b"", False) + + await self._wait("readchunk") + + async def readexactly(self, n: int) -> bytes: + if self._exception is not None: + raise self._exception + + blocks = [] # type: List[bytes] + while n > 0: + block = await self.read(n) + if not block: + partial = b"".join(blocks) + raise asyncio.IncompleteReadError(partial, len(partial) + n) + blocks.append(block) + n -= len(block) + + return b"".join(blocks) + + def read_nowait(self, n: int = -1) -> bytes: + # default was changed to be consistent with .read(-1) + # + # I believe the most users don't know about the method and + # they are not affected. + if self._exception is not None: + raise self._exception + + if self._waiter and not self._waiter.done(): + raise RuntimeError( + "Called while some coroutine is waiting for incoming data." + ) + + return self._read_nowait(n) + + def _read_nowait_chunk(self, n: int) -> bytes: + first_buffer = self._buffer[0] + offset = self._buffer_offset + if n != -1 and len(first_buffer) - offset > n: + data = first_buffer[offset : offset + n] + self._buffer_offset += n + + elif offset: + self._buffer.popleft() + data = first_buffer[offset:] + self._buffer_offset = 0 + + else: + data = self._buffer.popleft() + + self._size -= len(data) + self._cursor += len(data) + + chunk_splits = self._http_chunk_splits + # Prevent memory leak: drop useless chunk splits + while chunk_splits and chunk_splits[0] < self._cursor: + chunk_splits.pop(0) + + if self._size < self._low_water and self._protocol._reading_paused: + self._protocol.resume_reading() + return data + + def _read_nowait(self, n: int) -> bytes: + """Read not more than n bytes, or whole buffer if n == -1""" + chunks = [] + + while self._buffer: + chunk = self._read_nowait_chunk(n) + chunks.append(chunk) + if n != -1: + n -= len(chunk) + if n == 0: + break + + return b"".join(chunks) if chunks else b"" + + +class EmptyStreamReader(StreamReader): # lgtm [py/missing-call-to-init] + def __init__(self) -> None: + pass + + def exception(self) -> Optional[BaseException]: + return None + + def set_exception(self, exc: BaseException) -> None: + pass + + def on_eof(self, callback: Callable[[], None]) -> None: + try: + callback() + except Exception: + internal_logger.exception("Exception in eof callback") + + def feed_eof(self) -> None: + pass + + def is_eof(self) -> bool: + return True + + def at_eof(self) -> bool: + return True + + async def wait_eof(self) -> None: + return + + def feed_data(self, data: bytes, n: int = 0) -> None: + pass + + async def readline(self) -> bytes: + return b"" + + async def read(self, n: int = -1) -> bytes: + return b"" + + # TODO add async def readuntil + + async def readany(self) -> bytes: + return b"" + + async def readchunk(self) -> Tuple[bytes, bool]: + return (b"", True) + + async def readexactly(self, n: int) -> bytes: + raise asyncio.IncompleteReadError(b"", n) + + def read_nowait(self, n: int = -1) -> bytes: + return b"" + + +EMPTY_PAYLOAD: Final[StreamReader] = EmptyStreamReader() + + +class DataQueue(Generic[_T]): + """DataQueue is a general-purpose blocking queue with one reader.""" + + def __init__(self, loop: asyncio.AbstractEventLoop) -> None: + self._loop = loop + self._eof = False + self._waiter = None # type: Optional[asyncio.Future[None]] + self._exception = None # type: Optional[BaseException] + self._size = 0 + self._buffer = collections.deque() # type: Deque[Tuple[_T, int]] + + def __len__(self) -> int: + return len(self._buffer) + + def is_eof(self) -> bool: + return self._eof + + def at_eof(self) -> bool: + return self._eof and not self._buffer + + def exception(self) -> Optional[BaseException]: + return self._exception + + def set_exception(self, exc: BaseException) -> None: + self._eof = True + self._exception = exc + + waiter = self._waiter + if waiter is not None: + self._waiter = None + set_exception(waiter, exc) + + def feed_data(self, data: _T, size: int = 0) -> None: + self._size += size + self._buffer.append((data, size)) + + waiter = self._waiter + if waiter is not None: + self._waiter = None + set_result(waiter, None) + + def feed_eof(self) -> None: + self._eof = True + + waiter = self._waiter + if waiter is not None: + self._waiter = None + set_result(waiter, None) + + async def read(self) -> _T: + if not self._buffer and not self._eof: + assert not self._waiter + self._waiter = self._loop.create_future() + try: + await self._waiter + except (asyncio.CancelledError, asyncio.TimeoutError): + self._waiter = None + raise + + if self._buffer: + data, size = self._buffer.popleft() + self._size -= size + return data + else: + if self._exception is not None: + raise self._exception + else: + raise EofStream + + def __aiter__(self) -> AsyncStreamIterator[_T]: + return AsyncStreamIterator(self.read) + + +class FlowControlDataQueue(DataQueue[_T]): + """FlowControlDataQueue resumes and pauses an underlying stream. + + It is a destination for parsed data. + """ + + def __init__( + self, protocol: BaseProtocol, limit: int, *, loop: asyncio.AbstractEventLoop + ) -> None: + super().__init__(loop=loop) + + self._protocol = protocol + self._limit = limit * 2 + + def feed_data(self, data: _T, size: int = 0) -> None: + super().feed_data(data, size) + + if self._size > self._limit and not self._protocol._reading_paused: + self._protocol.pause_reading() + + async def read(self) -> _T: + try: + return await super().read() + finally: + if self._size < self._limit and self._protocol._reading_paused: + self._protocol.resume_reading() diff --git a/contrib/python/aiohttp/aiohttp/tcp_helpers.py b/contrib/python/aiohttp/aiohttp/tcp_helpers.py new file mode 100644 index 0000000000..0e1dbf1655 --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/tcp_helpers.py @@ -0,0 +1,38 @@ +"""Helper methods to tune a TCP connection""" + +import asyncio +import socket +from contextlib import suppress +from typing import Optional # noqa + +__all__ = ("tcp_keepalive", "tcp_nodelay") + + +if hasattr(socket, "SO_KEEPALIVE"): + + def tcp_keepalive(transport: asyncio.Transport) -> None: + sock = transport.get_extra_info("socket") + if sock is not None: + sock.setsockopt(socket.SOL_SOCKET, socket.SO_KEEPALIVE, 1) + + +else: + + def tcp_keepalive(transport: asyncio.Transport) -> None: # pragma: no cover + pass + + +def tcp_nodelay(transport: asyncio.Transport, value: bool) -> None: + sock = transport.get_extra_info("socket") + + if sock is None: + return + + if sock.family not in (socket.AF_INET, socket.AF_INET6): + return + + value = bool(value) + + # socket may be closed already, on windows OSError get raised + with suppress(OSError): + sock.setsockopt(socket.IPPROTO_TCP, socket.TCP_NODELAY, value) diff --git a/contrib/python/aiohttp/aiohttp/test_utils.py b/contrib/python/aiohttp/aiohttp/test_utils.py new file mode 100644 index 0000000000..361dae486c --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/test_utils.py @@ -0,0 +1,698 @@ +"""Utilities shared by tests.""" + +import asyncio +import contextlib +import gc +import inspect +import ipaddress +import os +import socket +import sys +import warnings +from abc import ABC, abstractmethod +from types import TracebackType +from typing import ( + TYPE_CHECKING, + Any, + Callable, + Iterator, + List, + Optional, + Type, + Union, + cast, +) +from unittest import mock + +from aiosignal import Signal +from multidict import CIMultiDict, CIMultiDictProxy +from yarl import URL + +import aiohttp +from aiohttp.client import _RequestContextManager, _WSRequestContextManager + +from . import ClientSession, hdrs +from .abc import AbstractCookieJar +from .client_reqrep import ClientResponse +from .client_ws import ClientWebSocketResponse +from .helpers import PY_38, sentinel +from .http import HttpVersion, RawRequestMessage +from .web import ( + Application, + AppRunner, + BaseRunner, + Request, + Server, + ServerRunner, + SockSite, + UrlMappingMatchInfo, +) +from .web_protocol import _RequestHandler + +if TYPE_CHECKING: # pragma: no cover + from ssl import SSLContext +else: + SSLContext = None + +if PY_38: + from unittest import IsolatedAsyncioTestCase as TestCase +else: + from asynctest import TestCase # type: ignore[no-redef] + +REUSE_ADDRESS = os.name == "posix" and sys.platform != "cygwin" + + +def get_unused_port_socket( + host: str, family: socket.AddressFamily = socket.AF_INET +) -> socket.socket: + return get_port_socket(host, 0, family) + + +def get_port_socket( + host: str, port: int, family: socket.AddressFamily +) -> socket.socket: + s = socket.socket(family, socket.SOCK_STREAM) + if REUSE_ADDRESS: + # Windows has different semantics for SO_REUSEADDR, + # so don't set it. Ref: + # https://docs.microsoft.com/en-us/windows/win32/winsock/using-so-reuseaddr-and-so-exclusiveaddruse + s.setsockopt(socket.SOL_SOCKET, socket.SO_REUSEADDR, 1) + s.bind((host, port)) + return s + + +def unused_port() -> int: + """Return a port that is unused on the current host.""" + with socket.socket(socket.AF_INET, socket.SOCK_STREAM) as s: + s.bind(("127.0.0.1", 0)) + return cast(int, s.getsockname()[1]) + + +class BaseTestServer(ABC): + __test__ = False + + def __init__( + self, + *, + scheme: Union[str, object] = sentinel, + loop: Optional[asyncio.AbstractEventLoop] = None, + host: str = "127.0.0.1", + port: Optional[int] = None, + skip_url_asserts: bool = False, + socket_factory: Callable[ + [str, int, socket.AddressFamily], socket.socket + ] = get_port_socket, + **kwargs: Any, + ) -> None: + self._loop = loop + self.runner = None # type: Optional[BaseRunner] + self._root = None # type: Optional[URL] + self.host = host + self.port = port + self._closed = False + self.scheme = scheme + self.skip_url_asserts = skip_url_asserts + self.socket_factory = socket_factory + + async def start_server( + self, loop: Optional[asyncio.AbstractEventLoop] = None, **kwargs: Any + ) -> None: + if self.runner: + return + self._loop = loop + self._ssl = kwargs.pop("ssl", None) + self.runner = await self._make_runner(**kwargs) + await self.runner.setup() + if not self.port: + self.port = 0 + try: + version = ipaddress.ip_address(self.host).version + except ValueError: + version = 4 + family = socket.AF_INET6 if version == 6 else socket.AF_INET + _sock = self.socket_factory(self.host, self.port, family) + self.host, self.port = _sock.getsockname()[:2] + site = SockSite(self.runner, sock=_sock, ssl_context=self._ssl) + await site.start() + server = site._server + assert server is not None + sockets = server.sockets + assert sockets is not None + self.port = sockets[0].getsockname()[1] + if self.scheme is sentinel: + if self._ssl: + scheme = "https" + else: + scheme = "http" + self.scheme = scheme + self._root = URL(f"{self.scheme}://{self.host}:{self.port}") + + @abstractmethod # pragma: no cover + async def _make_runner(self, **kwargs: Any) -> BaseRunner: + pass + + def make_url(self, path: str) -> URL: + assert self._root is not None + url = URL(path) + if not self.skip_url_asserts: + assert not url.is_absolute() + return self._root.join(url) + else: + return URL(str(self._root) + path) + + @property + def started(self) -> bool: + return self.runner is not None + + @property + def closed(self) -> bool: + return self._closed + + @property + def handler(self) -> Server: + # for backward compatibility + # web.Server instance + runner = self.runner + assert runner is not None + assert runner.server is not None + return runner.server + + async def close(self) -> None: + """Close all fixtures created by the test client. + + After that point, the TestClient is no longer usable. + + This is an idempotent function: running close multiple times + will not have any additional effects. + + close is also run when the object is garbage collected, and on + exit when used as a context manager. + + """ + if self.started and not self.closed: + assert self.runner is not None + await self.runner.cleanup() + self._root = None + self.port = None + self._closed = True + + def __enter__(self) -> None: + raise TypeError("Use async with instead") + + def __exit__( + self, + exc_type: Optional[Type[BaseException]], + exc_value: Optional[BaseException], + traceback: Optional[TracebackType], + ) -> None: + # __exit__ should exist in pair with __enter__ but never executed + pass # pragma: no cover + + async def __aenter__(self) -> "BaseTestServer": + await self.start_server(loop=self._loop) + return self + + async def __aexit__( + self, + exc_type: Optional[Type[BaseException]], + exc_value: Optional[BaseException], + traceback: Optional[TracebackType], + ) -> None: + await self.close() + + +class TestServer(BaseTestServer): + def __init__( + self, + app: Application, + *, + scheme: Union[str, object] = sentinel, + host: str = "127.0.0.1", + port: Optional[int] = None, + **kwargs: Any, + ): + self.app = app + super().__init__(scheme=scheme, host=host, port=port, **kwargs) + + async def _make_runner(self, **kwargs: Any) -> BaseRunner: + return AppRunner(self.app, **kwargs) + + +class RawTestServer(BaseTestServer): + def __init__( + self, + handler: _RequestHandler, + *, + scheme: Union[str, object] = sentinel, + host: str = "127.0.0.1", + port: Optional[int] = None, + **kwargs: Any, + ) -> None: + self._handler = handler + super().__init__(scheme=scheme, host=host, port=port, **kwargs) + + async def _make_runner(self, debug: bool = True, **kwargs: Any) -> ServerRunner: + srv = Server(self._handler, loop=self._loop, debug=debug, **kwargs) + return ServerRunner(srv, debug=debug, **kwargs) + + +class TestClient: + """ + A test client implementation. + + To write functional tests for aiohttp based servers. + + """ + + __test__ = False + + def __init__( + self, + server: BaseTestServer, + *, + cookie_jar: Optional[AbstractCookieJar] = None, + loop: Optional[asyncio.AbstractEventLoop] = None, + **kwargs: Any, + ) -> None: + if not isinstance(server, BaseTestServer): + raise TypeError( + "server must be TestServer " "instance, found type: %r" % type(server) + ) + self._server = server + self._loop = loop + if cookie_jar is None: + cookie_jar = aiohttp.CookieJar(unsafe=True, loop=loop) + self._session = ClientSession(loop=loop, cookie_jar=cookie_jar, **kwargs) + self._closed = False + self._responses = [] # type: List[ClientResponse] + self._websockets = [] # type: List[ClientWebSocketResponse] + + async def start_server(self) -> None: + await self._server.start_server(loop=self._loop) + + @property + def host(self) -> str: + return self._server.host + + @property + def port(self) -> Optional[int]: + return self._server.port + + @property + def server(self) -> BaseTestServer: + return self._server + + @property + def app(self) -> Optional[Application]: + return cast(Optional[Application], getattr(self._server, "app", None)) + + @property + def session(self) -> ClientSession: + """An internal aiohttp.ClientSession. + + Unlike the methods on the TestClient, client session requests + do not automatically include the host in the url queried, and + will require an absolute path to the resource. + + """ + return self._session + + def make_url(self, path: str) -> URL: + return self._server.make_url(path) + + async def _request(self, method: str, path: str, **kwargs: Any) -> ClientResponse: + resp = await self._session.request(method, self.make_url(path), **kwargs) + # save it to close later + self._responses.append(resp) + return resp + + def request(self, method: str, path: str, **kwargs: Any) -> _RequestContextManager: + """Routes a request to tested http server. + + The interface is identical to aiohttp.ClientSession.request, + except the loop kwarg is overridden by the instance used by the + test server. + + """ + return _RequestContextManager(self._request(method, path, **kwargs)) + + def get(self, path: str, **kwargs: Any) -> _RequestContextManager: + """Perform an HTTP GET request.""" + return _RequestContextManager(self._request(hdrs.METH_GET, path, **kwargs)) + + def post(self, path: str, **kwargs: Any) -> _RequestContextManager: + """Perform an HTTP POST request.""" + return _RequestContextManager(self._request(hdrs.METH_POST, path, **kwargs)) + + def options(self, path: str, **kwargs: Any) -> _RequestContextManager: + """Perform an HTTP OPTIONS request.""" + return _RequestContextManager(self._request(hdrs.METH_OPTIONS, path, **kwargs)) + + def head(self, path: str, **kwargs: Any) -> _RequestContextManager: + """Perform an HTTP HEAD request.""" + return _RequestContextManager(self._request(hdrs.METH_HEAD, path, **kwargs)) + + def put(self, path: str, **kwargs: Any) -> _RequestContextManager: + """Perform an HTTP PUT request.""" + return _RequestContextManager(self._request(hdrs.METH_PUT, path, **kwargs)) + + def patch(self, path: str, **kwargs: Any) -> _RequestContextManager: + """Perform an HTTP PATCH request.""" + return _RequestContextManager(self._request(hdrs.METH_PATCH, path, **kwargs)) + + def delete(self, path: str, **kwargs: Any) -> _RequestContextManager: + """Perform an HTTP PATCH request.""" + return _RequestContextManager(self._request(hdrs.METH_DELETE, path, **kwargs)) + + def ws_connect(self, path: str, **kwargs: Any) -> _WSRequestContextManager: + """Initiate websocket connection. + + The api corresponds to aiohttp.ClientSession.ws_connect. + + """ + return _WSRequestContextManager(self._ws_connect(path, **kwargs)) + + async def _ws_connect(self, path: str, **kwargs: Any) -> ClientWebSocketResponse: + ws = await self._session.ws_connect(self.make_url(path), **kwargs) + self._websockets.append(ws) + return ws + + async def close(self) -> None: + """Close all fixtures created by the test client. + + After that point, the TestClient is no longer usable. + + This is an idempotent function: running close multiple times + will not have any additional effects. + + close is also run on exit when used as a(n) (asynchronous) + context manager. + + """ + if not self._closed: + for resp in self._responses: + resp.close() + for ws in self._websockets: + await ws.close() + await self._session.close() + await self._server.close() + self._closed = True + + def __enter__(self) -> None: + raise TypeError("Use async with instead") + + def __exit__( + self, + exc_type: Optional[Type[BaseException]], + exc: Optional[BaseException], + tb: Optional[TracebackType], + ) -> None: + # __exit__ should exist in pair with __enter__ but never executed + pass # pragma: no cover + + async def __aenter__(self) -> "TestClient": + await self.start_server() + return self + + async def __aexit__( + self, + exc_type: Optional[Type[BaseException]], + exc: Optional[BaseException], + tb: Optional[TracebackType], + ) -> None: + await self.close() + + +class AioHTTPTestCase(TestCase): + """A base class to allow for unittest web applications using aiohttp. + + Provides the following: + + * self.client (aiohttp.test_utils.TestClient): an aiohttp test client. + * self.loop (asyncio.BaseEventLoop): the event loop in which the + application and server are running. + * self.app (aiohttp.web.Application): the application returned by + self.get_application() + + Note that the TestClient's methods are asynchronous: you have to + execute function on the test client using asynchronous methods. + """ + + async def get_application(self) -> Application: + """Get application. + + This method should be overridden + to return the aiohttp.web.Application + object to test. + """ + return self.get_app() + + def get_app(self) -> Application: + """Obsolete method used to constructing web application. + + Use .get_application() coroutine instead. + """ + raise RuntimeError("Did you forget to define get_application()?") + + def setUp(self) -> None: + try: + self.loop = asyncio.get_running_loop() + except (AttributeError, RuntimeError): # AttributeError->py36 + self.loop = asyncio.get_event_loop_policy().get_event_loop() + + self.loop.run_until_complete(self.setUpAsync()) + + async def setUpAsync(self) -> None: + self.app = await self.get_application() + self.server = await self.get_server(self.app) + self.client = await self.get_client(self.server) + + await self.client.start_server() + + def tearDown(self) -> None: + self.loop.run_until_complete(self.tearDownAsync()) + + async def tearDownAsync(self) -> None: + await self.client.close() + + async def get_server(self, app: Application) -> TestServer: + """Return a TestServer instance.""" + return TestServer(app, loop=self.loop) + + async def get_client(self, server: TestServer) -> TestClient: + """Return a TestClient instance.""" + return TestClient(server, loop=self.loop) + + +def unittest_run_loop(func: Any, *args: Any, **kwargs: Any) -> Any: + """ + A decorator dedicated to use with asynchronous AioHTTPTestCase test methods. + + In 3.8+, this does nothing. + """ + warnings.warn( + "Decorator `@unittest_run_loop` is no longer needed in aiohttp 3.8+", + DeprecationWarning, + stacklevel=2, + ) + return func + + +_LOOP_FACTORY = Callable[[], asyncio.AbstractEventLoop] + + +@contextlib.contextmanager +def loop_context( + loop_factory: _LOOP_FACTORY = asyncio.new_event_loop, fast: bool = False +) -> Iterator[asyncio.AbstractEventLoop]: + """A contextmanager that creates an event_loop, for test purposes. + + Handles the creation and cleanup of a test loop. + """ + loop = setup_test_loop(loop_factory) + yield loop + teardown_test_loop(loop, fast=fast) + + +def setup_test_loop( + loop_factory: _LOOP_FACTORY = asyncio.new_event_loop, +) -> asyncio.AbstractEventLoop: + """Create and return an asyncio.BaseEventLoop instance. + + The caller should also call teardown_test_loop, + once they are done with the loop. + """ + loop = loop_factory() + try: + module = loop.__class__.__module__ + skip_watcher = "uvloop" in module + except AttributeError: # pragma: no cover + # Just in case + skip_watcher = True + asyncio.set_event_loop(loop) + if sys.platform != "win32" and not skip_watcher: + policy = asyncio.get_event_loop_policy() + watcher: asyncio.AbstractChildWatcher + try: # Python >= 3.8 + # Refs: + # * https://github.com/pytest-dev/pytest-xdist/issues/620 + # * https://stackoverflow.com/a/58614689/595220 + # * https://bugs.python.org/issue35621 + # * https://github.com/python/cpython/pull/14344 + watcher = asyncio.ThreadedChildWatcher() + except AttributeError: # Python < 3.8 + watcher = asyncio.SafeChildWatcher() + watcher.attach_loop(loop) + with contextlib.suppress(NotImplementedError): + policy.set_child_watcher(watcher) + return loop + + +def teardown_test_loop(loop: asyncio.AbstractEventLoop, fast: bool = False) -> None: + """Teardown and cleanup an event_loop created by setup_test_loop.""" + closed = loop.is_closed() + if not closed: + loop.call_soon(loop.stop) + loop.run_forever() + loop.close() + + if not fast: + gc.collect() + + asyncio.set_event_loop(None) + + +def _create_app_mock() -> mock.MagicMock: + def get_dict(app: Any, key: str) -> Any: + return app.__app_dict[key] + + def set_dict(app: Any, key: str, value: Any) -> None: + app.__app_dict[key] = value + + app = mock.MagicMock() + app.__app_dict = {} + app.__getitem__ = get_dict + app.__setitem__ = set_dict + + app._debug = False + app.on_response_prepare = Signal(app) + app.on_response_prepare.freeze() + return app + + +def _create_transport(sslcontext: Optional[SSLContext] = None) -> mock.Mock: + transport = mock.Mock() + + def get_extra_info(key: str) -> Optional[SSLContext]: + if key == "sslcontext": + return sslcontext + else: + return None + + transport.get_extra_info.side_effect = get_extra_info + return transport + + +def make_mocked_request( + method: str, + path: str, + headers: Any = None, + *, + match_info: Any = sentinel, + version: HttpVersion = HttpVersion(1, 1), + closing: bool = False, + app: Any = None, + writer: Any = sentinel, + protocol: Any = sentinel, + transport: Any = sentinel, + payload: Any = sentinel, + sslcontext: Optional[SSLContext] = None, + client_max_size: int = 1024 ** 2, + loop: Any = ..., +) -> Request: + """Creates mocked web.Request testing purposes. + + Useful in unit tests, when spinning full web server is overkill or + specific conditions and errors are hard to trigger. + """ + task = mock.Mock() + if loop is ...: + loop = mock.Mock() + loop.create_future.return_value = () + + if version < HttpVersion(1, 1): + closing = True + + if headers: + headers = CIMultiDictProxy(CIMultiDict(headers)) + raw_hdrs = tuple( + (k.encode("utf-8"), v.encode("utf-8")) for k, v in headers.items() + ) + else: + headers = CIMultiDictProxy(CIMultiDict()) + raw_hdrs = () + + chunked = "chunked" in headers.get(hdrs.TRANSFER_ENCODING, "").lower() + + message = RawRequestMessage( + method, + path, + version, + headers, + raw_hdrs, + closing, + None, + False, + chunked, + URL(path), + ) + if app is None: + app = _create_app_mock() + + if transport is sentinel: + transport = _create_transport(sslcontext) + + if protocol is sentinel: + protocol = mock.Mock() + protocol.transport = transport + + if writer is sentinel: + writer = mock.Mock() + writer.write_headers = make_mocked_coro(None) + writer.write = make_mocked_coro(None) + writer.write_eof = make_mocked_coro(None) + writer.drain = make_mocked_coro(None) + writer.transport = transport + + protocol.transport = transport + protocol.writer = writer + + if payload is sentinel: + payload = mock.Mock() + + req = Request( + message, payload, protocol, writer, task, loop, client_max_size=client_max_size + ) + + match_info = UrlMappingMatchInfo( + {} if match_info is sentinel else match_info, mock.Mock() + ) + match_info.add_app(app) + req._match_info = match_info + + return req + + +def make_mocked_coro( + return_value: Any = sentinel, raise_exception: Any = sentinel +) -> Any: + """Creates a coroutine mock.""" + + async def mock_coro(*args: Any, **kwargs: Any) -> Any: + if raise_exception is not sentinel: + raise raise_exception + if not inspect.isawaitable(return_value): + return return_value + await return_value + + return mock.Mock(wraps=mock_coro) diff --git a/contrib/python/aiohttp/aiohttp/tracing.py b/contrib/python/aiohttp/aiohttp/tracing.py new file mode 100644 index 0000000000..0e118a3997 --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/tracing.py @@ -0,0 +1,472 @@ +from types import SimpleNamespace +from typing import TYPE_CHECKING, Awaitable, Optional, Type, TypeVar + +import attr +from aiosignal import Signal +from multidict import CIMultiDict +from yarl import URL + +from .client_reqrep import ClientResponse + +if TYPE_CHECKING: # pragma: no cover + from .client import ClientSession + from .typedefs import Protocol + + _ParamT_contra = TypeVar("_ParamT_contra", contravariant=True) + + class _SignalCallback(Protocol[_ParamT_contra]): + def __call__( + self, + __client_session: ClientSession, + __trace_config_ctx: SimpleNamespace, + __params: _ParamT_contra, + ) -> Awaitable[None]: + ... + + +__all__ = ( + "TraceConfig", + "TraceRequestStartParams", + "TraceRequestEndParams", + "TraceRequestExceptionParams", + "TraceConnectionQueuedStartParams", + "TraceConnectionQueuedEndParams", + "TraceConnectionCreateStartParams", + "TraceConnectionCreateEndParams", + "TraceConnectionReuseconnParams", + "TraceDnsResolveHostStartParams", + "TraceDnsResolveHostEndParams", + "TraceDnsCacheHitParams", + "TraceDnsCacheMissParams", + "TraceRequestRedirectParams", + "TraceRequestChunkSentParams", + "TraceResponseChunkReceivedParams", + "TraceRequestHeadersSentParams", +) + + +class TraceConfig: + """First-class used to trace requests launched via ClientSession objects.""" + + def __init__( + self, trace_config_ctx_factory: Type[SimpleNamespace] = SimpleNamespace + ) -> None: + self._on_request_start = Signal( + self + ) # type: Signal[_SignalCallback[TraceRequestStartParams]] + self._on_request_chunk_sent = Signal( + self + ) # type: Signal[_SignalCallback[TraceRequestChunkSentParams]] + self._on_response_chunk_received = Signal( + self + ) # type: Signal[_SignalCallback[TraceResponseChunkReceivedParams]] + self._on_request_end = Signal( + self + ) # type: Signal[_SignalCallback[TraceRequestEndParams]] + self._on_request_exception = Signal( + self + ) # type: Signal[_SignalCallback[TraceRequestExceptionParams]] + self._on_request_redirect = Signal( + self + ) # type: Signal[_SignalCallback[TraceRequestRedirectParams]] + self._on_connection_queued_start = Signal( + self + ) # type: Signal[_SignalCallback[TraceConnectionQueuedStartParams]] + self._on_connection_queued_end = Signal( + self + ) # type: Signal[_SignalCallback[TraceConnectionQueuedEndParams]] + self._on_connection_create_start = Signal( + self + ) # type: Signal[_SignalCallback[TraceConnectionCreateStartParams]] + self._on_connection_create_end = Signal( + self + ) # type: Signal[_SignalCallback[TraceConnectionCreateEndParams]] + self._on_connection_reuseconn = Signal( + self + ) # type: Signal[_SignalCallback[TraceConnectionReuseconnParams]] + self._on_dns_resolvehost_start = Signal( + self + ) # type: Signal[_SignalCallback[TraceDnsResolveHostStartParams]] + self._on_dns_resolvehost_end = Signal( + self + ) # type: Signal[_SignalCallback[TraceDnsResolveHostEndParams]] + self._on_dns_cache_hit = Signal( + self + ) # type: Signal[_SignalCallback[TraceDnsCacheHitParams]] + self._on_dns_cache_miss = Signal( + self + ) # type: Signal[_SignalCallback[TraceDnsCacheMissParams]] + self._on_request_headers_sent = Signal( + self + ) # type: Signal[_SignalCallback[TraceRequestHeadersSentParams]] + + self._trace_config_ctx_factory = trace_config_ctx_factory + + def trace_config_ctx( + self, trace_request_ctx: Optional[SimpleNamespace] = None + ) -> SimpleNamespace: + """Return a new trace_config_ctx instance""" + return self._trace_config_ctx_factory(trace_request_ctx=trace_request_ctx) + + def freeze(self) -> None: + self._on_request_start.freeze() + self._on_request_chunk_sent.freeze() + self._on_response_chunk_received.freeze() + self._on_request_end.freeze() + self._on_request_exception.freeze() + self._on_request_redirect.freeze() + self._on_connection_queued_start.freeze() + self._on_connection_queued_end.freeze() + self._on_connection_create_start.freeze() + self._on_connection_create_end.freeze() + self._on_connection_reuseconn.freeze() + self._on_dns_resolvehost_start.freeze() + self._on_dns_resolvehost_end.freeze() + self._on_dns_cache_hit.freeze() + self._on_dns_cache_miss.freeze() + self._on_request_headers_sent.freeze() + + @property + def on_request_start(self) -> "Signal[_SignalCallback[TraceRequestStartParams]]": + return self._on_request_start + + @property + def on_request_chunk_sent( + self, + ) -> "Signal[_SignalCallback[TraceRequestChunkSentParams]]": + return self._on_request_chunk_sent + + @property + def on_response_chunk_received( + self, + ) -> "Signal[_SignalCallback[TraceResponseChunkReceivedParams]]": + return self._on_response_chunk_received + + @property + def on_request_end(self) -> "Signal[_SignalCallback[TraceRequestEndParams]]": + return self._on_request_end + + @property + def on_request_exception( + self, + ) -> "Signal[_SignalCallback[TraceRequestExceptionParams]]": + return self._on_request_exception + + @property + def on_request_redirect( + self, + ) -> "Signal[_SignalCallback[TraceRequestRedirectParams]]": + return self._on_request_redirect + + @property + def on_connection_queued_start( + self, + ) -> "Signal[_SignalCallback[TraceConnectionQueuedStartParams]]": + return self._on_connection_queued_start + + @property + def on_connection_queued_end( + self, + ) -> "Signal[_SignalCallback[TraceConnectionQueuedEndParams]]": + return self._on_connection_queued_end + + @property + def on_connection_create_start( + self, + ) -> "Signal[_SignalCallback[TraceConnectionCreateStartParams]]": + return self._on_connection_create_start + + @property + def on_connection_create_end( + self, + ) -> "Signal[_SignalCallback[TraceConnectionCreateEndParams]]": + return self._on_connection_create_end + + @property + def on_connection_reuseconn( + self, + ) -> "Signal[_SignalCallback[TraceConnectionReuseconnParams]]": + return self._on_connection_reuseconn + + @property + def on_dns_resolvehost_start( + self, + ) -> "Signal[_SignalCallback[TraceDnsResolveHostStartParams]]": + return self._on_dns_resolvehost_start + + @property + def on_dns_resolvehost_end( + self, + ) -> "Signal[_SignalCallback[TraceDnsResolveHostEndParams]]": + return self._on_dns_resolvehost_end + + @property + def on_dns_cache_hit(self) -> "Signal[_SignalCallback[TraceDnsCacheHitParams]]": + return self._on_dns_cache_hit + + @property + def on_dns_cache_miss(self) -> "Signal[_SignalCallback[TraceDnsCacheMissParams]]": + return self._on_dns_cache_miss + + @property + def on_request_headers_sent( + self, + ) -> "Signal[_SignalCallback[TraceRequestHeadersSentParams]]": + return self._on_request_headers_sent + + +@attr.s(auto_attribs=True, frozen=True, slots=True) +class TraceRequestStartParams: + """Parameters sent by the `on_request_start` signal""" + + method: str + url: URL + headers: "CIMultiDict[str]" + + +@attr.s(auto_attribs=True, frozen=True, slots=True) +class TraceRequestChunkSentParams: + """Parameters sent by the `on_request_chunk_sent` signal""" + + method: str + url: URL + chunk: bytes + + +@attr.s(auto_attribs=True, frozen=True, slots=True) +class TraceResponseChunkReceivedParams: + """Parameters sent by the `on_response_chunk_received` signal""" + + method: str + url: URL + chunk: bytes + + +@attr.s(auto_attribs=True, frozen=True, slots=True) +class TraceRequestEndParams: + """Parameters sent by the `on_request_end` signal""" + + method: str + url: URL + headers: "CIMultiDict[str]" + response: ClientResponse + + +@attr.s(auto_attribs=True, frozen=True, slots=True) +class TraceRequestExceptionParams: + """Parameters sent by the `on_request_exception` signal""" + + method: str + url: URL + headers: "CIMultiDict[str]" + exception: BaseException + + +@attr.s(auto_attribs=True, frozen=True, slots=True) +class TraceRequestRedirectParams: + """Parameters sent by the `on_request_redirect` signal""" + + method: str + url: URL + headers: "CIMultiDict[str]" + response: ClientResponse + + +@attr.s(auto_attribs=True, frozen=True, slots=True) +class TraceConnectionQueuedStartParams: + """Parameters sent by the `on_connection_queued_start` signal""" + + +@attr.s(auto_attribs=True, frozen=True, slots=True) +class TraceConnectionQueuedEndParams: + """Parameters sent by the `on_connection_queued_end` signal""" + + +@attr.s(auto_attribs=True, frozen=True, slots=True) +class TraceConnectionCreateStartParams: + """Parameters sent by the `on_connection_create_start` signal""" + + +@attr.s(auto_attribs=True, frozen=True, slots=True) +class TraceConnectionCreateEndParams: + """Parameters sent by the `on_connection_create_end` signal""" + + +@attr.s(auto_attribs=True, frozen=True, slots=True) +class TraceConnectionReuseconnParams: + """Parameters sent by the `on_connection_reuseconn` signal""" + + +@attr.s(auto_attribs=True, frozen=True, slots=True) +class TraceDnsResolveHostStartParams: + """Parameters sent by the `on_dns_resolvehost_start` signal""" + + host: str + + +@attr.s(auto_attribs=True, frozen=True, slots=True) +class TraceDnsResolveHostEndParams: + """Parameters sent by the `on_dns_resolvehost_end` signal""" + + host: str + + +@attr.s(auto_attribs=True, frozen=True, slots=True) +class TraceDnsCacheHitParams: + """Parameters sent by the `on_dns_cache_hit` signal""" + + host: str + + +@attr.s(auto_attribs=True, frozen=True, slots=True) +class TraceDnsCacheMissParams: + """Parameters sent by the `on_dns_cache_miss` signal""" + + host: str + + +@attr.s(auto_attribs=True, frozen=True, slots=True) +class TraceRequestHeadersSentParams: + """Parameters sent by the `on_request_headers_sent` signal""" + + method: str + url: URL + headers: "CIMultiDict[str]" + + +class Trace: + """Internal dependency holder class. + + Used to keep together the main dependencies used + at the moment of send a signal. + """ + + def __init__( + self, + session: "ClientSession", + trace_config: TraceConfig, + trace_config_ctx: SimpleNamespace, + ) -> None: + self._trace_config = trace_config + self._trace_config_ctx = trace_config_ctx + self._session = session + + async def send_request_start( + self, method: str, url: URL, headers: "CIMultiDict[str]" + ) -> None: + return await self._trace_config.on_request_start.send( + self._session, + self._trace_config_ctx, + TraceRequestStartParams(method, url, headers), + ) + + async def send_request_chunk_sent( + self, method: str, url: URL, chunk: bytes + ) -> None: + return await self._trace_config.on_request_chunk_sent.send( + self._session, + self._trace_config_ctx, + TraceRequestChunkSentParams(method, url, chunk), + ) + + async def send_response_chunk_received( + self, method: str, url: URL, chunk: bytes + ) -> None: + return await self._trace_config.on_response_chunk_received.send( + self._session, + self._trace_config_ctx, + TraceResponseChunkReceivedParams(method, url, chunk), + ) + + async def send_request_end( + self, + method: str, + url: URL, + headers: "CIMultiDict[str]", + response: ClientResponse, + ) -> None: + return await self._trace_config.on_request_end.send( + self._session, + self._trace_config_ctx, + TraceRequestEndParams(method, url, headers, response), + ) + + async def send_request_exception( + self, + method: str, + url: URL, + headers: "CIMultiDict[str]", + exception: BaseException, + ) -> None: + return await self._trace_config.on_request_exception.send( + self._session, + self._trace_config_ctx, + TraceRequestExceptionParams(method, url, headers, exception), + ) + + async def send_request_redirect( + self, + method: str, + url: URL, + headers: "CIMultiDict[str]", + response: ClientResponse, + ) -> None: + return await self._trace_config._on_request_redirect.send( + self._session, + self._trace_config_ctx, + TraceRequestRedirectParams(method, url, headers, response), + ) + + async def send_connection_queued_start(self) -> None: + return await self._trace_config.on_connection_queued_start.send( + self._session, self._trace_config_ctx, TraceConnectionQueuedStartParams() + ) + + async def send_connection_queued_end(self) -> None: + return await self._trace_config.on_connection_queued_end.send( + self._session, self._trace_config_ctx, TraceConnectionQueuedEndParams() + ) + + async def send_connection_create_start(self) -> None: + return await self._trace_config.on_connection_create_start.send( + self._session, self._trace_config_ctx, TraceConnectionCreateStartParams() + ) + + async def send_connection_create_end(self) -> None: + return await self._trace_config.on_connection_create_end.send( + self._session, self._trace_config_ctx, TraceConnectionCreateEndParams() + ) + + async def send_connection_reuseconn(self) -> None: + return await self._trace_config.on_connection_reuseconn.send( + self._session, self._trace_config_ctx, TraceConnectionReuseconnParams() + ) + + async def send_dns_resolvehost_start(self, host: str) -> None: + return await self._trace_config.on_dns_resolvehost_start.send( + self._session, self._trace_config_ctx, TraceDnsResolveHostStartParams(host) + ) + + async def send_dns_resolvehost_end(self, host: str) -> None: + return await self._trace_config.on_dns_resolvehost_end.send( + self._session, self._trace_config_ctx, TraceDnsResolveHostEndParams(host) + ) + + async def send_dns_cache_hit(self, host: str) -> None: + return await self._trace_config.on_dns_cache_hit.send( + self._session, self._trace_config_ctx, TraceDnsCacheHitParams(host) + ) + + async def send_dns_cache_miss(self, host: str) -> None: + return await self._trace_config.on_dns_cache_miss.send( + self._session, self._trace_config_ctx, TraceDnsCacheMissParams(host) + ) + + async def send_request_headers( + self, method: str, url: URL, headers: "CIMultiDict[str]" + ) -> None: + return await self._trace_config._on_request_headers_sent.send( + self._session, + self._trace_config_ctx, + TraceRequestHeadersSentParams(method, url, headers), + ) diff --git a/contrib/python/aiohttp/aiohttp/typedefs.py b/contrib/python/aiohttp/aiohttp/typedefs.py new file mode 100644 index 0000000000..84283d9a46 --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/typedefs.py @@ -0,0 +1,64 @@ +import json +import os +import sys +from typing import ( + TYPE_CHECKING, + Any, + Awaitable, + Callable, + Iterable, + Mapping, + Tuple, + Union, +) + +from multidict import CIMultiDict, CIMultiDictProxy, MultiDict, MultiDictProxy, istr +from yarl import URL + +# These are for other modules to use (to avoid repeating the conditional import). +if sys.version_info >= (3, 8): + from typing import Final as Final, Protocol as Protocol, TypedDict as TypedDict +else: + from typing_extensions import ( # noqa: F401 + Final, + Protocol as Protocol, + TypedDict as TypedDict, + ) + +DEFAULT_JSON_ENCODER = json.dumps +DEFAULT_JSON_DECODER = json.loads + +if TYPE_CHECKING: # pragma: no cover + _CIMultiDict = CIMultiDict[str] + _CIMultiDictProxy = CIMultiDictProxy[str] + _MultiDict = MultiDict[str] + _MultiDictProxy = MultiDictProxy[str] + from http.cookies import BaseCookie, Morsel + + from .web import Request, StreamResponse +else: + _CIMultiDict = CIMultiDict + _CIMultiDictProxy = CIMultiDictProxy + _MultiDict = MultiDict + _MultiDictProxy = MultiDictProxy + +Byteish = Union[bytes, bytearray, memoryview] +JSONEncoder = Callable[[Any], str] +JSONDecoder = Callable[[str], Any] +LooseHeaders = Union[Mapping[Union[str, istr], str], _CIMultiDict, _CIMultiDictProxy] +RawHeaders = Tuple[Tuple[bytes, bytes], ...] +StrOrURL = Union[str, URL] + +LooseCookiesMappings = Mapping[str, Union[str, "BaseCookie[str]", "Morsel[Any]"]] +LooseCookiesIterables = Iterable[ + Tuple[str, Union[str, "BaseCookie[str]", "Morsel[Any]"]] +] +LooseCookies = Union[ + LooseCookiesMappings, + LooseCookiesIterables, + "BaseCookie[str]", +] + +Handler = Callable[["Request"], Awaitable["StreamResponse"]] + +PathLike = Union[str, "os.PathLike[str]"] diff --git a/contrib/python/aiohttp/aiohttp/web.py b/contrib/python/aiohttp/aiohttp/web.py new file mode 100644 index 0000000000..864428b49b --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/web.py @@ -0,0 +1,586 @@ +import asyncio +import logging +import socket +import sys +from argparse import ArgumentParser +from collections.abc import Iterable +from importlib import import_module +from typing import ( + Any, + Awaitable, + Callable, + Iterable as TypingIterable, + List, + Optional, + Set, + Type, + Union, + cast, +) + +from .abc import AbstractAccessLogger +from .helpers import all_tasks +from .log import access_logger +from .web_app import Application as Application, CleanupError as CleanupError +from .web_exceptions import ( + HTTPAccepted as HTTPAccepted, + HTTPBadGateway as HTTPBadGateway, + HTTPBadRequest as HTTPBadRequest, + HTTPClientError as HTTPClientError, + HTTPConflict as HTTPConflict, + HTTPCreated as HTTPCreated, + HTTPError as HTTPError, + HTTPException as HTTPException, + HTTPExpectationFailed as HTTPExpectationFailed, + HTTPFailedDependency as HTTPFailedDependency, + HTTPForbidden as HTTPForbidden, + HTTPFound as HTTPFound, + HTTPGatewayTimeout as HTTPGatewayTimeout, + HTTPGone as HTTPGone, + HTTPInsufficientStorage as HTTPInsufficientStorage, + HTTPInternalServerError as HTTPInternalServerError, + HTTPLengthRequired as HTTPLengthRequired, + HTTPMethodNotAllowed as HTTPMethodNotAllowed, + HTTPMisdirectedRequest as HTTPMisdirectedRequest, + HTTPMovedPermanently as HTTPMovedPermanently, + HTTPMultipleChoices as HTTPMultipleChoices, + HTTPNetworkAuthenticationRequired as HTTPNetworkAuthenticationRequired, + HTTPNoContent as HTTPNoContent, + HTTPNonAuthoritativeInformation as HTTPNonAuthoritativeInformation, + HTTPNotAcceptable as HTTPNotAcceptable, + HTTPNotExtended as HTTPNotExtended, + HTTPNotFound as HTTPNotFound, + HTTPNotImplemented as HTTPNotImplemented, + HTTPNotModified as HTTPNotModified, + HTTPOk as HTTPOk, + HTTPPartialContent as HTTPPartialContent, + HTTPPaymentRequired as HTTPPaymentRequired, + HTTPPermanentRedirect as HTTPPermanentRedirect, + HTTPPreconditionFailed as HTTPPreconditionFailed, + HTTPPreconditionRequired as HTTPPreconditionRequired, + HTTPProxyAuthenticationRequired as HTTPProxyAuthenticationRequired, + HTTPRedirection as HTTPRedirection, + HTTPRequestEntityTooLarge as HTTPRequestEntityTooLarge, + HTTPRequestHeaderFieldsTooLarge as HTTPRequestHeaderFieldsTooLarge, + HTTPRequestRangeNotSatisfiable as HTTPRequestRangeNotSatisfiable, + HTTPRequestTimeout as HTTPRequestTimeout, + HTTPRequestURITooLong as HTTPRequestURITooLong, + HTTPResetContent as HTTPResetContent, + HTTPSeeOther as HTTPSeeOther, + HTTPServerError as HTTPServerError, + HTTPServiceUnavailable as HTTPServiceUnavailable, + HTTPSuccessful as HTTPSuccessful, + HTTPTemporaryRedirect as HTTPTemporaryRedirect, + HTTPTooManyRequests as HTTPTooManyRequests, + HTTPUnauthorized as HTTPUnauthorized, + HTTPUnavailableForLegalReasons as HTTPUnavailableForLegalReasons, + HTTPUnprocessableEntity as HTTPUnprocessableEntity, + HTTPUnsupportedMediaType as HTTPUnsupportedMediaType, + HTTPUpgradeRequired as HTTPUpgradeRequired, + HTTPUseProxy as HTTPUseProxy, + HTTPVariantAlsoNegotiates as HTTPVariantAlsoNegotiates, + HTTPVersionNotSupported as HTTPVersionNotSupported, +) +from .web_fileresponse import FileResponse as FileResponse +from .web_log import AccessLogger +from .web_middlewares import ( + middleware as middleware, + normalize_path_middleware as normalize_path_middleware, +) +from .web_protocol import ( + PayloadAccessError as PayloadAccessError, + RequestHandler as RequestHandler, + RequestPayloadError as RequestPayloadError, +) +from .web_request import ( + BaseRequest as BaseRequest, + FileField as FileField, + Request as Request, +) +from .web_response import ( + ContentCoding as ContentCoding, + Response as Response, + StreamResponse as StreamResponse, + json_response as json_response, +) +from .web_routedef import ( + AbstractRouteDef as AbstractRouteDef, + RouteDef as RouteDef, + RouteTableDef as RouteTableDef, + StaticDef as StaticDef, + delete as delete, + get as get, + head as head, + options as options, + patch as patch, + post as post, + put as put, + route as route, + static as static, + view as view, +) +from .web_runner import ( + AppRunner as AppRunner, + BaseRunner as BaseRunner, + BaseSite as BaseSite, + GracefulExit as GracefulExit, + NamedPipeSite as NamedPipeSite, + ServerRunner as ServerRunner, + SockSite as SockSite, + TCPSite as TCPSite, + UnixSite as UnixSite, +) +from .web_server import Server as Server +from .web_urldispatcher import ( + AbstractResource as AbstractResource, + AbstractRoute as AbstractRoute, + DynamicResource as DynamicResource, + PlainResource as PlainResource, + Resource as Resource, + ResourceRoute as ResourceRoute, + StaticResource as StaticResource, + UrlDispatcher as UrlDispatcher, + UrlMappingMatchInfo as UrlMappingMatchInfo, + View as View, +) +from .web_ws import ( + WebSocketReady as WebSocketReady, + WebSocketResponse as WebSocketResponse, + WSMsgType as WSMsgType, +) + +__all__ = ( + # web_app + "Application", + "CleanupError", + # web_exceptions + "HTTPAccepted", + "HTTPBadGateway", + "HTTPBadRequest", + "HTTPClientError", + "HTTPConflict", + "HTTPCreated", + "HTTPError", + "HTTPException", + "HTTPExpectationFailed", + "HTTPFailedDependency", + "HTTPForbidden", + "HTTPFound", + "HTTPGatewayTimeout", + "HTTPGone", + "HTTPInsufficientStorage", + "HTTPInternalServerError", + "HTTPLengthRequired", + "HTTPMethodNotAllowed", + "HTTPMisdirectedRequest", + "HTTPMovedPermanently", + "HTTPMultipleChoices", + "HTTPNetworkAuthenticationRequired", + "HTTPNoContent", + "HTTPNonAuthoritativeInformation", + "HTTPNotAcceptable", + "HTTPNotExtended", + "HTTPNotFound", + "HTTPNotImplemented", + "HTTPNotModified", + "HTTPOk", + "HTTPPartialContent", + "HTTPPaymentRequired", + "HTTPPermanentRedirect", + "HTTPPreconditionFailed", + "HTTPPreconditionRequired", + "HTTPProxyAuthenticationRequired", + "HTTPRedirection", + "HTTPRequestEntityTooLarge", + "HTTPRequestHeaderFieldsTooLarge", + "HTTPRequestRangeNotSatisfiable", + "HTTPRequestTimeout", + "HTTPRequestURITooLong", + "HTTPResetContent", + "HTTPSeeOther", + "HTTPServerError", + "HTTPServiceUnavailable", + "HTTPSuccessful", + "HTTPTemporaryRedirect", + "HTTPTooManyRequests", + "HTTPUnauthorized", + "HTTPUnavailableForLegalReasons", + "HTTPUnprocessableEntity", + "HTTPUnsupportedMediaType", + "HTTPUpgradeRequired", + "HTTPUseProxy", + "HTTPVariantAlsoNegotiates", + "HTTPVersionNotSupported", + # web_fileresponse + "FileResponse", + # web_middlewares + "middleware", + "normalize_path_middleware", + # web_protocol + "PayloadAccessError", + "RequestHandler", + "RequestPayloadError", + # web_request + "BaseRequest", + "FileField", + "Request", + # web_response + "ContentCoding", + "Response", + "StreamResponse", + "json_response", + # web_routedef + "AbstractRouteDef", + "RouteDef", + "RouteTableDef", + "StaticDef", + "delete", + "get", + "head", + "options", + "patch", + "post", + "put", + "route", + "static", + "view", + # web_runner + "AppRunner", + "BaseRunner", + "BaseSite", + "GracefulExit", + "ServerRunner", + "SockSite", + "TCPSite", + "UnixSite", + "NamedPipeSite", + # web_server + "Server", + # web_urldispatcher + "AbstractResource", + "AbstractRoute", + "DynamicResource", + "PlainResource", + "Resource", + "ResourceRoute", + "StaticResource", + "UrlDispatcher", + "UrlMappingMatchInfo", + "View", + # web_ws + "WebSocketReady", + "WebSocketResponse", + "WSMsgType", + # web + "run_app", +) + + +try: + from ssl import SSLContext +except ImportError: # pragma: no cover + SSLContext = Any # type: ignore[misc,assignment] + +HostSequence = TypingIterable[str] + + +async def _run_app( + app: Union[Application, Awaitable[Application]], + *, + host: Optional[Union[str, HostSequence]] = None, + port: Optional[int] = None, + path: Optional[str] = None, + sock: Optional[socket.socket] = None, + shutdown_timeout: float = 60.0, + keepalive_timeout: float = 75.0, + ssl_context: Optional[SSLContext] = None, + print: Callable[..., None] = print, + backlog: int = 128, + access_log_class: Type[AbstractAccessLogger] = AccessLogger, + access_log_format: str = AccessLogger.LOG_FORMAT, + access_log: Optional[logging.Logger] = access_logger, + handle_signals: bool = True, + reuse_address: Optional[bool] = None, + reuse_port: Optional[bool] = None, +) -> None: + # A internal functio to actually do all dirty job for application running + if asyncio.iscoroutine(app): + app = await app # type: ignore[misc] + + app = cast(Application, app) + + runner = AppRunner( + app, + handle_signals=handle_signals, + access_log_class=access_log_class, + access_log_format=access_log_format, + access_log=access_log, + keepalive_timeout=keepalive_timeout, + ) + + await runner.setup() + + sites = [] # type: List[BaseSite] + + try: + if host is not None: + if isinstance(host, (str, bytes, bytearray, memoryview)): + sites.append( + TCPSite( + runner, + host, + port, + shutdown_timeout=shutdown_timeout, + ssl_context=ssl_context, + backlog=backlog, + reuse_address=reuse_address, + reuse_port=reuse_port, + ) + ) + else: + for h in host: + sites.append( + TCPSite( + runner, + h, + port, + shutdown_timeout=shutdown_timeout, + ssl_context=ssl_context, + backlog=backlog, + reuse_address=reuse_address, + reuse_port=reuse_port, + ) + ) + elif path is None and sock is None or port is not None: + sites.append( + TCPSite( + runner, + port=port, + shutdown_timeout=shutdown_timeout, + ssl_context=ssl_context, + backlog=backlog, + reuse_address=reuse_address, + reuse_port=reuse_port, + ) + ) + + if path is not None: + if isinstance(path, (str, bytes, bytearray, memoryview)): + sites.append( + UnixSite( + runner, + path, + shutdown_timeout=shutdown_timeout, + ssl_context=ssl_context, + backlog=backlog, + ) + ) + else: + for p in path: + sites.append( + UnixSite( + runner, + p, + shutdown_timeout=shutdown_timeout, + ssl_context=ssl_context, + backlog=backlog, + ) + ) + + if sock is not None: + if not isinstance(sock, Iterable): + sites.append( + SockSite( + runner, + sock, + shutdown_timeout=shutdown_timeout, + ssl_context=ssl_context, + backlog=backlog, + ) + ) + else: + for s in sock: + sites.append( + SockSite( + runner, + s, + shutdown_timeout=shutdown_timeout, + ssl_context=ssl_context, + backlog=backlog, + ) + ) + for site in sites: + await site.start() + + if print: # pragma: no branch + names = sorted(str(s.name) for s in runner.sites) + print( + "======== Running on {} ========\n" + "(Press CTRL+C to quit)".format(", ".join(names)) + ) + + # sleep forever by 1 hour intervals, + # on Windows before Python 3.8 wake up every 1 second to handle + # Ctrl+C smoothly + if sys.platform == "win32" and sys.version_info < (3, 8): + delay = 1 + else: + delay = 3600 + + while True: + await asyncio.sleep(delay) + finally: + await runner.cleanup() + + +def _cancel_tasks( + to_cancel: Set["asyncio.Task[Any]"], loop: asyncio.AbstractEventLoop +) -> None: + if not to_cancel: + return + + for task in to_cancel: + task.cancel() + + loop.run_until_complete(asyncio.gather(*to_cancel, return_exceptions=True)) + + for task in to_cancel: + if task.cancelled(): + continue + if task.exception() is not None: + loop.call_exception_handler( + { + "message": "unhandled exception during asyncio.run() shutdown", + "exception": task.exception(), + "task": task, + } + ) + + +def run_app( + app: Union[Application, Awaitable[Application]], + *, + host: Optional[Union[str, HostSequence]] = None, + port: Optional[int] = None, + path: Optional[str] = None, + sock: Optional[socket.socket] = None, + shutdown_timeout: float = 60.0, + keepalive_timeout: float = 75.0, + ssl_context: Optional[SSLContext] = None, + print: Callable[..., None] = print, + backlog: int = 128, + access_log_class: Type[AbstractAccessLogger] = AccessLogger, + access_log_format: str = AccessLogger.LOG_FORMAT, + access_log: Optional[logging.Logger] = access_logger, + handle_signals: bool = True, + reuse_address: Optional[bool] = None, + reuse_port: Optional[bool] = None, + loop: Optional[asyncio.AbstractEventLoop] = None, +) -> None: + """Run an app locally""" + if loop is None: + loop = asyncio.new_event_loop() + + # Configure if and only if in debugging mode and using the default logger + if loop.get_debug() and access_log and access_log.name == "aiohttp.access": + if access_log.level == logging.NOTSET: + access_log.setLevel(logging.DEBUG) + if not access_log.hasHandlers(): + access_log.addHandler(logging.StreamHandler()) + + main_task = loop.create_task( + _run_app( + app, + host=host, + port=port, + path=path, + sock=sock, + shutdown_timeout=shutdown_timeout, + keepalive_timeout=keepalive_timeout, + ssl_context=ssl_context, + print=print, + backlog=backlog, + access_log_class=access_log_class, + access_log_format=access_log_format, + access_log=access_log, + handle_signals=handle_signals, + reuse_address=reuse_address, + reuse_port=reuse_port, + ) + ) + + try: + asyncio.set_event_loop(loop) + loop.run_until_complete(main_task) + except (GracefulExit, KeyboardInterrupt): # pragma: no cover + pass + finally: + _cancel_tasks({main_task}, loop) + _cancel_tasks(all_tasks(loop), loop) + loop.run_until_complete(loop.shutdown_asyncgens()) + loop.close() + + +def main(argv: List[str]) -> None: + arg_parser = ArgumentParser( + description="aiohttp.web Application server", prog="aiohttp.web" + ) + arg_parser.add_argument( + "entry_func", + help=( + "Callable returning the `aiohttp.web.Application` instance to " + "run. Should be specified in the 'module:function' syntax." + ), + metavar="entry-func", + ) + arg_parser.add_argument( + "-H", + "--hostname", + help="TCP/IP hostname to serve on (default: %(default)r)", + default="localhost", + ) + arg_parser.add_argument( + "-P", + "--port", + help="TCP/IP port to serve on (default: %(default)r)", + type=int, + default="8080", + ) + arg_parser.add_argument( + "-U", + "--path", + help="Unix file system path to serve on. Specifying a path will cause " + "hostname and port arguments to be ignored.", + ) + args, extra_argv = arg_parser.parse_known_args(argv) + + # Import logic + mod_str, _, func_str = args.entry_func.partition(":") + if not func_str or not mod_str: + arg_parser.error("'entry-func' not in 'module:function' syntax") + if mod_str.startswith("."): + arg_parser.error("relative module names not supported") + try: + module = import_module(mod_str) + except ImportError as ex: + arg_parser.error(f"unable to import {mod_str}: {ex}") + try: + func = getattr(module, func_str) + except AttributeError: + arg_parser.error(f"module {mod_str!r} has no attribute {func_str!r}") + + # Compatibility logic + if args.path is not None and not hasattr(socket, "AF_UNIX"): + arg_parser.error( + "file system paths not supported by your operating" " environment" + ) + + logging.basicConfig(level=logging.DEBUG) + + app = func(extra_argv) + run_app(app, host=args.hostname, port=args.port, path=args.path) + arg_parser.exit(message="Stopped\n") + + +if __name__ == "__main__": # pragma: no branch + main(sys.argv[1:]) # pragma: no cover diff --git a/contrib/python/aiohttp/aiohttp/web_app.py b/contrib/python/aiohttp/aiohttp/web_app.py new file mode 100644 index 0000000000..d5dc90ed42 --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/web_app.py @@ -0,0 +1,557 @@ +import asyncio +import logging +import warnings +from functools import partial, update_wrapper +from typing import ( + TYPE_CHECKING, + Any, + AsyncIterator, + Awaitable, + Callable, + Dict, + Iterable, + Iterator, + List, + Mapping, + MutableMapping, + Optional, + Sequence, + Tuple, + Type, + Union, + cast, +) + +from aiosignal import Signal +from frozenlist import FrozenList + +from . import hdrs +from .abc import ( + AbstractAccessLogger, + AbstractMatchInfo, + AbstractRouter, + AbstractStreamWriter, +) +from .helpers import DEBUG +from .http_parser import RawRequestMessage +from .log import web_logger +from .streams import StreamReader +from .web_log import AccessLogger +from .web_middlewares import _fix_request_current_app +from .web_protocol import RequestHandler +from .web_request import Request +from .web_response import StreamResponse +from .web_routedef import AbstractRouteDef +from .web_server import Server +from .web_urldispatcher import ( + AbstractResource, + AbstractRoute, + Domain, + MaskDomain, + MatchedSubAppResource, + PrefixedSubAppResource, + UrlDispatcher, +) + +__all__ = ("Application", "CleanupError") + + +if TYPE_CHECKING: # pragma: no cover + from .typedefs import Handler + + _AppSignal = Signal[Callable[["Application"], Awaitable[None]]] + _RespPrepareSignal = Signal[Callable[[Request, StreamResponse], Awaitable[None]]] + _Middleware = Union[ + Callable[[Request, Handler], Awaitable[StreamResponse]], + Callable[["Application", Handler], Awaitable[Handler]], # old-style + ] + _Middlewares = FrozenList[_Middleware] + _MiddlewaresHandlers = Optional[Sequence[Tuple[_Middleware, bool]]] + _Subapps = List["Application"] +else: + # No type checker mode, skip types + _AppSignal = Signal + _RespPrepareSignal = Signal + _Middleware = Callable + _Middlewares = FrozenList + _MiddlewaresHandlers = Optional[Sequence] + _Subapps = List + + +class Application(MutableMapping[str, Any]): + ATTRS = frozenset( + [ + "logger", + "_debug", + "_router", + "_loop", + "_handler_args", + "_middlewares", + "_middlewares_handlers", + "_run_middlewares", + "_state", + "_frozen", + "_pre_frozen", + "_subapps", + "_on_response_prepare", + "_on_startup", + "_on_shutdown", + "_on_cleanup", + "_client_max_size", + "_cleanup_ctx", + ] + ) + + def __init__( + self, + *, + logger: logging.Logger = web_logger, + router: Optional[UrlDispatcher] = None, + middlewares: Iterable[_Middleware] = (), + handler_args: Optional[Mapping[str, Any]] = None, + client_max_size: int = 1024 ** 2, + loop: Optional[asyncio.AbstractEventLoop] = None, + debug: Any = ..., # mypy doesn't support ellipsis + ) -> None: + if router is None: + router = UrlDispatcher() + else: + warnings.warn( + "router argument is deprecated", DeprecationWarning, stacklevel=2 + ) + assert isinstance(router, AbstractRouter), router + + if loop is not None: + warnings.warn( + "loop argument is deprecated", DeprecationWarning, stacklevel=2 + ) + + if debug is not ...: + warnings.warn( + "debug argument is deprecated", DeprecationWarning, stacklevel=2 + ) + self._debug = debug + self._router = router # type: UrlDispatcher + self._loop = loop + self._handler_args = handler_args + self.logger = logger + + self._middlewares = FrozenList(middlewares) # type: _Middlewares + + # initialized on freezing + self._middlewares_handlers = None # type: _MiddlewaresHandlers + # initialized on freezing + self._run_middlewares = None # type: Optional[bool] + + self._state = {} # type: Dict[str, Any] + self._frozen = False + self._pre_frozen = False + self._subapps = [] # type: _Subapps + + self._on_response_prepare = Signal(self) # type: _RespPrepareSignal + self._on_startup = Signal(self) # type: _AppSignal + self._on_shutdown = Signal(self) # type: _AppSignal + self._on_cleanup = Signal(self) # type: _AppSignal + self._cleanup_ctx = CleanupContext() + self._on_startup.append(self._cleanup_ctx._on_startup) + self._on_cleanup.append(self._cleanup_ctx._on_cleanup) + self._client_max_size = client_max_size + + def __init_subclass__(cls: Type["Application"]) -> None: + warnings.warn( + "Inheritance class {} from web.Application " + "is discouraged".format(cls.__name__), + DeprecationWarning, + stacklevel=2, + ) + + if DEBUG: # pragma: no cover + + def __setattr__(self, name: str, val: Any) -> None: + if name not in self.ATTRS: + warnings.warn( + "Setting custom web.Application.{} attribute " + "is discouraged".format(name), + DeprecationWarning, + stacklevel=2, + ) + super().__setattr__(name, val) + + # MutableMapping API + + def __eq__(self, other: object) -> bool: + return self is other + + def __getitem__(self, key: str) -> Any: + return self._state[key] + + def _check_frozen(self) -> None: + if self._frozen: + warnings.warn( + "Changing state of started or joined " "application is deprecated", + DeprecationWarning, + stacklevel=3, + ) + + def __setitem__(self, key: str, value: Any) -> None: + self._check_frozen() + self._state[key] = value + + def __delitem__(self, key: str) -> None: + self._check_frozen() + del self._state[key] + + def __len__(self) -> int: + return len(self._state) + + def __iter__(self) -> Iterator[str]: + return iter(self._state) + + ######## + @property + def loop(self) -> asyncio.AbstractEventLoop: + # Technically the loop can be None + # but we mask it by explicit type cast + # to provide more convinient type annotation + warnings.warn("loop property is deprecated", DeprecationWarning, stacklevel=2) + return cast(asyncio.AbstractEventLoop, self._loop) + + def _set_loop(self, loop: Optional[asyncio.AbstractEventLoop]) -> None: + if loop is None: + loop = asyncio.get_event_loop() + if self._loop is not None and self._loop is not loop: + raise RuntimeError( + "web.Application instance initialized with different loop" + ) + + self._loop = loop + + # set loop debug + if self._debug is ...: + self._debug = loop.get_debug() + + # set loop to sub applications + for subapp in self._subapps: + subapp._set_loop(loop) + + @property + def pre_frozen(self) -> bool: + return self._pre_frozen + + def pre_freeze(self) -> None: + if self._pre_frozen: + return + + self._pre_frozen = True + self._middlewares.freeze() + self._router.freeze() + self._on_response_prepare.freeze() + self._cleanup_ctx.freeze() + self._on_startup.freeze() + self._on_shutdown.freeze() + self._on_cleanup.freeze() + self._middlewares_handlers = tuple(self._prepare_middleware()) + + # If current app and any subapp do not have middlewares avoid run all + # of the code footprint that it implies, which have a middleware + # hardcoded per app that sets up the current_app attribute. If no + # middlewares are configured the handler will receive the proper + # current_app without needing all of this code. + self._run_middlewares = True if self.middlewares else False + + for subapp in self._subapps: + subapp.pre_freeze() + self._run_middlewares = self._run_middlewares or subapp._run_middlewares + + @property + def frozen(self) -> bool: + return self._frozen + + def freeze(self) -> None: + if self._frozen: + return + + self.pre_freeze() + self._frozen = True + for subapp in self._subapps: + subapp.freeze() + + @property + def debug(self) -> bool: + warnings.warn("debug property is deprecated", DeprecationWarning, stacklevel=2) + return self._debug # type: ignore[no-any-return] + + def _reg_subapp_signals(self, subapp: "Application") -> None: + def reg_handler(signame: str) -> None: + subsig = getattr(subapp, signame) + + async def handler(app: "Application") -> None: + await subsig.send(subapp) + + appsig = getattr(self, signame) + appsig.append(handler) + + reg_handler("on_startup") + reg_handler("on_shutdown") + reg_handler("on_cleanup") + + def add_subapp(self, prefix: str, subapp: "Application") -> AbstractResource: + if not isinstance(prefix, str): + raise TypeError("Prefix must be str") + prefix = prefix.rstrip("/") + if not prefix: + raise ValueError("Prefix cannot be empty") + factory = partial(PrefixedSubAppResource, prefix, subapp) + return self._add_subapp(factory, subapp) + + def _add_subapp( + self, resource_factory: Callable[[], AbstractResource], subapp: "Application" + ) -> AbstractResource: + if self.frozen: + raise RuntimeError("Cannot add sub application to frozen application") + if subapp.frozen: + raise RuntimeError("Cannot add frozen application") + resource = resource_factory() + self.router.register_resource(resource) + self._reg_subapp_signals(subapp) + self._subapps.append(subapp) + subapp.pre_freeze() + if self._loop is not None: + subapp._set_loop(self._loop) + return resource + + def add_domain(self, domain: str, subapp: "Application") -> AbstractResource: + if not isinstance(domain, str): + raise TypeError("Domain must be str") + elif "*" in domain: + rule = MaskDomain(domain) # type: Domain + else: + rule = Domain(domain) + factory = partial(MatchedSubAppResource, rule, subapp) + return self._add_subapp(factory, subapp) + + def add_routes(self, routes: Iterable[AbstractRouteDef]) -> List[AbstractRoute]: + return self.router.add_routes(routes) + + @property + def on_response_prepare(self) -> _RespPrepareSignal: + return self._on_response_prepare + + @property + def on_startup(self) -> _AppSignal: + return self._on_startup + + @property + def on_shutdown(self) -> _AppSignal: + return self._on_shutdown + + @property + def on_cleanup(self) -> _AppSignal: + return self._on_cleanup + + @property + def cleanup_ctx(self) -> "CleanupContext": + return self._cleanup_ctx + + @property + def router(self) -> UrlDispatcher: + return self._router + + @property + def middlewares(self) -> _Middlewares: + return self._middlewares + + def _make_handler( + self, + *, + loop: Optional[asyncio.AbstractEventLoop] = None, + access_log_class: Type[AbstractAccessLogger] = AccessLogger, + **kwargs: Any, + ) -> Server: + + if not issubclass(access_log_class, AbstractAccessLogger): + raise TypeError( + "access_log_class must be subclass of " + "aiohttp.abc.AbstractAccessLogger, got {}".format(access_log_class) + ) + + self._set_loop(loop) + self.freeze() + + kwargs["debug"] = self._debug + kwargs["access_log_class"] = access_log_class + if self._handler_args: + for k, v in self._handler_args.items(): + kwargs[k] = v + + return Server( + self._handle, # type: ignore[arg-type] + request_factory=self._make_request, + loop=self._loop, + **kwargs, + ) + + def make_handler( + self, + *, + loop: Optional[asyncio.AbstractEventLoop] = None, + access_log_class: Type[AbstractAccessLogger] = AccessLogger, + **kwargs: Any, + ) -> Server: + + warnings.warn( + "Application.make_handler(...) is deprecated, " "use AppRunner API instead", + DeprecationWarning, + stacklevel=2, + ) + + return self._make_handler( + loop=loop, access_log_class=access_log_class, **kwargs + ) + + async def startup(self) -> None: + """Causes on_startup signal + + Should be called in the event loop along with the request handler. + """ + await self.on_startup.send(self) + + async def shutdown(self) -> None: + """Causes on_shutdown signal + + Should be called before cleanup() + """ + await self.on_shutdown.send(self) + + async def cleanup(self) -> None: + """Causes on_cleanup signal + + Should be called after shutdown() + """ + if self.on_cleanup.frozen: + await self.on_cleanup.send(self) + else: + # If an exception occurs in startup, ensure cleanup contexts are completed. + await self._cleanup_ctx._on_cleanup(self) + + def _make_request( + self, + message: RawRequestMessage, + payload: StreamReader, + protocol: RequestHandler, + writer: AbstractStreamWriter, + task: "asyncio.Task[None]", + _cls: Type[Request] = Request, + ) -> Request: + return _cls( + message, + payload, + protocol, + writer, + task, + self._loop, + client_max_size=self._client_max_size, + ) + + def _prepare_middleware(self) -> Iterator[Tuple[_Middleware, bool]]: + for m in reversed(self._middlewares): + if getattr(m, "__middleware_version__", None) == 1: + yield m, True + else: + warnings.warn( + 'old-style middleware "{!r}" deprecated, ' "see #2252".format(m), + DeprecationWarning, + stacklevel=2, + ) + yield m, False + + yield _fix_request_current_app(self), True + + async def _handle(self, request: Request) -> StreamResponse: + loop = asyncio.get_event_loop() + debug = loop.get_debug() + match_info = await self._router.resolve(request) + if debug: # pragma: no cover + if not isinstance(match_info, AbstractMatchInfo): + raise TypeError( + "match_info should be AbstractMatchInfo " + "instance, not {!r}".format(match_info) + ) + match_info.add_app(self) + + match_info.freeze() + + resp = None + request._match_info = match_info + expect = request.headers.get(hdrs.EXPECT) + if expect: + resp = await match_info.expect_handler(request) + await request.writer.drain() + + if resp is None: + handler = match_info.handler + + if self._run_middlewares: + for app in match_info.apps[::-1]: + for m, new_style in app._middlewares_handlers: # type: ignore[union-attr] # noqa + if new_style: + handler = update_wrapper( + partial(m, handler=handler), handler + ) + else: + handler = await m(app, handler) # type: ignore[arg-type] + + resp = await handler(request) + + return resp + + def __call__(self) -> "Application": + """gunicorn compatibility""" + return self + + def __repr__(self) -> str: + return f"<Application 0x{id(self):x}>" + + def __bool__(self) -> bool: + return True + + +class CleanupError(RuntimeError): + @property + def exceptions(self) -> List[BaseException]: + return cast(List[BaseException], self.args[1]) + + +if TYPE_CHECKING: # pragma: no cover + _CleanupContextBase = FrozenList[Callable[[Application], AsyncIterator[None]]] +else: + _CleanupContextBase = FrozenList + + +class CleanupContext(_CleanupContextBase): + def __init__(self) -> None: + super().__init__() + self._exits = [] # type: List[AsyncIterator[None]] + + async def _on_startup(self, app: Application) -> None: + for cb in self: + it = cb(app).__aiter__() + await it.__anext__() + self._exits.append(it) + + async def _on_cleanup(self, app: Application) -> None: + errors = [] + for it in reversed(self._exits): + try: + await it.__anext__() + except StopAsyncIteration: + pass + except Exception as exc: + errors.append(exc) + else: + errors.append(RuntimeError(f"{it!r} has more than one 'yield'")) + if errors: + if len(errors) == 1: + raise errors[0] + else: + raise CleanupError("Multiple errors on cleanup stage", errors) diff --git a/contrib/python/aiohttp/aiohttp/web_exceptions.py b/contrib/python/aiohttp/aiohttp/web_exceptions.py new file mode 100644 index 0000000000..2eadca0386 --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/web_exceptions.py @@ -0,0 +1,441 @@ +import warnings +from typing import Any, Dict, Iterable, List, Optional, Set # noqa + +from yarl import URL + +from .typedefs import LooseHeaders, StrOrURL +from .web_response import Response + +__all__ = ( + "HTTPException", + "HTTPError", + "HTTPRedirection", + "HTTPSuccessful", + "HTTPOk", + "HTTPCreated", + "HTTPAccepted", + "HTTPNonAuthoritativeInformation", + "HTTPNoContent", + "HTTPResetContent", + "HTTPPartialContent", + "HTTPMultipleChoices", + "HTTPMovedPermanently", + "HTTPFound", + "HTTPSeeOther", + "HTTPNotModified", + "HTTPUseProxy", + "HTTPTemporaryRedirect", + "HTTPPermanentRedirect", + "HTTPClientError", + "HTTPBadRequest", + "HTTPUnauthorized", + "HTTPPaymentRequired", + "HTTPForbidden", + "HTTPNotFound", + "HTTPMethodNotAllowed", + "HTTPNotAcceptable", + "HTTPProxyAuthenticationRequired", + "HTTPRequestTimeout", + "HTTPConflict", + "HTTPGone", + "HTTPLengthRequired", + "HTTPPreconditionFailed", + "HTTPRequestEntityTooLarge", + "HTTPRequestURITooLong", + "HTTPUnsupportedMediaType", + "HTTPRequestRangeNotSatisfiable", + "HTTPExpectationFailed", + "HTTPMisdirectedRequest", + "HTTPUnprocessableEntity", + "HTTPFailedDependency", + "HTTPUpgradeRequired", + "HTTPPreconditionRequired", + "HTTPTooManyRequests", + "HTTPRequestHeaderFieldsTooLarge", + "HTTPUnavailableForLegalReasons", + "HTTPServerError", + "HTTPInternalServerError", + "HTTPNotImplemented", + "HTTPBadGateway", + "HTTPServiceUnavailable", + "HTTPGatewayTimeout", + "HTTPVersionNotSupported", + "HTTPVariantAlsoNegotiates", + "HTTPInsufficientStorage", + "HTTPNotExtended", + "HTTPNetworkAuthenticationRequired", +) + + +############################################################ +# HTTP Exceptions +############################################################ + + +class HTTPException(Response, Exception): + + # You should set in subclasses: + # status = 200 + + status_code = -1 + empty_body = False + + __http_exception__ = True + + def __init__( + self, + *, + headers: Optional[LooseHeaders] = None, + reason: Optional[str] = None, + body: Any = None, + text: Optional[str] = None, + content_type: Optional[str] = None, + ) -> None: + if body is not None: + warnings.warn( + "body argument is deprecated for http web exceptions", + DeprecationWarning, + ) + Response.__init__( + self, + status=self.status_code, + headers=headers, + reason=reason, + body=body, + text=text, + content_type=content_type, + ) + Exception.__init__(self, self.reason) + if self.body is None and not self.empty_body: + self.text = f"{self.status}: {self.reason}" + + def __bool__(self) -> bool: + return True + + +class HTTPError(HTTPException): + """Base class for exceptions with status codes in the 400s and 500s.""" + + +class HTTPRedirection(HTTPException): + """Base class for exceptions with status codes in the 300s.""" + + +class HTTPSuccessful(HTTPException): + """Base class for exceptions with status codes in the 200s.""" + + +class HTTPOk(HTTPSuccessful): + status_code = 200 + + +class HTTPCreated(HTTPSuccessful): + status_code = 201 + + +class HTTPAccepted(HTTPSuccessful): + status_code = 202 + + +class HTTPNonAuthoritativeInformation(HTTPSuccessful): + status_code = 203 + + +class HTTPNoContent(HTTPSuccessful): + status_code = 204 + empty_body = True + + +class HTTPResetContent(HTTPSuccessful): + status_code = 205 + empty_body = True + + +class HTTPPartialContent(HTTPSuccessful): + status_code = 206 + + +############################################################ +# 3xx redirection +############################################################ + + +class _HTTPMove(HTTPRedirection): + def __init__( + self, + location: StrOrURL, + *, + headers: Optional[LooseHeaders] = None, + reason: Optional[str] = None, + body: Any = None, + text: Optional[str] = None, + content_type: Optional[str] = None, + ) -> None: + if not location: + raise ValueError("HTTP redirects need a location to redirect to.") + super().__init__( + headers=headers, + reason=reason, + body=body, + text=text, + content_type=content_type, + ) + self.headers["Location"] = str(URL(location)) + self.location = location + + +class HTTPMultipleChoices(_HTTPMove): + status_code = 300 + + +class HTTPMovedPermanently(_HTTPMove): + status_code = 301 + + +class HTTPFound(_HTTPMove): + status_code = 302 + + +# This one is safe after a POST (the redirected location will be +# retrieved with GET): +class HTTPSeeOther(_HTTPMove): + status_code = 303 + + +class HTTPNotModified(HTTPRedirection): + # FIXME: this should include a date or etag header + status_code = 304 + empty_body = True + + +class HTTPUseProxy(_HTTPMove): + # Not a move, but looks a little like one + status_code = 305 + + +class HTTPTemporaryRedirect(_HTTPMove): + status_code = 307 + + +class HTTPPermanentRedirect(_HTTPMove): + status_code = 308 + + +############################################################ +# 4xx client error +############################################################ + + +class HTTPClientError(HTTPError): + pass + + +class HTTPBadRequest(HTTPClientError): + status_code = 400 + + +class HTTPUnauthorized(HTTPClientError): + status_code = 401 + + +class HTTPPaymentRequired(HTTPClientError): + status_code = 402 + + +class HTTPForbidden(HTTPClientError): + status_code = 403 + + +class HTTPNotFound(HTTPClientError): + status_code = 404 + + +class HTTPMethodNotAllowed(HTTPClientError): + status_code = 405 + + def __init__( + self, + method: str, + allowed_methods: Iterable[str], + *, + headers: Optional[LooseHeaders] = None, + reason: Optional[str] = None, + body: Any = None, + text: Optional[str] = None, + content_type: Optional[str] = None, + ) -> None: + allow = ",".join(sorted(allowed_methods)) + super().__init__( + headers=headers, + reason=reason, + body=body, + text=text, + content_type=content_type, + ) + self.headers["Allow"] = allow + self.allowed_methods = set(allowed_methods) # type: Set[str] + self.method = method.upper() + + +class HTTPNotAcceptable(HTTPClientError): + status_code = 406 + + +class HTTPProxyAuthenticationRequired(HTTPClientError): + status_code = 407 + + +class HTTPRequestTimeout(HTTPClientError): + status_code = 408 + + +class HTTPConflict(HTTPClientError): + status_code = 409 + + +class HTTPGone(HTTPClientError): + status_code = 410 + + +class HTTPLengthRequired(HTTPClientError): + status_code = 411 + + +class HTTPPreconditionFailed(HTTPClientError): + status_code = 412 + + +class HTTPRequestEntityTooLarge(HTTPClientError): + status_code = 413 + + def __init__(self, max_size: float, actual_size: float, **kwargs: Any) -> None: + kwargs.setdefault( + "text", + "Maximum request body size {} exceeded, " + "actual body size {}".format(max_size, actual_size), + ) + super().__init__(**kwargs) + + +class HTTPRequestURITooLong(HTTPClientError): + status_code = 414 + + +class HTTPUnsupportedMediaType(HTTPClientError): + status_code = 415 + + +class HTTPRequestRangeNotSatisfiable(HTTPClientError): + status_code = 416 + + +class HTTPExpectationFailed(HTTPClientError): + status_code = 417 + + +class HTTPMisdirectedRequest(HTTPClientError): + status_code = 421 + + +class HTTPUnprocessableEntity(HTTPClientError): + status_code = 422 + + +class HTTPFailedDependency(HTTPClientError): + status_code = 424 + + +class HTTPUpgradeRequired(HTTPClientError): + status_code = 426 + + +class HTTPPreconditionRequired(HTTPClientError): + status_code = 428 + + +class HTTPTooManyRequests(HTTPClientError): + status_code = 429 + + +class HTTPRequestHeaderFieldsTooLarge(HTTPClientError): + status_code = 431 + + +class HTTPUnavailableForLegalReasons(HTTPClientError): + status_code = 451 + + def __init__( + self, + link: str, + *, + headers: Optional[LooseHeaders] = None, + reason: Optional[str] = None, + body: Any = None, + text: Optional[str] = None, + content_type: Optional[str] = None, + ) -> None: + super().__init__( + headers=headers, + reason=reason, + body=body, + text=text, + content_type=content_type, + ) + self.headers["Link"] = '<%s>; rel="blocked-by"' % link + self.link = link + + +############################################################ +# 5xx Server Error +############################################################ +# Response status codes beginning with the digit "5" indicate cases in +# which the server is aware that it has erred or is incapable of +# performing the request. Except when responding to a HEAD request, the +# server SHOULD include an entity containing an explanation of the error +# situation, and whether it is a temporary or permanent condition. User +# agents SHOULD display any included entity to the user. These response +# codes are applicable to any request method. + + +class HTTPServerError(HTTPError): + pass + + +class HTTPInternalServerError(HTTPServerError): + status_code = 500 + + +class HTTPNotImplemented(HTTPServerError): + status_code = 501 + + +class HTTPBadGateway(HTTPServerError): + status_code = 502 + + +class HTTPServiceUnavailable(HTTPServerError): + status_code = 503 + + +class HTTPGatewayTimeout(HTTPServerError): + status_code = 504 + + +class HTTPVersionNotSupported(HTTPServerError): + status_code = 505 + + +class HTTPVariantAlsoNegotiates(HTTPServerError): + status_code = 506 + + +class HTTPInsufficientStorage(HTTPServerError): + status_code = 507 + + +class HTTPNotExtended(HTTPServerError): + status_code = 510 + + +class HTTPNetworkAuthenticationRequired(HTTPServerError): + status_code = 511 diff --git a/contrib/python/aiohttp/aiohttp/web_fileresponse.py b/contrib/python/aiohttp/aiohttp/web_fileresponse.py new file mode 100644 index 0000000000..f41ed3fd0a --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/web_fileresponse.py @@ -0,0 +1,288 @@ +import asyncio +import mimetypes +import os +import pathlib +import sys +from typing import ( # noqa + IO, + TYPE_CHECKING, + Any, + Awaitable, + Callable, + Iterator, + List, + Optional, + Tuple, + Union, + cast, +) + +from . import hdrs +from .abc import AbstractStreamWriter +from .helpers import ETAG_ANY, ETag +from .typedefs import Final, LooseHeaders +from .web_exceptions import ( + HTTPNotModified, + HTTPPartialContent, + HTTPPreconditionFailed, + HTTPRequestRangeNotSatisfiable, +) +from .web_response import StreamResponse + +__all__ = ("FileResponse",) + +if TYPE_CHECKING: # pragma: no cover + from .web_request import BaseRequest + + +_T_OnChunkSent = Optional[Callable[[bytes], Awaitable[None]]] + + +NOSENDFILE: Final[bool] = bool(os.environ.get("AIOHTTP_NOSENDFILE")) + + +class FileResponse(StreamResponse): + """A response object can be used to send files.""" + + def __init__( + self, + path: Union[str, pathlib.Path], + chunk_size: int = 256 * 1024, + status: int = 200, + reason: Optional[str] = None, + headers: Optional[LooseHeaders] = None, + ) -> None: + super().__init__(status=status, reason=reason, headers=headers) + + if isinstance(path, str): + path = pathlib.Path(path) + + self._path = path + self._chunk_size = chunk_size + + async def _sendfile_fallback( + self, writer: AbstractStreamWriter, fobj: IO[Any], offset: int, count: int + ) -> AbstractStreamWriter: + # To keep memory usage low,fobj is transferred in chunks + # controlled by the constructor's chunk_size argument. + + chunk_size = self._chunk_size + loop = asyncio.get_event_loop() + + await loop.run_in_executor(None, fobj.seek, offset) + + chunk = await loop.run_in_executor(None, fobj.read, chunk_size) + while chunk: + await writer.write(chunk) + count = count - chunk_size + if count <= 0: + break + chunk = await loop.run_in_executor(None, fobj.read, min(chunk_size, count)) + + await writer.drain() + return writer + + async def _sendfile( + self, request: "BaseRequest", fobj: IO[Any], offset: int, count: int + ) -> AbstractStreamWriter: + writer = await super().prepare(request) + assert writer is not None + + if NOSENDFILE or sys.version_info < (3, 7) or self.compression: + return await self._sendfile_fallback(writer, fobj, offset, count) + + loop = request._loop + transport = request.transport + assert transport is not None + + try: + await loop.sendfile(transport, fobj, offset, count) + except NotImplementedError: + return await self._sendfile_fallback(writer, fobj, offset, count) + + await super().write_eof() + return writer + + @staticmethod + def _strong_etag_match(etag_value: str, etags: Tuple[ETag, ...]) -> bool: + if len(etags) == 1 and etags[0].value == ETAG_ANY: + return True + return any(etag.value == etag_value for etag in etags if not etag.is_weak) + + async def _not_modified( + self, request: "BaseRequest", etag_value: str, last_modified: float + ) -> Optional[AbstractStreamWriter]: + self.set_status(HTTPNotModified.status_code) + self._length_check = False + self.etag = etag_value # type: ignore[assignment] + self.last_modified = last_modified # type: ignore[assignment] + # Delete any Content-Length headers provided by user. HTTP 304 + # should always have empty response body + return await super().prepare(request) + + async def _precondition_failed( + self, request: "BaseRequest" + ) -> Optional[AbstractStreamWriter]: + self.set_status(HTTPPreconditionFailed.status_code) + self.content_length = 0 + return await super().prepare(request) + + async def prepare(self, request: "BaseRequest") -> Optional[AbstractStreamWriter]: + filepath = self._path + + gzip = False + if "gzip" in request.headers.get(hdrs.ACCEPT_ENCODING, ""): + gzip_path = filepath.with_name(filepath.name + ".gz") + + if gzip_path.is_file(): + filepath = gzip_path + gzip = True + + loop = asyncio.get_event_loop() + st: os.stat_result = await loop.run_in_executor(None, filepath.stat) + + etag_value = f"{st.st_mtime_ns:x}-{st.st_size:x}" + last_modified = st.st_mtime + + # https://tools.ietf.org/html/rfc7232#section-6 + ifmatch = request.if_match + if ifmatch is not None and not self._strong_etag_match(etag_value, ifmatch): + return await self._precondition_failed(request) + + unmodsince = request.if_unmodified_since + if ( + unmodsince is not None + and ifmatch is None + and st.st_mtime > unmodsince.timestamp() + ): + return await self._precondition_failed(request) + + ifnonematch = request.if_none_match + if ifnonematch is not None and self._strong_etag_match(etag_value, ifnonematch): + return await self._not_modified(request, etag_value, last_modified) + + modsince = request.if_modified_since + if ( + modsince is not None + and ifnonematch is None + and st.st_mtime <= modsince.timestamp() + ): + return await self._not_modified(request, etag_value, last_modified) + + if hdrs.CONTENT_TYPE not in self.headers: + ct, encoding = mimetypes.guess_type(str(filepath)) + if not ct: + ct = "application/octet-stream" + should_set_ct = True + else: + encoding = "gzip" if gzip else None + should_set_ct = False + + status = self._status + file_size = st.st_size + count = file_size + + start = None + + ifrange = request.if_range + if ifrange is None or st.st_mtime <= ifrange.timestamp(): + # If-Range header check: + # condition = cached date >= last modification date + # return 206 if True else 200. + # if False: + # Range header would not be processed, return 200 + # if True but Range header missing + # return 200 + try: + rng = request.http_range + start = rng.start + end = rng.stop + except ValueError: + # https://tools.ietf.org/html/rfc7233: + # A server generating a 416 (Range Not Satisfiable) response to + # a byte-range request SHOULD send a Content-Range header field + # with an unsatisfied-range value. + # The complete-length in a 416 response indicates the current + # length of the selected representation. + # + # Will do the same below. Many servers ignore this and do not + # send a Content-Range header with HTTP 416 + self.headers[hdrs.CONTENT_RANGE] = f"bytes */{file_size}" + self.set_status(HTTPRequestRangeNotSatisfiable.status_code) + return await super().prepare(request) + + # If a range request has been made, convert start, end slice + # notation into file pointer offset and count + if start is not None or end is not None: + if start < 0 and end is None: # return tail of file + start += file_size + if start < 0: + # if Range:bytes=-1000 in request header but file size + # is only 200, there would be trouble without this + start = 0 + count = file_size - start + else: + # rfc7233:If the last-byte-pos value is + # absent, or if the value is greater than or equal to + # the current length of the representation data, + # the byte range is interpreted as the remainder + # of the representation (i.e., the server replaces the + # value of last-byte-pos with a value that is one less than + # the current length of the selected representation). + count = ( + min(end if end is not None else file_size, file_size) - start + ) + + if start >= file_size: + # HTTP 416 should be returned in this case. + # + # According to https://tools.ietf.org/html/rfc7233: + # If a valid byte-range-set includes at least one + # byte-range-spec with a first-byte-pos that is less than + # the current length of the representation, or at least one + # suffix-byte-range-spec with a non-zero suffix-length, + # then the byte-range-set is satisfiable. Otherwise, the + # byte-range-set is unsatisfiable. + self.headers[hdrs.CONTENT_RANGE] = f"bytes */{file_size}" + self.set_status(HTTPRequestRangeNotSatisfiable.status_code) + return await super().prepare(request) + + status = HTTPPartialContent.status_code + # Even though you are sending the whole file, you should still + # return a HTTP 206 for a Range request. + self.set_status(status) + + if should_set_ct: + self.content_type = ct # type: ignore[assignment] + if encoding: + self.headers[hdrs.CONTENT_ENCODING] = encoding + if gzip: + self.headers[hdrs.VARY] = hdrs.ACCEPT_ENCODING + + self.etag = etag_value # type: ignore[assignment] + self.last_modified = st.st_mtime # type: ignore[assignment] + self.content_length = count + + self.headers[hdrs.ACCEPT_RANGES] = "bytes" + + real_start = cast(int, start) + + if status == HTTPPartialContent.status_code: + self.headers[hdrs.CONTENT_RANGE] = "bytes {}-{}/{}".format( + real_start, real_start + count - 1, file_size + ) + + # If we are sending 0 bytes calling sendfile() will throw a ValueError + if count == 0 or request.method == hdrs.METH_HEAD or self.status in [204, 304]: + return await super().prepare(request) + + fobj = await loop.run_in_executor(None, filepath.open, "rb") + if start: # be aware that start could be None or int=0 here. + offset = start + else: + offset = 0 + + try: + return await self._sendfile(request, fobj, offset, count) + finally: + await loop.run_in_executor(None, fobj.close) diff --git a/contrib/python/aiohttp/aiohttp/web_log.py b/contrib/python/aiohttp/aiohttp/web_log.py new file mode 100644 index 0000000000..a977c1ba5c --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/web_log.py @@ -0,0 +1,208 @@ +import datetime +import functools +import logging +import os +import re +from collections import namedtuple +from typing import Any, Callable, Dict, Iterable, List, Tuple # noqa + +from .abc import AbstractAccessLogger +from .web_request import BaseRequest +from .web_response import StreamResponse + +KeyMethod = namedtuple("KeyMethod", "key method") + + +class AccessLogger(AbstractAccessLogger): + """Helper object to log access. + + Usage: + log = logging.getLogger("spam") + log_format = "%a %{User-Agent}i" + access_logger = AccessLogger(log, log_format) + access_logger.log(request, response, time) + + Format: + %% The percent sign + %a Remote IP-address (IP-address of proxy if using reverse proxy) + %t Time when the request was started to process + %P The process ID of the child that serviced the request + %r First line of request + %s Response status code + %b Size of response in bytes, including HTTP headers + %T Time taken to serve the request, in seconds + %Tf Time taken to serve the request, in seconds with floating fraction + in .06f format + %D Time taken to serve the request, in microseconds + %{FOO}i request.headers['FOO'] + %{FOO}o response.headers['FOO'] + %{FOO}e os.environ['FOO'] + + """ + + LOG_FORMAT_MAP = { + "a": "remote_address", + "t": "request_start_time", + "P": "process_id", + "r": "first_request_line", + "s": "response_status", + "b": "response_size", + "T": "request_time", + "Tf": "request_time_frac", + "D": "request_time_micro", + "i": "request_header", + "o": "response_header", + } + + LOG_FORMAT = '%a %t "%r" %s %b "%{Referer}i" "%{User-Agent}i"' + FORMAT_RE = re.compile(r"%(\{([A-Za-z0-9\-_]+)\}([ioe])|[atPrsbOD]|Tf?)") + CLEANUP_RE = re.compile(r"(%[^s])") + _FORMAT_CACHE = {} # type: Dict[str, Tuple[str, List[KeyMethod]]] + + def __init__(self, logger: logging.Logger, log_format: str = LOG_FORMAT) -> None: + """Initialise the logger. + + logger is a logger object to be used for logging. + log_format is a string with apache compatible log format description. + + """ + super().__init__(logger, log_format=log_format) + + _compiled_format = AccessLogger._FORMAT_CACHE.get(log_format) + if not _compiled_format: + _compiled_format = self.compile_format(log_format) + AccessLogger._FORMAT_CACHE[log_format] = _compiled_format + + self._log_format, self._methods = _compiled_format + + def compile_format(self, log_format: str) -> Tuple[str, List[KeyMethod]]: + """Translate log_format into form usable by modulo formatting + + All known atoms will be replaced with %s + Also methods for formatting of those atoms will be added to + _methods in appropriate order + + For example we have log_format = "%a %t" + This format will be translated to "%s %s" + Also contents of _methods will be + [self._format_a, self._format_t] + These method will be called and results will be passed + to translated string format. + + Each _format_* method receive 'args' which is list of arguments + given to self.log + + Exceptions are _format_e, _format_i and _format_o methods which + also receive key name (by functools.partial) + + """ + # list of (key, method) tuples, we don't use an OrderedDict as users + # can repeat the same key more than once + methods = list() + + for atom in self.FORMAT_RE.findall(log_format): + if atom[1] == "": + format_key1 = self.LOG_FORMAT_MAP[atom[0]] + m = getattr(AccessLogger, "_format_%s" % atom[0]) + key_method = KeyMethod(format_key1, m) + else: + format_key2 = (self.LOG_FORMAT_MAP[atom[2]], atom[1]) + m = getattr(AccessLogger, "_format_%s" % atom[2]) + key_method = KeyMethod(format_key2, functools.partial(m, atom[1])) + + methods.append(key_method) + + log_format = self.FORMAT_RE.sub(r"%s", log_format) + log_format = self.CLEANUP_RE.sub(r"%\1", log_format) + return log_format, methods + + @staticmethod + def _format_i( + key: str, request: BaseRequest, response: StreamResponse, time: float + ) -> str: + if request is None: + return "(no headers)" + + # suboptimal, make istr(key) once + return request.headers.get(key, "-") + + @staticmethod + def _format_o( + key: str, request: BaseRequest, response: StreamResponse, time: float + ) -> str: + # suboptimal, make istr(key) once + return response.headers.get(key, "-") + + @staticmethod + def _format_a(request: BaseRequest, response: StreamResponse, time: float) -> str: + if request is None: + return "-" + ip = request.remote + return ip if ip is not None else "-" + + @staticmethod + def _format_t(request: BaseRequest, response: StreamResponse, time: float) -> str: + now = datetime.datetime.utcnow() + start_time = now - datetime.timedelta(seconds=time) + return start_time.strftime("[%d/%b/%Y:%H:%M:%S +0000]") + + @staticmethod + def _format_P(request: BaseRequest, response: StreamResponse, time: float) -> str: + return "<%s>" % os.getpid() + + @staticmethod + def _format_r(request: BaseRequest, response: StreamResponse, time: float) -> str: + if request is None: + return "-" + return "{} {} HTTP/{}.{}".format( + request.method, + request.path_qs, + request.version.major, + request.version.minor, + ) + + @staticmethod + def _format_s(request: BaseRequest, response: StreamResponse, time: float) -> int: + return response.status + + @staticmethod + def _format_b(request: BaseRequest, response: StreamResponse, time: float) -> int: + return response.body_length + + @staticmethod + def _format_T(request: BaseRequest, response: StreamResponse, time: float) -> str: + return str(round(time)) + + @staticmethod + def _format_Tf(request: BaseRequest, response: StreamResponse, time: float) -> str: + return "%06f" % time + + @staticmethod + def _format_D(request: BaseRequest, response: StreamResponse, time: float) -> str: + return str(round(time * 1000000)) + + def _format_line( + self, request: BaseRequest, response: StreamResponse, time: float + ) -> Iterable[Tuple[str, Callable[[BaseRequest, StreamResponse, float], str]]]: + return [(key, method(request, response, time)) for key, method in self._methods] + + def log(self, request: BaseRequest, response: StreamResponse, time: float) -> None: + try: + fmt_info = self._format_line(request, response, time) + + values = list() + extra = dict() + for key, value in fmt_info: + values.append(value) + + if key.__class__ is str: + extra[key] = value + else: + k1, k2 = key # type: ignore[misc] + dct = extra.get(k1, {}) # type: ignore[var-annotated,has-type] + dct[k2] = value # type: ignore[index,has-type] + extra[k1] = dct # type: ignore[has-type,assignment] + + self.logger.info(self._log_format % tuple(values), extra=extra) + except Exception: + self.logger.exception("Error in logging") diff --git a/contrib/python/aiohttp/aiohttp/web_middlewares.py b/contrib/python/aiohttp/aiohttp/web_middlewares.py new file mode 100644 index 0000000000..fabcc449a2 --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/web_middlewares.py @@ -0,0 +1,119 @@ +import re +from typing import TYPE_CHECKING, Awaitable, Callable, Tuple, Type, TypeVar + +from .typedefs import Handler +from .web_exceptions import HTTPPermanentRedirect, _HTTPMove +from .web_request import Request +from .web_response import StreamResponse +from .web_urldispatcher import SystemRoute + +__all__ = ( + "middleware", + "normalize_path_middleware", +) + +if TYPE_CHECKING: # pragma: no cover + from .web_app import Application + +_Func = TypeVar("_Func") + + +async def _check_request_resolves(request: Request, path: str) -> Tuple[bool, Request]: + alt_request = request.clone(rel_url=path) + + match_info = await request.app.router.resolve(alt_request) + alt_request._match_info = match_info + + if match_info.http_exception is None: + return True, alt_request + + return False, request + + +def middleware(f: _Func) -> _Func: + f.__middleware_version__ = 1 # type: ignore[attr-defined] + return f + + +_Middleware = Callable[[Request, Handler], Awaitable[StreamResponse]] + + +def normalize_path_middleware( + *, + append_slash: bool = True, + remove_slash: bool = False, + merge_slashes: bool = True, + redirect_class: Type[_HTTPMove] = HTTPPermanentRedirect, +) -> _Middleware: + """Factory for producing a middleware that normalizes the path of a request. + + Normalizing means: + - Add or remove a trailing slash to the path. + - Double slashes are replaced by one. + + The middleware returns as soon as it finds a path that resolves + correctly. The order if both merge and append/remove are enabled is + 1) merge slashes + 2) append/remove slash + 3) both merge slashes and append/remove slash. + If the path resolves with at least one of those conditions, it will + redirect to the new path. + + Only one of `append_slash` and `remove_slash` can be enabled. If both + are `True` the factory will raise an assertion error + + If `append_slash` is `True` the middleware will append a slash when + needed. If a resource is defined with trailing slash and the request + comes without it, it will append it automatically. + + If `remove_slash` is `True`, `append_slash` must be `False`. When enabled + the middleware will remove trailing slashes and redirect if the resource + is defined + + If merge_slashes is True, merge multiple consecutive slashes in the + path into one. + """ + correct_configuration = not (append_slash and remove_slash) + assert correct_configuration, "Cannot both remove and append slash" + + @middleware + async def impl(request: Request, handler: Handler) -> StreamResponse: + if isinstance(request.match_info.route, SystemRoute): + paths_to_check = [] + if "?" in request.raw_path: + path, query = request.raw_path.split("?", 1) + query = "?" + query + else: + query = "" + path = request.raw_path + + if merge_slashes: + paths_to_check.append(re.sub("//+", "/", path)) + if append_slash and not request.path.endswith("/"): + paths_to_check.append(path + "/") + if remove_slash and request.path.endswith("/"): + paths_to_check.append(path[:-1]) + if merge_slashes and append_slash: + paths_to_check.append(re.sub("//+", "/", path + "/")) + if merge_slashes and remove_slash: + merged_slashes = re.sub("//+", "/", path) + paths_to_check.append(merged_slashes[:-1]) + + for path in paths_to_check: + path = re.sub("^//+", "/", path) # SECURITY: GHSA-v6wp-4m6f-gcjg + resolves, request = await _check_request_resolves(request, path) + if resolves: + raise redirect_class(request.raw_path + query) + + return await handler(request) + + return impl + + +def _fix_request_current_app(app: "Application") -> _Middleware: + @middleware + async def impl(request: Request, handler: Handler) -> StreamResponse: + with request.match_info.set_current_app(app): + return await handler(request) + + return impl diff --git a/contrib/python/aiohttp/aiohttp/web_protocol.py b/contrib/python/aiohttp/aiohttp/web_protocol.py new file mode 100644 index 0000000000..ad0c0498e3 --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/web_protocol.py @@ -0,0 +1,681 @@ +import asyncio +import asyncio.streams +import traceback +import warnings +from collections import deque +from contextlib import suppress +from html import escape as html_escape +from http import HTTPStatus +from logging import Logger +from typing import ( + TYPE_CHECKING, + Any, + Awaitable, + Callable, + Deque, + Optional, + Sequence, + Tuple, + Type, + Union, + cast, +) + +import attr +import yarl + +from .abc import AbstractAccessLogger, AbstractStreamWriter +from .base_protocol import BaseProtocol +from .helpers import ceil_timeout +from .http import ( + HttpProcessingError, + HttpRequestParser, + HttpVersion10, + RawRequestMessage, + StreamWriter, +) +from .log import access_logger, server_logger +from .streams import EMPTY_PAYLOAD, StreamReader +from .tcp_helpers import tcp_keepalive +from .web_exceptions import HTTPException +from .web_log import AccessLogger +from .web_request import BaseRequest +from .web_response import Response, StreamResponse + +__all__ = ("RequestHandler", "RequestPayloadError", "PayloadAccessError") + +if TYPE_CHECKING: # pragma: no cover + from .web_server import Server + + +_RequestFactory = Callable[ + [ + RawRequestMessage, + StreamReader, + "RequestHandler", + AbstractStreamWriter, + "asyncio.Task[None]", + ], + BaseRequest, +] + +_RequestHandler = Callable[[BaseRequest], Awaitable[StreamResponse]] + +ERROR = RawRequestMessage( + "UNKNOWN", + "/", + HttpVersion10, + {}, # type: ignore[arg-type] + {}, # type: ignore[arg-type] + True, + None, + False, + False, + yarl.URL("/"), +) + + +class RequestPayloadError(Exception): + """Payload parsing error.""" + + +class PayloadAccessError(Exception): + """Payload was accessed after response was sent.""" + + +@attr.s(auto_attribs=True, frozen=True, slots=True) +class _ErrInfo: + status: int + exc: BaseException + message: str + + +_MsgType = Tuple[Union[RawRequestMessage, _ErrInfo], StreamReader] + + +class RequestHandler(BaseProtocol): + """HTTP protocol implementation. + + RequestHandler handles incoming HTTP request. It reads request line, + request headers and request payload and calls handle_request() method. + By default it always returns with 404 response. + + RequestHandler handles errors in incoming request, like bad + status line, bad headers or incomplete payload. If any error occurs, + connection gets closed. + + keepalive_timeout -- number of seconds before closing + keep-alive connection + + tcp_keepalive -- TCP keep-alive is on, default is on + + debug -- enable debug mode + + logger -- custom logger object + + access_log_class -- custom class for access_logger + + access_log -- custom logging object + + access_log_format -- access log format string + + loop -- Optional event loop + + max_line_size -- Optional maximum header line size + + max_field_size -- Optional maximum header field size + + max_headers -- Optional maximum header size + + """ + + KEEPALIVE_RESCHEDULE_DELAY = 1 + + __slots__ = ( + "_request_count", + "_keepalive", + "_manager", + "_request_handler", + "_request_factory", + "_tcp_keepalive", + "_keepalive_time", + "_keepalive_handle", + "_keepalive_timeout", + "_lingering_time", + "_messages", + "_message_tail", + "_waiter", + "_task_handler", + "_upgrade", + "_payload_parser", + "_request_parser", + "_reading_paused", + "logger", + "debug", + "access_log", + "access_logger", + "_close", + "_force_close", + "_current_request", + ) + + def __init__( + self, + manager: "Server", + *, + loop: asyncio.AbstractEventLoop, + keepalive_timeout: float = 75.0, # NGINX default is 75 secs + tcp_keepalive: bool = True, + logger: Logger = server_logger, + access_log_class: Type[AbstractAccessLogger] = AccessLogger, + access_log: Logger = access_logger, + access_log_format: str = AccessLogger.LOG_FORMAT, + debug: bool = False, + max_line_size: int = 8190, + max_headers: int = 32768, + max_field_size: int = 8190, + lingering_time: float = 10.0, + read_bufsize: int = 2 ** 16, + auto_decompress: bool = True, + ): + super().__init__(loop) + + self._request_count = 0 + self._keepalive = False + self._current_request = None # type: Optional[BaseRequest] + self._manager = manager # type: Optional[Server] + self._request_handler: Optional[_RequestHandler] = manager.request_handler + self._request_factory: Optional[_RequestFactory] = manager.request_factory + + self._tcp_keepalive = tcp_keepalive + # placeholder to be replaced on keepalive timeout setup + self._keepalive_time = 0.0 + self._keepalive_handle = None # type: Optional[asyncio.Handle] + self._keepalive_timeout = keepalive_timeout + self._lingering_time = float(lingering_time) + + self._messages: Deque[_MsgType] = deque() + self._message_tail = b"" + + self._waiter = None # type: Optional[asyncio.Future[None]] + self._task_handler = None # type: Optional[asyncio.Task[None]] + + self._upgrade = False + self._payload_parser = None # type: Any + self._request_parser = HttpRequestParser( + self, + loop, + read_bufsize, + max_line_size=max_line_size, + max_field_size=max_field_size, + max_headers=max_headers, + payload_exception=RequestPayloadError, + auto_decompress=auto_decompress, + ) # type: Optional[HttpRequestParser] + + self.logger = logger + self.debug = debug + self.access_log = access_log + if access_log: + self.access_logger = access_log_class( + access_log, access_log_format + ) # type: Optional[AbstractAccessLogger] + else: + self.access_logger = None + + self._close = False + self._force_close = False + + def __repr__(self) -> str: + return "<{} {}>".format( + self.__class__.__name__, + "connected" if self.transport is not None else "disconnected", + ) + + @property + def keepalive_timeout(self) -> float: + return self._keepalive_timeout + + async def shutdown(self, timeout: Optional[float] = 15.0) -> None: + """Do worker process exit preparations. + + We need to clean up everything and stop accepting requests. + It is especially important for keep-alive connections. + """ + self._force_close = True + + if self._keepalive_handle is not None: + self._keepalive_handle.cancel() + + if self._waiter: + self._waiter.cancel() + + # wait for handlers + with suppress(asyncio.CancelledError, asyncio.TimeoutError): + async with ceil_timeout(timeout): + if self._current_request is not None: + self._current_request._cancel(asyncio.CancelledError()) + + if self._task_handler is not None and not self._task_handler.done(): + await self._task_handler + + # force-close non-idle handler + if self._task_handler is not None: + self._task_handler.cancel() + + if self.transport is not None: + self.transport.close() + self.transport = None + + def connection_made(self, transport: asyncio.BaseTransport) -> None: + super().connection_made(transport) + + real_transport = cast(asyncio.Transport, transport) + if self._tcp_keepalive: + tcp_keepalive(real_transport) + + self._task_handler = self._loop.create_task(self.start()) + assert self._manager is not None + self._manager.connection_made(self, real_transport) + + def connection_lost(self, exc: Optional[BaseException]) -> None: + if self._manager is None: + return + self._manager.connection_lost(self, exc) + + super().connection_lost(exc) + + self._manager = None + self._force_close = True + self._request_factory = None + self._request_handler = None + self._request_parser = None + + if self._keepalive_handle is not None: + self._keepalive_handle.cancel() + + if self._current_request is not None: + if exc is None: + exc = ConnectionResetError("Connection lost") + self._current_request._cancel(exc) + + if self._task_handler is not None: + self._task_handler.cancel() + if self._waiter is not None: + self._waiter.cancel() + + self._task_handler = None + + if self._payload_parser is not None: + self._payload_parser.feed_eof() + self._payload_parser = None + + def set_parser(self, parser: Any) -> None: + # Actual type is WebReader + assert self._payload_parser is None + + self._payload_parser = parser + + if self._message_tail: + self._payload_parser.feed_data(self._message_tail) + self._message_tail = b"" + + def eof_received(self) -> None: + pass + + def data_received(self, data: bytes) -> None: + if self._force_close or self._close: + return + # parse http messages + messages: Sequence[_MsgType] + if self._payload_parser is None and not self._upgrade: + assert self._request_parser is not None + try: + messages, upgraded, tail = self._request_parser.feed_data(data) + except HttpProcessingError as exc: + messages = [ + (_ErrInfo(status=400, exc=exc, message=exc.message), EMPTY_PAYLOAD) + ] + upgraded = False + tail = b"" + + for msg, payload in messages or (): + self._request_count += 1 + self._messages.append((msg, payload)) + + waiter = self._waiter + if messages and waiter is not None and not waiter.done(): + # don't set result twice + waiter.set_result(None) + + self._upgrade = upgraded + if upgraded and tail: + self._message_tail = tail + + # no parser, just store + elif self._payload_parser is None and self._upgrade and data: + self._message_tail += data + + # feed payload + elif data: + eof, tail = self._payload_parser.feed_data(data) + if eof: + self.close() + + def keep_alive(self, val: bool) -> None: + """Set keep-alive connection mode. + + :param bool val: new state. + """ + self._keepalive = val + if self._keepalive_handle: + self._keepalive_handle.cancel() + self._keepalive_handle = None + + def close(self) -> None: + """Close connection. + + Stop accepting new pipelining messages and close + connection when handlers done processing messages. + """ + self._close = True + if self._waiter: + self._waiter.cancel() + + def force_close(self) -> None: + """Forcefully close connection.""" + self._force_close = True + if self._waiter: + self._waiter.cancel() + if self.transport is not None: + self.transport.close() + self.transport = None + + def log_access( + self, request: BaseRequest, response: StreamResponse, time: float + ) -> None: + if self.access_logger is not None: + self.access_logger.log(request, response, self._loop.time() - time) + + def log_debug(self, *args: Any, **kw: Any) -> None: + if self.debug: + self.logger.debug(*args, **kw) + + def log_exception(self, *args: Any, **kw: Any) -> None: + self.logger.exception(*args, **kw) + + def _process_keepalive(self) -> None: + if self._force_close or not self._keepalive: + return + + next = self._keepalive_time + self._keepalive_timeout + + # handler in idle state + if self._waiter: + if self._loop.time() > next: + self.force_close() + return + + # not all request handlers are done, + # reschedule itself to next second + self._keepalive_handle = self._loop.call_later( + self.KEEPALIVE_RESCHEDULE_DELAY, self._process_keepalive + ) + + async def _handle_request( + self, + request: BaseRequest, + start_time: float, + request_handler: Callable[[BaseRequest], Awaitable[StreamResponse]], + ) -> Tuple[StreamResponse, bool]: + assert self._request_handler is not None + try: + try: + self._current_request = request + resp = await request_handler(request) + finally: + self._current_request = None + except HTTPException as exc: + resp = exc + reset = await self.finish_response(request, resp, start_time) + except asyncio.CancelledError: + raise + except asyncio.TimeoutError as exc: + self.log_debug("Request handler timed out.", exc_info=exc) + resp = self.handle_error(request, 504) + reset = await self.finish_response(request, resp, start_time) + except Exception as exc: + resp = self.handle_error(request, 500, exc) + reset = await self.finish_response(request, resp, start_time) + else: + # Deprecation warning (See #2415) + if getattr(resp, "__http_exception__", False): + warnings.warn( + "returning HTTPException object is deprecated " + "(#2415) and will be removed, " + "please raise the exception instead", + DeprecationWarning, + ) + + reset = await self.finish_response(request, resp, start_time) + + return resp, reset + + async def start(self) -> None: + """Process incoming request. + + It reads request line, request headers and request payload, then + calls handle_request() method. Subclass has to override + handle_request(). start() handles various exceptions in request + or response handling. Connection is being closed always unless + keep_alive(True) specified. + """ + loop = self._loop + handler = self._task_handler + assert handler is not None + manager = self._manager + assert manager is not None + keepalive_timeout = self._keepalive_timeout + resp = None + assert self._request_factory is not None + assert self._request_handler is not None + + while not self._force_close: + if not self._messages: + try: + # wait for next request + self._waiter = loop.create_future() + await self._waiter + except asyncio.CancelledError: + break + finally: + self._waiter = None + + message, payload = self._messages.popleft() + + start = loop.time() + + manager.requests_count += 1 + writer = StreamWriter(self, loop) + if isinstance(message, _ErrInfo): + # make request_factory work + request_handler = self._make_error_handler(message) + message = ERROR + else: + request_handler = self._request_handler + + request = self._request_factory(message, payload, self, writer, handler) + try: + # a new task is used for copy context vars (#3406) + task = self._loop.create_task( + self._handle_request(request, start, request_handler) + ) + try: + resp, reset = await task + except (asyncio.CancelledError, ConnectionError): + self.log_debug("Ignored premature client disconnection") + break + + # Drop the processed task from asyncio.Task.all_tasks() early + del task + if reset: + self.log_debug("Ignored premature client disconnection 2") + break + + # notify server about keep-alive + self._keepalive = bool(resp.keep_alive) + + # check payload + if not payload.is_eof(): + lingering_time = self._lingering_time + if not self._force_close and lingering_time: + self.log_debug( + "Start lingering close timer for %s sec.", lingering_time + ) + + now = loop.time() + end_t = now + lingering_time + + with suppress(asyncio.TimeoutError, asyncio.CancelledError): + while not payload.is_eof() and now < end_t: + async with ceil_timeout(end_t - now): + # read and ignore + await payload.readany() + now = loop.time() + + # if payload still uncompleted + if not payload.is_eof() and not self._force_close: + self.log_debug("Uncompleted request.") + self.close() + + payload.set_exception(PayloadAccessError()) + + except asyncio.CancelledError: + self.log_debug("Ignored premature client disconnection ") + break + except RuntimeError as exc: + if self.debug: + self.log_exception("Unhandled runtime exception", exc_info=exc) + self.force_close() + except Exception as exc: + self.log_exception("Unhandled exception", exc_info=exc) + self.force_close() + finally: + if self.transport is None and resp is not None: + self.log_debug("Ignored premature client disconnection.") + elif not self._force_close: + if self._keepalive and not self._close: + # start keep-alive timer + if keepalive_timeout is not None: + now = self._loop.time() + self._keepalive_time = now + if self._keepalive_handle is None: + self._keepalive_handle = loop.call_at( + now + keepalive_timeout, self._process_keepalive + ) + else: + break + + # remove handler, close transport if no handlers left + if not self._force_close: + self._task_handler = None + if self.transport is not None: + self.transport.close() + + async def finish_response( + self, request: BaseRequest, resp: StreamResponse, start_time: float + ) -> bool: + """Prepare the response and write_eof, then log access. + + This has to + be called within the context of any exception so the access logger + can get exception information. Returns True if the client disconnects + prematurely. + """ + if self._request_parser is not None: + self._request_parser.set_upgraded(False) + self._upgrade = False + if self._message_tail: + self._request_parser.feed_data(self._message_tail) + self._message_tail = b"" + try: + prepare_meth = resp.prepare + except AttributeError: + if resp is None: + raise RuntimeError("Missing return " "statement on request handler") + else: + raise RuntimeError( + "Web-handler should return " + "a response instance, " + "got {!r}".format(resp) + ) + try: + await prepare_meth(request) + await resp.write_eof() + except ConnectionError: + self.log_access(request, resp, start_time) + return True + else: + self.log_access(request, resp, start_time) + return False + + def handle_error( + self, + request: BaseRequest, + status: int = 500, + exc: Optional[BaseException] = None, + message: Optional[str] = None, + ) -> StreamResponse: + """Handle errors. + + Returns HTTP response with specific status code. Logs additional + information. It always closes current connection. + """ + self.log_exception("Error handling request", exc_info=exc) + + # some data already got sent, connection is broken + if request.writer.output_size > 0: + raise ConnectionError( + "Response is sent already, cannot send another response " + "with the error message" + ) + + ct = "text/plain" + if status == HTTPStatus.INTERNAL_SERVER_ERROR: + title = "{0.value} {0.phrase}".format(HTTPStatus.INTERNAL_SERVER_ERROR) + msg = HTTPStatus.INTERNAL_SERVER_ERROR.description + tb = None + if self.debug: + with suppress(Exception): + tb = traceback.format_exc() + + if "text/html" in request.headers.get("Accept", ""): + if tb: + tb = html_escape(tb) + msg = f"<h2>Traceback:</h2>\n<pre>{tb}</pre>" + message = ( + "<html><head>" + "<title>{title}</title>" + "</head><body>\n<h1>{title}</h1>" + "\n{msg}\n</body></html>\n" + ).format(title=title, msg=msg) + ct = "text/html" + else: + if tb: + msg = tb + message = title + "\n\n" + msg + + resp = Response(status=status, text=message, content_type=ct) + resp.force_close() + + return resp + + def _make_error_handler( + self, err_info: _ErrInfo + ) -> Callable[[BaseRequest], Awaitable[StreamResponse]]: + async def handler(request: BaseRequest) -> StreamResponse: + return self.handle_error( + request, err_info.status, err_info.exc, err_info.message + ) + + return handler diff --git a/contrib/python/aiohttp/aiohttp/web_request.py b/contrib/python/aiohttp/aiohttp/web_request.py new file mode 100644 index 0000000000..b3574cafb3 --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/web_request.py @@ -0,0 +1,874 @@ +import asyncio +import datetime +import io +import re +import socket +import string +import tempfile +import types +import warnings +from http.cookies import SimpleCookie +from types import MappingProxyType +from typing import ( + TYPE_CHECKING, + Any, + Dict, + Iterator, + Mapping, + MutableMapping, + Optional, + Pattern, + Tuple, + Union, + cast, +) +from urllib.parse import parse_qsl + +import attr +from multidict import CIMultiDict, CIMultiDictProxy, MultiDict, MultiDictProxy +from yarl import URL + +from . import hdrs +from .abc import AbstractStreamWriter +from .helpers import ( + DEBUG, + ETAG_ANY, + LIST_QUOTED_ETAG_RE, + ChainMapProxy, + ETag, + HeadersMixin, + parse_http_date, + reify, + sentinel, +) +from .http_parser import RawRequestMessage +from .http_writer import HttpVersion +from .multipart import BodyPartReader, MultipartReader +from .streams import EmptyStreamReader, StreamReader +from .typedefs import ( + DEFAULT_JSON_DECODER, + Final, + JSONDecoder, + LooseHeaders, + RawHeaders, + StrOrURL, +) +from .web_exceptions import HTTPRequestEntityTooLarge +from .web_response import StreamResponse + +__all__ = ("BaseRequest", "FileField", "Request") + + +if TYPE_CHECKING: # pragma: no cover + from .web_app import Application + from .web_protocol import RequestHandler + from .web_urldispatcher import UrlMappingMatchInfo + + +@attr.s(auto_attribs=True, frozen=True, slots=True) +class FileField: + name: str + filename: str + file: io.BufferedReader + content_type: str + headers: "CIMultiDictProxy[str]" + + +_TCHAR: Final[str] = string.digits + string.ascii_letters + r"!#$%&'*+.^_`|~-" +# '-' at the end to prevent interpretation as range in a char class + +_TOKEN: Final[str] = fr"[{_TCHAR}]+" + +_QDTEXT: Final[str] = r"[{}]".format( + r"".join(chr(c) for c in (0x09, 0x20, 0x21) + tuple(range(0x23, 0x7F))) +) +# qdtext includes 0x5C to escape 0x5D ('\]') +# qdtext excludes obs-text (because obsoleted, and encoding not specified) + +_QUOTED_PAIR: Final[str] = r"\\[\t !-~]" + +_QUOTED_STRING: Final[str] = r'"(?:{quoted_pair}|{qdtext})*"'.format( + qdtext=_QDTEXT, quoted_pair=_QUOTED_PAIR +) + +_FORWARDED_PAIR: Final[ + str +] = r"({token})=({token}|{quoted_string})(:\d{{1,4}})?".format( + token=_TOKEN, quoted_string=_QUOTED_STRING +) + +_QUOTED_PAIR_REPLACE_RE: Final[Pattern[str]] = re.compile(r"\\([\t !-~])") +# same pattern as _QUOTED_PAIR but contains a capture group + +_FORWARDED_PAIR_RE: Final[Pattern[str]] = re.compile(_FORWARDED_PAIR) + +############################################################ +# HTTP Request +############################################################ + + +class BaseRequest(MutableMapping[str, Any], HeadersMixin): + + POST_METHODS = { + hdrs.METH_PATCH, + hdrs.METH_POST, + hdrs.METH_PUT, + hdrs.METH_TRACE, + hdrs.METH_DELETE, + } + + ATTRS = HeadersMixin.ATTRS | frozenset( + [ + "_message", + "_protocol", + "_payload_writer", + "_payload", + "_headers", + "_method", + "_version", + "_rel_url", + "_post", + "_read_bytes", + "_state", + "_cache", + "_task", + "_client_max_size", + "_loop", + "_transport_sslcontext", + "_transport_peername", + ] + ) + + def __init__( + self, + message: RawRequestMessage, + payload: StreamReader, + protocol: "RequestHandler", + payload_writer: AbstractStreamWriter, + task: "asyncio.Task[None]", + loop: asyncio.AbstractEventLoop, + *, + client_max_size: int = 1024 ** 2, + state: Optional[Dict[str, Any]] = None, + scheme: Optional[str] = None, + host: Optional[str] = None, + remote: Optional[str] = None, + ) -> None: + if state is None: + state = {} + self._message = message + self._protocol = protocol + self._payload_writer = payload_writer + + self._payload = payload + self._headers = message.headers + self._method = message.method + self._version = message.version + self._rel_url = message.url + self._post = ( + None + ) # type: Optional[MultiDictProxy[Union[str, bytes, FileField]]] + self._read_bytes = None # type: Optional[bytes] + + self._state = state + self._cache = {} # type: Dict[str, Any] + self._task = task + self._client_max_size = client_max_size + self._loop = loop + + transport = self._protocol.transport + assert transport is not None + self._transport_sslcontext = transport.get_extra_info("sslcontext") + self._transport_peername = transport.get_extra_info("peername") + + if scheme is not None: + self._cache["scheme"] = scheme + if host is not None: + self._cache["host"] = host + if remote is not None: + self._cache["remote"] = remote + + def clone( + self, + *, + method: str = sentinel, + rel_url: StrOrURL = sentinel, + headers: LooseHeaders = sentinel, + scheme: str = sentinel, + host: str = sentinel, + remote: str = sentinel, + ) -> "BaseRequest": + """Clone itself with replacement some attributes. + + Creates and returns a new instance of Request object. If no parameters + are given, an exact copy is returned. If a parameter is not passed, it + will reuse the one from the current request object. + """ + if self._read_bytes: + raise RuntimeError("Cannot clone request " "after reading its content") + + dct = {} # type: Dict[str, Any] + if method is not sentinel: + dct["method"] = method + if rel_url is not sentinel: + new_url = URL(rel_url) + dct["url"] = new_url + dct["path"] = str(new_url) + if headers is not sentinel: + # a copy semantic + dct["headers"] = CIMultiDictProxy(CIMultiDict(headers)) + dct["raw_headers"] = tuple( + (k.encode("utf-8"), v.encode("utf-8")) for k, v in headers.items() + ) + + message = self._message._replace(**dct) + + kwargs = {} + if scheme is not sentinel: + kwargs["scheme"] = scheme + if host is not sentinel: + kwargs["host"] = host + if remote is not sentinel: + kwargs["remote"] = remote + + return self.__class__( + message, + self._payload, + self._protocol, + self._payload_writer, + self._task, + self._loop, + client_max_size=self._client_max_size, + state=self._state.copy(), + **kwargs, + ) + + @property + def task(self) -> "asyncio.Task[None]": + return self._task + + @property + def protocol(self) -> "RequestHandler": + return self._protocol + + @property + def transport(self) -> Optional[asyncio.Transport]: + if self._protocol is None: + return None + return self._protocol.transport + + @property + def writer(self) -> AbstractStreamWriter: + return self._payload_writer + + @reify + def message(self) -> RawRequestMessage: + warnings.warn("Request.message is deprecated", DeprecationWarning, stacklevel=3) + return self._message + + @reify + def rel_url(self) -> URL: + return self._rel_url + + @reify + def loop(self) -> asyncio.AbstractEventLoop: + warnings.warn( + "request.loop property is deprecated", DeprecationWarning, stacklevel=2 + ) + return self._loop + + # MutableMapping API + + def __getitem__(self, key: str) -> Any: + return self._state[key] + + def __setitem__(self, key: str, value: Any) -> None: + self._state[key] = value + + def __delitem__(self, key: str) -> None: + del self._state[key] + + def __len__(self) -> int: + return len(self._state) + + def __iter__(self) -> Iterator[str]: + return iter(self._state) + + ######## + + @reify + def secure(self) -> bool: + """A bool indicating if the request is handled with SSL.""" + return self.scheme == "https" + + @reify + def forwarded(self) -> Tuple[Mapping[str, str], ...]: + """A tuple containing all parsed Forwarded header(s). + + Makes an effort to parse Forwarded headers as specified by RFC 7239: + + - It adds one (immutable) dictionary per Forwarded 'field-value', ie + per proxy. The element corresponds to the data in the Forwarded + field-value added by the first proxy encountered by the client. Each + subsequent item corresponds to those added by later proxies. + - It checks that every value has valid syntax in general as specified + in section 4: either a 'token' or a 'quoted-string'. + - It un-escapes found escape sequences. + - It does NOT validate 'by' and 'for' contents as specified in section + 6. + - It does NOT validate 'host' contents (Host ABNF). + - It does NOT validate 'proto' contents for valid URI scheme names. + + Returns a tuple containing one or more immutable dicts + """ + elems = [] + for field_value in self._message.headers.getall(hdrs.FORWARDED, ()): + length = len(field_value) + pos = 0 + need_separator = False + elem = {} # type: Dict[str, str] + elems.append(types.MappingProxyType(elem)) + while 0 <= pos < length: + match = _FORWARDED_PAIR_RE.match(field_value, pos) + if match is not None: # got a valid forwarded-pair + if need_separator: + # bad syntax here, skip to next comma + pos = field_value.find(",", pos) + else: + name, value, port = match.groups() + if value[0] == '"': + # quoted string: remove quotes and unescape + value = _QUOTED_PAIR_REPLACE_RE.sub(r"\1", value[1:-1]) + if port: + value += port + elem[name.lower()] = value + pos += len(match.group(0)) + need_separator = True + elif field_value[pos] == ",": # next forwarded-element + need_separator = False + elem = {} + elems.append(types.MappingProxyType(elem)) + pos += 1 + elif field_value[pos] == ";": # next forwarded-pair + need_separator = False + pos += 1 + elif field_value[pos] in " \t": + # Allow whitespace even between forwarded-pairs, though + # RFC 7239 doesn't. This simplifies code and is in line + # with Postel's law. + pos += 1 + else: + # bad syntax here, skip to next comma + pos = field_value.find(",", pos) + return tuple(elems) + + @reify + def scheme(self) -> str: + """A string representing the scheme of the request. + + Hostname is resolved in this order: + + - overridden value by .clone(scheme=new_scheme) call. + - type of connection to peer: HTTPS if socket is SSL, HTTP otherwise. + + 'http' or 'https'. + """ + if self._transport_sslcontext: + return "https" + else: + return "http" + + @reify + def method(self) -> str: + """Read only property for getting HTTP method. + + The value is upper-cased str like 'GET', 'POST', 'PUT' etc. + """ + return self._method + + @reify + def version(self) -> HttpVersion: + """Read only property for getting HTTP version of request. + + Returns aiohttp.protocol.HttpVersion instance. + """ + return self._version + + @reify + def host(self) -> str: + """Hostname of the request. + + Hostname is resolved in this order: + + - overridden value by .clone(host=new_host) call. + - HOST HTTP header + - socket.getfqdn() value + """ + host = self._message.headers.get(hdrs.HOST) + if host is not None: + return host + return socket.getfqdn() + + @reify + def remote(self) -> Optional[str]: + """Remote IP of client initiated HTTP request. + + The IP is resolved in this order: + + - overridden value by .clone(remote=new_remote) call. + - peername of opened socket + """ + if self._transport_peername is None: + return None + if isinstance(self._transport_peername, (list, tuple)): + return str(self._transport_peername[0]) + return str(self._transport_peername) + + @reify + def url(self) -> URL: + url = URL.build(scheme=self.scheme, host=self.host) + return url.join(self._rel_url) + + @reify + def path(self) -> str: + """The URL including *PATH INFO* without the host or scheme. + + E.g., ``/app/blog`` + """ + return self._rel_url.path + + @reify + def path_qs(self) -> str: + """The URL including PATH_INFO and the query string. + + E.g, /app/blog?id=10 + """ + return str(self._rel_url) + + @reify + def raw_path(self) -> str: + """The URL including raw *PATH INFO* without the host or scheme. + + Warning, the path is unquoted and may contains non valid URL characters + + E.g., ``/my%2Fpath%7Cwith%21some%25strange%24characters`` + """ + return self._message.path + + @reify + def query(self) -> "MultiDictProxy[str]": + """A multidict with all the variables in the query string.""" + return MultiDictProxy(self._rel_url.query) + + @reify + def query_string(self) -> str: + """The query string in the URL. + + E.g., id=10 + """ + return self._rel_url.query_string + + @reify + def headers(self) -> "CIMultiDictProxy[str]": + """A case-insensitive multidict proxy with all headers.""" + return self._headers + + @reify + def raw_headers(self) -> RawHeaders: + """A sequence of pairs for all headers.""" + return self._message.raw_headers + + @reify + def if_modified_since(self) -> Optional[datetime.datetime]: + """The value of If-Modified-Since HTTP header, or None. + + This header is represented as a `datetime` object. + """ + return parse_http_date(self.headers.get(hdrs.IF_MODIFIED_SINCE)) + + @reify + def if_unmodified_since(self) -> Optional[datetime.datetime]: + """The value of If-Unmodified-Since HTTP header, or None. + + This header is represented as a `datetime` object. + """ + return parse_http_date(self.headers.get(hdrs.IF_UNMODIFIED_SINCE)) + + @staticmethod + def _etag_values(etag_header: str) -> Iterator[ETag]: + """Extract `ETag` objects from raw header.""" + if etag_header == ETAG_ANY: + yield ETag( + is_weak=False, + value=ETAG_ANY, + ) + else: + for match in LIST_QUOTED_ETAG_RE.finditer(etag_header): + is_weak, value, garbage = match.group(2, 3, 4) + # Any symbol captured by 4th group means + # that the following sequence is invalid. + if garbage: + break + + yield ETag( + is_weak=bool(is_weak), + value=value, + ) + + @classmethod + def _if_match_or_none_impl( + cls, header_value: Optional[str] + ) -> Optional[Tuple[ETag, ...]]: + if not header_value: + return None + + return tuple(cls._etag_values(header_value)) + + @reify + def if_match(self) -> Optional[Tuple[ETag, ...]]: + """The value of If-Match HTTP header, or None. + + This header is represented as a `tuple` of `ETag` objects. + """ + return self._if_match_or_none_impl(self.headers.get(hdrs.IF_MATCH)) + + @reify + def if_none_match(self) -> Optional[Tuple[ETag, ...]]: + """The value of If-None-Match HTTP header, or None. + + This header is represented as a `tuple` of `ETag` objects. + """ + return self._if_match_or_none_impl(self.headers.get(hdrs.IF_NONE_MATCH)) + + @reify + def if_range(self) -> Optional[datetime.datetime]: + """The value of If-Range HTTP header, or None. + + This header is represented as a `datetime` object. + """ + return parse_http_date(self.headers.get(hdrs.IF_RANGE)) + + @reify + def keep_alive(self) -> bool: + """Is keepalive enabled by client?""" + return not self._message.should_close + + @reify + def cookies(self) -> Mapping[str, str]: + """Return request cookies. + + A read-only dictionary-like object. + """ + raw = self.headers.get(hdrs.COOKIE, "") + parsed = SimpleCookie(raw) # type: SimpleCookie[str] + return MappingProxyType({key: val.value for key, val in parsed.items()}) + + @reify + def http_range(self) -> slice: + """The content of Range HTTP header. + + Return a slice instance. + + """ + rng = self._headers.get(hdrs.RANGE) + start, end = None, None + if rng is not None: + try: + pattern = r"^bytes=(\d*)-(\d*)$" + start, end = re.findall(pattern, rng)[0] + except IndexError: # pattern was not found in header + raise ValueError("range not in acceptable format") + + end = int(end) if end else None + start = int(start) if start else None + + if start is None and end is not None: + # end with no start is to return tail of content + start = -end + end = None + + if start is not None and end is not None: + # end is inclusive in range header, exclusive for slice + end += 1 + + if start >= end: + raise ValueError("start cannot be after end") + + if start is end is None: # No valid range supplied + raise ValueError("No start or end of range specified") + + return slice(start, end, 1) + + @reify + def content(self) -> StreamReader: + """Return raw payload stream.""" + return self._payload + + @property + def has_body(self) -> bool: + """Return True if request's HTTP BODY can be read, False otherwise.""" + warnings.warn( + "Deprecated, use .can_read_body #2005", DeprecationWarning, stacklevel=2 + ) + return not self._payload.at_eof() + + @property + def can_read_body(self) -> bool: + """Return True if request's HTTP BODY can be read, False otherwise.""" + return not self._payload.at_eof() + + @reify + def body_exists(self) -> bool: + """Return True if request has HTTP BODY, False otherwise.""" + return type(self._payload) is not EmptyStreamReader + + async def release(self) -> None: + """Release request. + + Eat unread part of HTTP BODY if present. + """ + while not self._payload.at_eof(): + await self._payload.readany() + + async def read(self) -> bytes: + """Read request body if present. + + Returns bytes object with full request content. + """ + if self._read_bytes is None: + body = bytearray() + while True: + chunk = await self._payload.readany() + body.extend(chunk) + if self._client_max_size: + body_size = len(body) + if body_size >= self._client_max_size: + raise HTTPRequestEntityTooLarge( + max_size=self._client_max_size, actual_size=body_size + ) + if not chunk: + break + self._read_bytes = bytes(body) + return self._read_bytes + + async def text(self) -> str: + """Return BODY as text using encoding from .charset.""" + bytes_body = await self.read() + encoding = self.charset or "utf-8" + return bytes_body.decode(encoding) + + async def json(self, *, loads: JSONDecoder = DEFAULT_JSON_DECODER) -> Any: + """Return BODY as JSON.""" + body = await self.text() + return loads(body) + + async def multipart(self) -> MultipartReader: + """Return async iterator to process BODY as multipart.""" + return MultipartReader(self._headers, self._payload) + + async def post(self) -> "MultiDictProxy[Union[str, bytes, FileField]]": + """Return POST parameters.""" + if self._post is not None: + return self._post + if self._method not in self.POST_METHODS: + self._post = MultiDictProxy(MultiDict()) + return self._post + + content_type = self.content_type + if content_type not in ( + "", + "application/x-www-form-urlencoded", + "multipart/form-data", + ): + self._post = MultiDictProxy(MultiDict()) + return self._post + + out = MultiDict() # type: MultiDict[Union[str, bytes, FileField]] + + if content_type == "multipart/form-data": + multipart = await self.multipart() + max_size = self._client_max_size + + field = await multipart.next() + while field is not None: + size = 0 + field_ct = field.headers.get(hdrs.CONTENT_TYPE) + + if isinstance(field, BodyPartReader): + assert field.name is not None + + # Note that according to RFC 7578, the Content-Type header + # is optional, even for files, so we can't assume it's + # present. + # https://tools.ietf.org/html/rfc7578#section-4.4 + if field.filename: + # store file in temp file + tmp = tempfile.TemporaryFile() + chunk = await field.read_chunk(size=2 ** 16) + while chunk: + chunk = field.decode(chunk) + tmp.write(chunk) + size += len(chunk) + if 0 < max_size < size: + tmp.close() + raise HTTPRequestEntityTooLarge( + max_size=max_size, actual_size=size + ) + chunk = await field.read_chunk(size=2 ** 16) + tmp.seek(0) + + if field_ct is None: + field_ct = "application/octet-stream" + + ff = FileField( + field.name, + field.filename, + cast(io.BufferedReader, tmp), + field_ct, + field.headers, + ) + out.add(field.name, ff) + else: + # deal with ordinary data + value = await field.read(decode=True) + if field_ct is None or field_ct.startswith("text/"): + charset = field.get_charset(default="utf-8") + out.add(field.name, value.decode(charset)) + else: + out.add(field.name, value) + size += len(value) + if 0 < max_size < size: + raise HTTPRequestEntityTooLarge( + max_size=max_size, actual_size=size + ) + else: + raise ValueError( + "To decode nested multipart you need " "to use custom reader", + ) + + field = await multipart.next() + else: + data = await self.read() + if data: + charset = self.charset or "utf-8" + out.extend( + parse_qsl( + data.rstrip().decode(charset), + keep_blank_values=True, + encoding=charset, + ) + ) + + self._post = MultiDictProxy(out) + return self._post + + def get_extra_info(self, name: str, default: Any = None) -> Any: + """Extra info from protocol transport""" + protocol = self._protocol + if protocol is None: + return default + + transport = protocol.transport + if transport is None: + return default + + return transport.get_extra_info(name, default) + + def __repr__(self) -> str: + ascii_encodable_path = self.path.encode("ascii", "backslashreplace").decode( + "ascii" + ) + return "<{} {} {} >".format( + self.__class__.__name__, self._method, ascii_encodable_path + ) + + def __eq__(self, other: object) -> bool: + return id(self) == id(other) + + def __bool__(self) -> bool: + return True + + async def _prepare_hook(self, response: StreamResponse) -> None: + return + + def _cancel(self, exc: BaseException) -> None: + self._payload.set_exception(exc) + + +class Request(BaseRequest): + + ATTRS = BaseRequest.ATTRS | frozenset(["_match_info"]) + + def __init__(self, *args: Any, **kwargs: Any) -> None: + super().__init__(*args, **kwargs) + + # matchdict, route_name, handler + # or information about traversal lookup + + # initialized after route resolving + self._match_info = None # type: Optional[UrlMappingMatchInfo] + + if DEBUG: + + def __setattr__(self, name: str, val: Any) -> None: + if name not in self.ATTRS: + warnings.warn( + "Setting custom {}.{} attribute " + "is discouraged".format(self.__class__.__name__, name), + DeprecationWarning, + stacklevel=2, + ) + super().__setattr__(name, val) + + def clone( + self, + *, + method: str = sentinel, + rel_url: StrOrURL = sentinel, + headers: LooseHeaders = sentinel, + scheme: str = sentinel, + host: str = sentinel, + remote: str = sentinel, + ) -> "Request": + ret = super().clone( + method=method, + rel_url=rel_url, + headers=headers, + scheme=scheme, + host=host, + remote=remote, + ) + new_ret = cast(Request, ret) + new_ret._match_info = self._match_info + return new_ret + + @reify + def match_info(self) -> "UrlMappingMatchInfo": + """Result of route resolving.""" + match_info = self._match_info + assert match_info is not None + return match_info + + @property + def app(self) -> "Application": + """Application instance.""" + match_info = self._match_info + assert match_info is not None + return match_info.current_app + + @property + def config_dict(self) -> ChainMapProxy: + match_info = self._match_info + assert match_info is not None + lst = match_info.apps + app = self.app + idx = lst.index(app) + sublist = list(reversed(lst[: idx + 1])) + return ChainMapProxy(sublist) + + async def _prepare_hook(self, response: StreamResponse) -> None: + match_info = self._match_info + if match_info is None: + return + for app in match_info._apps: + await app.on_response_prepare.send(self, response) diff --git a/contrib/python/aiohttp/aiohttp/web_response.py b/contrib/python/aiohttp/aiohttp/web_response.py new file mode 100644 index 0000000000..7880ab2d02 --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/web_response.py @@ -0,0 +1,825 @@ +import asyncio +import collections.abc +import datetime +import enum +import json +import math +import time +import warnings +import zlib +from concurrent.futures import Executor +from http.cookies import Morsel, SimpleCookie +from typing import ( + TYPE_CHECKING, + Any, + Dict, + Iterator, + Mapping, + MutableMapping, + Optional, + Tuple, + Union, + cast, +) + +from multidict import CIMultiDict, istr + +from . import hdrs, payload +from .abc import AbstractStreamWriter +from .helpers import ( + ETAG_ANY, + PY_38, + QUOTED_ETAG_RE, + ETag, + HeadersMixin, + parse_http_date, + rfc822_formatted_time, + sentinel, + validate_etag_value, +) +from .http import RESPONSES, SERVER_SOFTWARE, HttpVersion10, HttpVersion11 +from .payload import Payload +from .typedefs import JSONEncoder, LooseHeaders + +__all__ = ("ContentCoding", "StreamResponse", "Response", "json_response") + + +if TYPE_CHECKING: # pragma: no cover + from .web_request import BaseRequest + + BaseClass = MutableMapping[str, Any] +else: + BaseClass = collections.abc.MutableMapping + + +if not PY_38: + # allow samesite to be used in python < 3.8 + # already permitted in python 3.8, see https://bugs.python.org/issue29613 + Morsel._reserved["samesite"] = "SameSite" # type: ignore[attr-defined] + + +class ContentCoding(enum.Enum): + # The content codings that we have support for. + # + # Additional registered codings are listed at: + # https://www.iana.org/assignments/http-parameters/http-parameters.xhtml#content-coding + deflate = "deflate" + gzip = "gzip" + identity = "identity" + + +############################################################ +# HTTP Response classes +############################################################ + + +class StreamResponse(BaseClass, HeadersMixin): + + _length_check = True + + def __init__( + self, + *, + status: int = 200, + reason: Optional[str] = None, + headers: Optional[LooseHeaders] = None, + ) -> None: + self._body = None + self._keep_alive = None # type: Optional[bool] + self._chunked = False + self._compression = False + self._compression_force = None # type: Optional[ContentCoding] + self._cookies = SimpleCookie() # type: SimpleCookie[str] + + self._req = None # type: Optional[BaseRequest] + self._payload_writer = None # type: Optional[AbstractStreamWriter] + self._eof_sent = False + self._body_length = 0 + self._state = {} # type: Dict[str, Any] + + if headers is not None: + self._headers = CIMultiDict(headers) # type: CIMultiDict[str] + else: + self._headers = CIMultiDict() + + self.set_status(status, reason) + + @property + def prepared(self) -> bool: + return self._payload_writer is not None + + @property + def task(self) -> "Optional[asyncio.Task[None]]": + if self._req: + return self._req.task + else: + return None + + @property + def status(self) -> int: + return self._status + + @property + def chunked(self) -> bool: + return self._chunked + + @property + def compression(self) -> bool: + return self._compression + + @property + def reason(self) -> str: + return self._reason + + def set_status( + self, + status: int, + reason: Optional[str] = None, + _RESPONSES: Mapping[int, Tuple[str, str]] = RESPONSES, + ) -> None: + assert not self.prepared, ( + "Cannot change the response status code after " "the headers have been sent" + ) + self._status = int(status) + if reason is None: + try: + reason = _RESPONSES[self._status][0] + except Exception: + reason = "" + self._reason = reason + + @property + def keep_alive(self) -> Optional[bool]: + return self._keep_alive + + def force_close(self) -> None: + self._keep_alive = False + + @property + def body_length(self) -> int: + return self._body_length + + @property + def output_length(self) -> int: + warnings.warn("output_length is deprecated", DeprecationWarning) + assert self._payload_writer + return self._payload_writer.buffer_size + + def enable_chunked_encoding(self, chunk_size: Optional[int] = None) -> None: + """Enables automatic chunked transfer encoding.""" + self._chunked = True + + if hdrs.CONTENT_LENGTH in self._headers: + raise RuntimeError( + "You can't enable chunked encoding when " "a content length is set" + ) + if chunk_size is not None: + warnings.warn("Chunk size is deprecated #1615", DeprecationWarning) + + def enable_compression( + self, force: Optional[Union[bool, ContentCoding]] = None + ) -> None: + """Enables response compression encoding.""" + # Backwards compatibility for when force was a bool <0.17. + if type(force) == bool: + force = ContentCoding.deflate if force else ContentCoding.identity + warnings.warn( + "Using boolean for force is deprecated #3318", DeprecationWarning + ) + elif force is not None: + assert isinstance(force, ContentCoding), ( + "force should one of " "None, bool or " "ContentEncoding" + ) + + self._compression = True + self._compression_force = force + + @property + def headers(self) -> "CIMultiDict[str]": + return self._headers + + @property + def cookies(self) -> "SimpleCookie[str]": + return self._cookies + + def set_cookie( + self, + name: str, + value: str, + *, + expires: Optional[str] = None, + domain: Optional[str] = None, + max_age: Optional[Union[int, str]] = None, + path: str = "/", + secure: Optional[bool] = None, + httponly: Optional[bool] = None, + version: Optional[str] = None, + samesite: Optional[str] = None, + ) -> None: + """Set or update response cookie. + + Sets new cookie or updates existent with new value. + Also updates only those params which are not None. + """ + old = self._cookies.get(name) + if old is not None and old.coded_value == "": + # deleted cookie + self._cookies.pop(name, None) + + self._cookies[name] = value + c = self._cookies[name] + + if expires is not None: + c["expires"] = expires + elif c.get("expires") == "Thu, 01 Jan 1970 00:00:00 GMT": + del c["expires"] + + if domain is not None: + c["domain"] = domain + + if max_age is not None: + c["max-age"] = str(max_age) + elif "max-age" in c: + del c["max-age"] + + c["path"] = path + + if secure is not None: + c["secure"] = secure + if httponly is not None: + c["httponly"] = httponly + if version is not None: + c["version"] = version + if samesite is not None: + c["samesite"] = samesite + + def del_cookie( + self, name: str, *, domain: Optional[str] = None, path: str = "/" + ) -> None: + """Delete cookie. + + Creates new empty expired cookie. + """ + # TODO: do we need domain/path here? + self._cookies.pop(name, None) + self.set_cookie( + name, + "", + max_age=0, + expires="Thu, 01 Jan 1970 00:00:00 GMT", + domain=domain, + path=path, + ) + + @property + def content_length(self) -> Optional[int]: + # Just a placeholder for adding setter + return super().content_length + + @content_length.setter + def content_length(self, value: Optional[int]) -> None: + if value is not None: + value = int(value) + if self._chunked: + raise RuntimeError( + "You can't set content length when " "chunked encoding is enable" + ) + self._headers[hdrs.CONTENT_LENGTH] = str(value) + else: + self._headers.pop(hdrs.CONTENT_LENGTH, None) + + @property + def content_type(self) -> str: + # Just a placeholder for adding setter + return super().content_type + + @content_type.setter + def content_type(self, value: str) -> None: + self.content_type # read header values if needed + self._content_type = str(value) + self._generate_content_type_header() + + @property + def charset(self) -> Optional[str]: + # Just a placeholder for adding setter + return super().charset + + @charset.setter + def charset(self, value: Optional[str]) -> None: + ctype = self.content_type # read header values if needed + if ctype == "application/octet-stream": + raise RuntimeError( + "Setting charset for application/octet-stream " + "doesn't make sense, setup content_type first" + ) + assert self._content_dict is not None + if value is None: + self._content_dict.pop("charset", None) + else: + self._content_dict["charset"] = str(value).lower() + self._generate_content_type_header() + + @property + def last_modified(self) -> Optional[datetime.datetime]: + """The value of Last-Modified HTTP header, or None. + + This header is represented as a `datetime` object. + """ + return parse_http_date(self._headers.get(hdrs.LAST_MODIFIED)) + + @last_modified.setter + def last_modified( + self, value: Optional[Union[int, float, datetime.datetime, str]] + ) -> None: + if value is None: + self._headers.pop(hdrs.LAST_MODIFIED, None) + elif isinstance(value, (int, float)): + self._headers[hdrs.LAST_MODIFIED] = time.strftime( + "%a, %d %b %Y %H:%M:%S GMT", time.gmtime(math.ceil(value)) + ) + elif isinstance(value, datetime.datetime): + self._headers[hdrs.LAST_MODIFIED] = time.strftime( + "%a, %d %b %Y %H:%M:%S GMT", value.utctimetuple() + ) + elif isinstance(value, str): + self._headers[hdrs.LAST_MODIFIED] = value + + @property + def etag(self) -> Optional[ETag]: + quoted_value = self._headers.get(hdrs.ETAG) + if not quoted_value: + return None + elif quoted_value == ETAG_ANY: + return ETag(value=ETAG_ANY) + match = QUOTED_ETAG_RE.fullmatch(quoted_value) + if not match: + return None + is_weak, value = match.group(1, 2) + return ETag( + is_weak=bool(is_weak), + value=value, + ) + + @etag.setter + def etag(self, value: Optional[Union[ETag, str]]) -> None: + if value is None: + self._headers.pop(hdrs.ETAG, None) + elif (isinstance(value, str) and value == ETAG_ANY) or ( + isinstance(value, ETag) and value.value == ETAG_ANY + ): + self._headers[hdrs.ETAG] = ETAG_ANY + elif isinstance(value, str): + validate_etag_value(value) + self._headers[hdrs.ETAG] = f'"{value}"' + elif isinstance(value, ETag) and isinstance(value.value, str): + validate_etag_value(value.value) + hdr_value = f'W/"{value.value}"' if value.is_weak else f'"{value.value}"' + self._headers[hdrs.ETAG] = hdr_value + else: + raise ValueError( + f"Unsupported etag type: {type(value)}. " + f"etag must be str, ETag or None" + ) + + def _generate_content_type_header( + self, CONTENT_TYPE: istr = hdrs.CONTENT_TYPE + ) -> None: + assert self._content_dict is not None + assert self._content_type is not None + params = "; ".join(f"{k}={v}" for k, v in self._content_dict.items()) + if params: + ctype = self._content_type + "; " + params + else: + ctype = self._content_type + self._headers[CONTENT_TYPE] = ctype + + async def _do_start_compression(self, coding: ContentCoding) -> None: + if coding != ContentCoding.identity: + assert self._payload_writer is not None + self._headers[hdrs.CONTENT_ENCODING] = coding.value + self._payload_writer.enable_compression(coding.value) + # Compressed payload may have different content length, + # remove the header + self._headers.popall(hdrs.CONTENT_LENGTH, None) + + async def _start_compression(self, request: "BaseRequest") -> None: + if self._compression_force: + await self._do_start_compression(self._compression_force) + else: + accept_encoding = request.headers.get(hdrs.ACCEPT_ENCODING, "").lower() + for coding in ContentCoding: + if coding.value in accept_encoding: + await self._do_start_compression(coding) + return + + async def prepare(self, request: "BaseRequest") -> Optional[AbstractStreamWriter]: + if self._eof_sent: + return None + if self._payload_writer is not None: + return self._payload_writer + + return await self._start(request) + + async def _start(self, request: "BaseRequest") -> AbstractStreamWriter: + self._req = request + writer = self._payload_writer = request._payload_writer + + await self._prepare_headers() + await request._prepare_hook(self) + await self._write_headers() + + return writer + + async def _prepare_headers(self) -> None: + request = self._req + assert request is not None + writer = self._payload_writer + assert writer is not None + keep_alive = self._keep_alive + if keep_alive is None: + keep_alive = request.keep_alive + self._keep_alive = keep_alive + + version = request.version + + headers = self._headers + for cookie in self._cookies.values(): + value = cookie.output(header="")[1:] + headers.add(hdrs.SET_COOKIE, value) + + if self._compression: + await self._start_compression(request) + + if self._chunked: + if version != HttpVersion11: + raise RuntimeError( + "Using chunked encoding is forbidden " + "for HTTP/{0.major}.{0.minor}".format(request.version) + ) + writer.enable_chunking() + headers[hdrs.TRANSFER_ENCODING] = "chunked" + if hdrs.CONTENT_LENGTH in headers: + del headers[hdrs.CONTENT_LENGTH] + elif self._length_check: + writer.length = self.content_length + if writer.length is None: + if version >= HttpVersion11 and self.status != 204: + writer.enable_chunking() + headers[hdrs.TRANSFER_ENCODING] = "chunked" + if hdrs.CONTENT_LENGTH in headers: + del headers[hdrs.CONTENT_LENGTH] + else: + keep_alive = False + # HTTP 1.1: https://tools.ietf.org/html/rfc7230#section-3.3.2 + # HTTP 1.0: https://tools.ietf.org/html/rfc1945#section-10.4 + elif version >= HttpVersion11 and self.status in (100, 101, 102, 103, 204): + del headers[hdrs.CONTENT_LENGTH] + + if self.status not in (204, 304): + headers.setdefault(hdrs.CONTENT_TYPE, "application/octet-stream") + headers.setdefault(hdrs.DATE, rfc822_formatted_time()) + headers.setdefault(hdrs.SERVER, SERVER_SOFTWARE) + + # connection header + if hdrs.CONNECTION not in headers: + if keep_alive: + if version == HttpVersion10: + headers[hdrs.CONNECTION] = "keep-alive" + else: + if version == HttpVersion11: + headers[hdrs.CONNECTION] = "close" + + async def _write_headers(self) -> None: + request = self._req + assert request is not None + writer = self._payload_writer + assert writer is not None + # status line + version = request.version + status_line = "HTTP/{}.{} {} {}".format( + version[0], version[1], self._status, self._reason + ) + await writer.write_headers(status_line, self._headers) + + async def write(self, data: bytes) -> None: + assert isinstance( + data, (bytes, bytearray, memoryview) + ), "data argument must be byte-ish (%r)" % type(data) + + if self._eof_sent: + raise RuntimeError("Cannot call write() after write_eof()") + if self._payload_writer is None: + raise RuntimeError("Cannot call write() before prepare()") + + await self._payload_writer.write(data) + + async def drain(self) -> None: + assert not self._eof_sent, "EOF has already been sent" + assert self._payload_writer is not None, "Response has not been started" + warnings.warn( + "drain method is deprecated, use await resp.write()", + DeprecationWarning, + stacklevel=2, + ) + await self._payload_writer.drain() + + async def write_eof(self, data: bytes = b"") -> None: + assert isinstance( + data, (bytes, bytearray, memoryview) + ), "data argument must be byte-ish (%r)" % type(data) + + if self._eof_sent: + return + + assert self._payload_writer is not None, "Response has not been started" + + await self._payload_writer.write_eof(data) + self._eof_sent = True + self._req = None + self._body_length = self._payload_writer.output_size + self._payload_writer = None + + def __repr__(self) -> str: + if self._eof_sent: + info = "eof" + elif self.prepared: + assert self._req is not None + info = f"{self._req.method} {self._req.path} " + else: + info = "not prepared" + return f"<{self.__class__.__name__} {self.reason} {info}>" + + def __getitem__(self, key: str) -> Any: + return self._state[key] + + def __setitem__(self, key: str, value: Any) -> None: + self._state[key] = value + + def __delitem__(self, key: str) -> None: + del self._state[key] + + def __len__(self) -> int: + return len(self._state) + + def __iter__(self) -> Iterator[str]: + return iter(self._state) + + def __hash__(self) -> int: + return hash(id(self)) + + def __eq__(self, other: object) -> bool: + return self is other + + +class Response(StreamResponse): + def __init__( + self, + *, + body: Any = None, + status: int = 200, + reason: Optional[str] = None, + text: Optional[str] = None, + headers: Optional[LooseHeaders] = None, + content_type: Optional[str] = None, + charset: Optional[str] = None, + zlib_executor_size: Optional[int] = None, + zlib_executor: Optional[Executor] = None, + ) -> None: + if body is not None and text is not None: + raise ValueError("body and text are not allowed together") + + if headers is None: + real_headers = CIMultiDict() # type: CIMultiDict[str] + elif not isinstance(headers, CIMultiDict): + real_headers = CIMultiDict(headers) + else: + real_headers = headers # = cast('CIMultiDict[str]', headers) + + if content_type is not None and "charset" in content_type: + raise ValueError("charset must not be in content_type " "argument") + + if text is not None: + if hdrs.CONTENT_TYPE in real_headers: + if content_type or charset: + raise ValueError( + "passing both Content-Type header and " + "content_type or charset params " + "is forbidden" + ) + else: + # fast path for filling headers + if not isinstance(text, str): + raise TypeError("text argument must be str (%r)" % type(text)) + if content_type is None: + content_type = "text/plain" + if charset is None: + charset = "utf-8" + real_headers[hdrs.CONTENT_TYPE] = content_type + "; charset=" + charset + body = text.encode(charset) + text = None + else: + if hdrs.CONTENT_TYPE in real_headers: + if content_type is not None or charset is not None: + raise ValueError( + "passing both Content-Type header and " + "content_type or charset params " + "is forbidden" + ) + else: + if content_type is not None: + if charset is not None: + content_type += "; charset=" + charset + real_headers[hdrs.CONTENT_TYPE] = content_type + + super().__init__(status=status, reason=reason, headers=real_headers) + + if text is not None: + self.text = text + else: + self.body = body + + self._compressed_body = None # type: Optional[bytes] + self._zlib_executor_size = zlib_executor_size + self._zlib_executor = zlib_executor + + @property + def body(self) -> Optional[Union[bytes, Payload]]: + return self._body + + @body.setter + def body( + self, + body: bytes, + CONTENT_TYPE: istr = hdrs.CONTENT_TYPE, + CONTENT_LENGTH: istr = hdrs.CONTENT_LENGTH, + ) -> None: + if body is None: + self._body = None # type: Optional[bytes] + self._body_payload = False # type: bool + elif isinstance(body, (bytes, bytearray)): + self._body = body + self._body_payload = False + else: + try: + self._body = body = payload.PAYLOAD_REGISTRY.get(body) + except payload.LookupError: + raise ValueError("Unsupported body type %r" % type(body)) + + self._body_payload = True + + headers = self._headers + + # set content-length header if needed + if not self._chunked and CONTENT_LENGTH not in headers: + size = body.size + if size is not None: + headers[CONTENT_LENGTH] = str(size) + + # set content-type + if CONTENT_TYPE not in headers: + headers[CONTENT_TYPE] = body.content_type + + # copy payload headers + if body.headers: + for (key, value) in body.headers.items(): + if key not in headers: + headers[key] = value + + self._compressed_body = None + + @property + def text(self) -> Optional[str]: + if self._body is None: + return None + return self._body.decode(self.charset or "utf-8") + + @text.setter + def text(self, text: str) -> None: + assert text is None or isinstance( + text, str + ), "text argument must be str (%r)" % type(text) + + if self.content_type == "application/octet-stream": + self.content_type = "text/plain" + if self.charset is None: + self.charset = "utf-8" + + self._body = text.encode(self.charset) + self._body_payload = False + self._compressed_body = None + + @property + def content_length(self) -> Optional[int]: + if self._chunked: + return None + + if hdrs.CONTENT_LENGTH in self._headers: + return super().content_length + + if self._compressed_body is not None: + # Return length of the compressed body + return len(self._compressed_body) + elif self._body_payload: + # A payload without content length, or a compressed payload + return None + elif self._body is not None: + return len(self._body) + else: + return 0 + + @content_length.setter + def content_length(self, value: Optional[int]) -> None: + raise RuntimeError("Content length is set automatically") + + async def write_eof(self, data: bytes = b"") -> None: + if self._eof_sent: + return + if self._compressed_body is None: + body = self._body # type: Optional[Union[bytes, Payload]] + else: + body = self._compressed_body + assert not data, f"data arg is not supported, got {data!r}" + assert self._req is not None + assert self._payload_writer is not None + if body is not None: + if self._req._method == hdrs.METH_HEAD or self._status in [204, 304]: + await super().write_eof() + elif self._body_payload: + payload = cast(Payload, body) + await payload.write(self._payload_writer) + await super().write_eof() + else: + await super().write_eof(cast(bytes, body)) + else: + await super().write_eof() + + async def _start(self, request: "BaseRequest") -> AbstractStreamWriter: + if not self._chunked and hdrs.CONTENT_LENGTH not in self._headers: + if not self._body_payload: + if self._body is not None: + self._headers[hdrs.CONTENT_LENGTH] = str(len(self._body)) + else: + self._headers[hdrs.CONTENT_LENGTH] = "0" + + return await super()._start(request) + + def _compress_body(self, zlib_mode: int) -> None: + assert zlib_mode > 0 + compressobj = zlib.compressobj(wbits=zlib_mode) + body_in = self._body + assert body_in is not None + self._compressed_body = compressobj.compress(body_in) + compressobj.flush() + + async def _do_start_compression(self, coding: ContentCoding) -> None: + if self._body_payload or self._chunked: + return await super()._do_start_compression(coding) + + if coding != ContentCoding.identity: + # Instead of using _payload_writer.enable_compression, + # compress the whole body + zlib_mode = ( + 16 + zlib.MAX_WBITS if coding == ContentCoding.gzip else zlib.MAX_WBITS + ) + body_in = self._body + assert body_in is not None + if ( + self._zlib_executor_size is not None + and len(body_in) > self._zlib_executor_size + ): + await asyncio.get_event_loop().run_in_executor( + self._zlib_executor, self._compress_body, zlib_mode + ) + else: + self._compress_body(zlib_mode) + + body_out = self._compressed_body + assert body_out is not None + + self._headers[hdrs.CONTENT_ENCODING] = coding.value + self._headers[hdrs.CONTENT_LENGTH] = str(len(body_out)) + + +def json_response( + data: Any = sentinel, + *, + text: Optional[str] = None, + body: Optional[bytes] = None, + status: int = 200, + reason: Optional[str] = None, + headers: Optional[LooseHeaders] = None, + content_type: str = "application/json", + dumps: JSONEncoder = json.dumps, +) -> Response: + if data is not sentinel: + if text or body: + raise ValueError("only one of data, text, or body should be specified") + else: + text = dumps(data) + return Response( + text=text, + body=body, + status=status, + reason=reason, + headers=headers, + content_type=content_type, + ) diff --git a/contrib/python/aiohttp/aiohttp/web_routedef.py b/contrib/python/aiohttp/aiohttp/web_routedef.py new file mode 100644 index 0000000000..671e5c7f46 --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/web_routedef.py @@ -0,0 +1,213 @@ +import abc +import os # noqa +from typing import ( + TYPE_CHECKING, + Any, + Callable, + Dict, + Iterator, + List, + Optional, + Sequence, + Type, + Union, + overload, +) + +import attr + +from . import hdrs +from .abc import AbstractView +from .typedefs import Handler, PathLike + +if TYPE_CHECKING: # pragma: no cover + from .web_request import Request + from .web_response import StreamResponse + from .web_urldispatcher import AbstractRoute, UrlDispatcher +else: + Request = StreamResponse = UrlDispatcher = AbstractRoute = None + + +__all__ = ( + "AbstractRouteDef", + "RouteDef", + "StaticDef", + "RouteTableDef", + "head", + "options", + "get", + "post", + "patch", + "put", + "delete", + "route", + "view", + "static", +) + + +class AbstractRouteDef(abc.ABC): + @abc.abstractmethod + def register(self, router: UrlDispatcher) -> List[AbstractRoute]: + pass # pragma: no cover + + +_HandlerType = Union[Type[AbstractView], Handler] + + +@attr.s(auto_attribs=True, frozen=True, repr=False, slots=True) +class RouteDef(AbstractRouteDef): + method: str + path: str + handler: _HandlerType + kwargs: Dict[str, Any] + + def __repr__(self) -> str: + info = [] + for name, value in sorted(self.kwargs.items()): + info.append(f", {name}={value!r}") + return "<RouteDef {method} {path} -> {handler.__name__!r}" "{info}>".format( + method=self.method, path=self.path, handler=self.handler, info="".join(info) + ) + + def register(self, router: UrlDispatcher) -> List[AbstractRoute]: + if self.method in hdrs.METH_ALL: + reg = getattr(router, "add_" + self.method.lower()) + return [reg(self.path, self.handler, **self.kwargs)] + else: + return [ + router.add_route(self.method, self.path, self.handler, **self.kwargs) + ] + + +@attr.s(auto_attribs=True, frozen=True, repr=False, slots=True) +class StaticDef(AbstractRouteDef): + prefix: str + path: PathLike + kwargs: Dict[str, Any] + + def __repr__(self) -> str: + info = [] + for name, value in sorted(self.kwargs.items()): + info.append(f", {name}={value!r}") + return "<StaticDef {prefix} -> {path}" "{info}>".format( + prefix=self.prefix, path=self.path, info="".join(info) + ) + + def register(self, router: UrlDispatcher) -> List[AbstractRoute]: + resource = router.add_static(self.prefix, self.path, **self.kwargs) + routes = resource.get_info().get("routes", {}) + return list(routes.values()) + + +def route(method: str, path: str, handler: _HandlerType, **kwargs: Any) -> RouteDef: + return RouteDef(method, path, handler, kwargs) + + +def head(path: str, handler: _HandlerType, **kwargs: Any) -> RouteDef: + return route(hdrs.METH_HEAD, path, handler, **kwargs) + + +def options(path: str, handler: _HandlerType, **kwargs: Any) -> RouteDef: + return route(hdrs.METH_OPTIONS, path, handler, **kwargs) + + +def get( + path: str, + handler: _HandlerType, + *, + name: Optional[str] = None, + allow_head: bool = True, + **kwargs: Any, +) -> RouteDef: + return route( + hdrs.METH_GET, path, handler, name=name, allow_head=allow_head, **kwargs + ) + + +def post(path: str, handler: _HandlerType, **kwargs: Any) -> RouteDef: + return route(hdrs.METH_POST, path, handler, **kwargs) + + +def put(path: str, handler: _HandlerType, **kwargs: Any) -> RouteDef: + return route(hdrs.METH_PUT, path, handler, **kwargs) + + +def patch(path: str, handler: _HandlerType, **kwargs: Any) -> RouteDef: + return route(hdrs.METH_PATCH, path, handler, **kwargs) + + +def delete(path: str, handler: _HandlerType, **kwargs: Any) -> RouteDef: + return route(hdrs.METH_DELETE, path, handler, **kwargs) + + +def view(path: str, handler: Type[AbstractView], **kwargs: Any) -> RouteDef: + return route(hdrs.METH_ANY, path, handler, **kwargs) + + +def static(prefix: str, path: PathLike, **kwargs: Any) -> StaticDef: + return StaticDef(prefix, path, kwargs) + + +_Deco = Callable[[_HandlerType], _HandlerType] + + +class RouteTableDef(Sequence[AbstractRouteDef]): + """Route definition table""" + + def __init__(self) -> None: + self._items = [] # type: List[AbstractRouteDef] + + def __repr__(self) -> str: + return f"<RouteTableDef count={len(self._items)}>" + + @overload + def __getitem__(self, index: int) -> AbstractRouteDef: + ... + + @overload + def __getitem__(self, index: slice) -> List[AbstractRouteDef]: + ... + + def __getitem__(self, index): # type: ignore[no-untyped-def] + return self._items[index] + + def __iter__(self) -> Iterator[AbstractRouteDef]: + return iter(self._items) + + def __len__(self) -> int: + return len(self._items) + + def __contains__(self, item: object) -> bool: + return item in self._items + + def route(self, method: str, path: str, **kwargs: Any) -> _Deco: + def inner(handler: _HandlerType) -> _HandlerType: + self._items.append(RouteDef(method, path, handler, kwargs)) + return handler + + return inner + + def head(self, path: str, **kwargs: Any) -> _Deco: + return self.route(hdrs.METH_HEAD, path, **kwargs) + + def get(self, path: str, **kwargs: Any) -> _Deco: + return self.route(hdrs.METH_GET, path, **kwargs) + + def post(self, path: str, **kwargs: Any) -> _Deco: + return self.route(hdrs.METH_POST, path, **kwargs) + + def put(self, path: str, **kwargs: Any) -> _Deco: + return self.route(hdrs.METH_PUT, path, **kwargs) + + def patch(self, path: str, **kwargs: Any) -> _Deco: + return self.route(hdrs.METH_PATCH, path, **kwargs) + + def delete(self, path: str, **kwargs: Any) -> _Deco: + return self.route(hdrs.METH_DELETE, path, **kwargs) + + def view(self, path: str, **kwargs: Any) -> _Deco: + return self.route(hdrs.METH_ANY, path, **kwargs) + + def static(self, prefix: str, path: PathLike, **kwargs: Any) -> None: + self._items.append(StaticDef(prefix, path, kwargs)) diff --git a/contrib/python/aiohttp/aiohttp/web_runner.py b/contrib/python/aiohttp/aiohttp/web_runner.py new file mode 100644 index 0000000000..f4a64bff66 --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/web_runner.py @@ -0,0 +1,381 @@ +import asyncio +import signal +import socket +from abc import ABC, abstractmethod +from typing import Any, List, Optional, Set + +from yarl import URL + +from .web_app import Application +from .web_server import Server + +try: + from ssl import SSLContext +except ImportError: + SSLContext = object # type: ignore[misc,assignment] + + +__all__ = ( + "BaseSite", + "TCPSite", + "UnixSite", + "NamedPipeSite", + "SockSite", + "BaseRunner", + "AppRunner", + "ServerRunner", + "GracefulExit", +) + + +class GracefulExit(SystemExit): + code = 1 + + +def _raise_graceful_exit() -> None: + raise GracefulExit() + + +class BaseSite(ABC): + __slots__ = ("_runner", "_shutdown_timeout", "_ssl_context", "_backlog", "_server") + + def __init__( + self, + runner: "BaseRunner", + *, + shutdown_timeout: float = 60.0, + ssl_context: Optional[SSLContext] = None, + backlog: int = 128, + ) -> None: + if runner.server is None: + raise RuntimeError("Call runner.setup() before making a site") + self._runner = runner + self._shutdown_timeout = shutdown_timeout + self._ssl_context = ssl_context + self._backlog = backlog + self._server = None # type: Optional[asyncio.AbstractServer] + + @property + @abstractmethod + def name(self) -> str: + pass # pragma: no cover + + @abstractmethod + async def start(self) -> None: + self._runner._reg_site(self) + + async def stop(self) -> None: + self._runner._check_site(self) + if self._server is None: + self._runner._unreg_site(self) + return # not started yet + self._server.close() + # named pipes do not have wait_closed property + if hasattr(self._server, "wait_closed"): + await self._server.wait_closed() + await self._runner.shutdown() + assert self._runner.server + await self._runner.server.shutdown(self._shutdown_timeout) + self._runner._unreg_site(self) + + +class TCPSite(BaseSite): + __slots__ = ("_host", "_port", "_reuse_address", "_reuse_port") + + def __init__( + self, + runner: "BaseRunner", + host: Optional[str] = None, + port: Optional[int] = None, + *, + shutdown_timeout: float = 60.0, + ssl_context: Optional[SSLContext] = None, + backlog: int = 128, + reuse_address: Optional[bool] = None, + reuse_port: Optional[bool] = None, + ) -> None: + super().__init__( + runner, + shutdown_timeout=shutdown_timeout, + ssl_context=ssl_context, + backlog=backlog, + ) + self._host = host + if port is None: + port = 8443 if self._ssl_context else 8080 + self._port = port + self._reuse_address = reuse_address + self._reuse_port = reuse_port + + @property + def name(self) -> str: + scheme = "https" if self._ssl_context else "http" + host = "0.0.0.0" if self._host is None else self._host + return str(URL.build(scheme=scheme, host=host, port=self._port)) + + async def start(self) -> None: + await super().start() + loop = asyncio.get_event_loop() + server = self._runner.server + assert server is not None + self._server = await loop.create_server( + server, + self._host, + self._port, + ssl=self._ssl_context, + backlog=self._backlog, + reuse_address=self._reuse_address, + reuse_port=self._reuse_port, + ) + + +class UnixSite(BaseSite): + __slots__ = ("_path",) + + def __init__( + self, + runner: "BaseRunner", + path: str, + *, + shutdown_timeout: float = 60.0, + ssl_context: Optional[SSLContext] = None, + backlog: int = 128, + ) -> None: + super().__init__( + runner, + shutdown_timeout=shutdown_timeout, + ssl_context=ssl_context, + backlog=backlog, + ) + self._path = path + + @property + def name(self) -> str: + scheme = "https" if self._ssl_context else "http" + return f"{scheme}://unix:{self._path}:" + + async def start(self) -> None: + await super().start() + loop = asyncio.get_event_loop() + server = self._runner.server + assert server is not None + self._server = await loop.create_unix_server( + server, self._path, ssl=self._ssl_context, backlog=self._backlog + ) + + +class NamedPipeSite(BaseSite): + __slots__ = ("_path",) + + def __init__( + self, runner: "BaseRunner", path: str, *, shutdown_timeout: float = 60.0 + ) -> None: + loop = asyncio.get_event_loop() + if not isinstance( + loop, asyncio.ProactorEventLoop # type: ignore[attr-defined] + ): + raise RuntimeError( + "Named Pipes only available in proactor" "loop under windows" + ) + super().__init__(runner, shutdown_timeout=shutdown_timeout) + self._path = path + + @property + def name(self) -> str: + return self._path + + async def start(self) -> None: + await super().start() + loop = asyncio.get_event_loop() + server = self._runner.server + assert server is not None + _server = await loop.start_serving_pipe( # type: ignore[attr-defined] + server, self._path + ) + self._server = _server[0] + + +class SockSite(BaseSite): + __slots__ = ("_sock", "_name") + + def __init__( + self, + runner: "BaseRunner", + sock: socket.socket, + *, + shutdown_timeout: float = 60.0, + ssl_context: Optional[SSLContext] = None, + backlog: int = 128, + ) -> None: + super().__init__( + runner, + shutdown_timeout=shutdown_timeout, + ssl_context=ssl_context, + backlog=backlog, + ) + self._sock = sock + scheme = "https" if self._ssl_context else "http" + if hasattr(socket, "AF_UNIX") and sock.family == socket.AF_UNIX: + name = f"{scheme}://unix:{sock.getsockname()}:" + else: + host, port = sock.getsockname()[:2] + name = str(URL.build(scheme=scheme, host=host, port=port)) + self._name = name + + @property + def name(self) -> str: + return self._name + + async def start(self) -> None: + await super().start() + loop = asyncio.get_event_loop() + server = self._runner.server + assert server is not None + self._server = await loop.create_server( + server, sock=self._sock, ssl=self._ssl_context, backlog=self._backlog + ) + + +class BaseRunner(ABC): + __slots__ = ("_handle_signals", "_kwargs", "_server", "_sites") + + def __init__(self, *, handle_signals: bool = False, **kwargs: Any) -> None: + self._handle_signals = handle_signals + self._kwargs = kwargs + self._server = None # type: Optional[Server] + self._sites = [] # type: List[BaseSite] + + @property + def server(self) -> Optional[Server]: + return self._server + + @property + def addresses(self) -> List[Any]: + ret = [] # type: List[Any] + for site in self._sites: + server = site._server + if server is not None: + sockets = server.sockets + if sockets is not None: + for sock in sockets: + ret.append(sock.getsockname()) + return ret + + @property + def sites(self) -> Set[BaseSite]: + return set(self._sites) + + async def setup(self) -> None: + loop = asyncio.get_event_loop() + + if self._handle_signals: + try: + loop.add_signal_handler(signal.SIGINT, _raise_graceful_exit) + loop.add_signal_handler(signal.SIGTERM, _raise_graceful_exit) + except NotImplementedError: # pragma: no cover + # add_signal_handler is not implemented on Windows + pass + + self._server = await self._make_server() + + @abstractmethod + async def shutdown(self) -> None: + pass # pragma: no cover + + async def cleanup(self) -> None: + loop = asyncio.get_event_loop() + + # The loop over sites is intentional, an exception on gather() + # leaves self._sites in unpredictable state. + # The loop guaranties that a site is either deleted on success or + # still present on failure + for site in list(self._sites): + await site.stop() + await self._cleanup_server() + self._server = None + if self._handle_signals: + try: + loop.remove_signal_handler(signal.SIGINT) + loop.remove_signal_handler(signal.SIGTERM) + except NotImplementedError: # pragma: no cover + # remove_signal_handler is not implemented on Windows + pass + + @abstractmethod + async def _make_server(self) -> Server: + pass # pragma: no cover + + @abstractmethod + async def _cleanup_server(self) -> None: + pass # pragma: no cover + + def _reg_site(self, site: BaseSite) -> None: + if site in self._sites: + raise RuntimeError(f"Site {site} is already registered in runner {self}") + self._sites.append(site) + + def _check_site(self, site: BaseSite) -> None: + if site not in self._sites: + raise RuntimeError(f"Site {site} is not registered in runner {self}") + + def _unreg_site(self, site: BaseSite) -> None: + if site not in self._sites: + raise RuntimeError(f"Site {site} is not registered in runner {self}") + self._sites.remove(site) + + +class ServerRunner(BaseRunner): + """Low-level web server runner""" + + __slots__ = ("_web_server",) + + def __init__( + self, web_server: Server, *, handle_signals: bool = False, **kwargs: Any + ) -> None: + super().__init__(handle_signals=handle_signals, **kwargs) + self._web_server = web_server + + async def shutdown(self) -> None: + pass + + async def _make_server(self) -> Server: + return self._web_server + + async def _cleanup_server(self) -> None: + pass + + +class AppRunner(BaseRunner): + """Web Application runner""" + + __slots__ = ("_app",) + + def __init__( + self, app: Application, *, handle_signals: bool = False, **kwargs: Any + ) -> None: + super().__init__(handle_signals=handle_signals, **kwargs) + if not isinstance(app, Application): + raise TypeError( + "The first argument should be web.Application " + "instance, got {!r}".format(app) + ) + self._app = app + + @property + def app(self) -> Application: + return self._app + + async def shutdown(self) -> None: + await self._app.shutdown() + + async def _make_server(self) -> Server: + loop = asyncio.get_event_loop() + self._app._set_loop(loop) + self._app.on_startup.freeze() + await self._app.startup() + self._app.freeze() + + return self._app._make_handler(loop=loop, **self._kwargs) + + async def _cleanup_server(self) -> None: + await self._app.cleanup() diff --git a/contrib/python/aiohttp/aiohttp/web_server.py b/contrib/python/aiohttp/aiohttp/web_server.py new file mode 100644 index 0000000000..5657ed9c80 --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/web_server.py @@ -0,0 +1,62 @@ +"""Low level HTTP server.""" +import asyncio +from typing import Any, Awaitable, Callable, Dict, List, Optional # noqa + +from .abc import AbstractStreamWriter +from .helpers import get_running_loop +from .http_parser import RawRequestMessage +from .streams import StreamReader +from .web_protocol import RequestHandler, _RequestFactory, _RequestHandler +from .web_request import BaseRequest + +__all__ = ("Server",) + + +class Server: + def __init__( + self, + handler: _RequestHandler, + *, + request_factory: Optional[_RequestFactory] = None, + loop: Optional[asyncio.AbstractEventLoop] = None, + **kwargs: Any + ) -> None: + self._loop = get_running_loop(loop) + self._connections = {} # type: Dict[RequestHandler, asyncio.Transport] + self._kwargs = kwargs + self.requests_count = 0 + self.request_handler = handler + self.request_factory = request_factory or self._make_request + + @property + def connections(self) -> List[RequestHandler]: + return list(self._connections.keys()) + + def connection_made( + self, handler: RequestHandler, transport: asyncio.Transport + ) -> None: + self._connections[handler] = transport + + def connection_lost( + self, handler: RequestHandler, exc: Optional[BaseException] = None + ) -> None: + if handler in self._connections: + del self._connections[handler] + + def _make_request( + self, + message: RawRequestMessage, + payload: StreamReader, + protocol: RequestHandler, + writer: AbstractStreamWriter, + task: "asyncio.Task[None]", + ) -> BaseRequest: + return BaseRequest(message, payload, protocol, writer, task, self._loop) + + async def shutdown(self, timeout: Optional[float] = None) -> None: + coros = [conn.shutdown(timeout) for conn in self._connections] + await asyncio.gather(*coros) + self._connections.clear() + + def __call__(self) -> RequestHandler: + return RequestHandler(self, loop=self._loop, **self._kwargs) diff --git a/contrib/python/aiohttp/aiohttp/web_urldispatcher.py b/contrib/python/aiohttp/aiohttp/web_urldispatcher.py new file mode 100644 index 0000000000..73ec4c05d0 --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/web_urldispatcher.py @@ -0,0 +1,1220 @@ +import abc +import asyncio +import base64 +import hashlib +import inspect +import keyword +import os +import re +import warnings +from contextlib import contextmanager +from functools import wraps +from pathlib import Path +from types import MappingProxyType +from typing import ( + TYPE_CHECKING, + Any, + Awaitable, + Callable, + Container, + Dict, + Generator, + Iterable, + Iterator, + List, + Mapping, + Optional, + Pattern, + Set, + Sized, + Tuple, + Type, + Union, + cast, +) + +from yarl import URL, __version__ as yarl_version # type: ignore[attr-defined] + +from . import hdrs +from .abc import AbstractMatchInfo, AbstractRouter, AbstractView +from .helpers import DEBUG +from .http import HttpVersion11 +from .typedefs import Final, Handler, PathLike, TypedDict +from .web_exceptions import ( + HTTPException, + HTTPExpectationFailed, + HTTPForbidden, + HTTPMethodNotAllowed, + HTTPNotFound, +) +from .web_fileresponse import FileResponse +from .web_request import Request +from .web_response import Response, StreamResponse +from .web_routedef import AbstractRouteDef + +__all__ = ( + "UrlDispatcher", + "UrlMappingMatchInfo", + "AbstractResource", + "Resource", + "PlainResource", + "DynamicResource", + "AbstractRoute", + "ResourceRoute", + "StaticResource", + "View", +) + + +if TYPE_CHECKING: # pragma: no cover + from .web_app import Application + + BaseDict = Dict[str, str] +else: + BaseDict = dict + +YARL_VERSION: Final[Tuple[int, ...]] = tuple(map(int, yarl_version.split(".")[:2])) + +HTTP_METHOD_RE: Final[Pattern[str]] = re.compile( + r"^[0-9A-Za-z!#\$%&'\*\+\-\.\^_`\|~]+$" +) +ROUTE_RE: Final[Pattern[str]] = re.compile( + r"(\{[_a-zA-Z][^{}]*(?:\{[^{}]*\}[^{}]*)*\})" +) +PATH_SEP: Final[str] = re.escape("/") + + +_ExpectHandler = Callable[[Request], Awaitable[None]] +_Resolve = Tuple[Optional["UrlMappingMatchInfo"], Set[str]] + + +class _InfoDict(TypedDict, total=False): + path: str + + formatter: str + pattern: Pattern[str] + + directory: Path + prefix: str + routes: Mapping[str, "AbstractRoute"] + + app: "Application" + + domain: str + + rule: "AbstractRuleMatching" + + http_exception: HTTPException + + +class AbstractResource(Sized, Iterable["AbstractRoute"]): + def __init__(self, *, name: Optional[str] = None) -> None: + self._name = name + + @property + def name(self) -> Optional[str]: + return self._name + + @property + @abc.abstractmethod + def canonical(self) -> str: + """Exposes the resource's canonical path. + + For example '/foo/bar/{name}' + + """ + + @abc.abstractmethod # pragma: no branch + def url_for(self, **kwargs: str) -> URL: + """Construct url for resource with additional params.""" + + @abc.abstractmethod # pragma: no branch + async def resolve(self, request: Request) -> _Resolve: + """Resolve resource. + + Return (UrlMappingMatchInfo, allowed_methods) pair. + """ + + @abc.abstractmethod + def add_prefix(self, prefix: str) -> None: + """Add a prefix to processed URLs. + + Required for subapplications support. + """ + + @abc.abstractmethod + def get_info(self) -> _InfoDict: + """Return a dict with additional info useful for introspection""" + + def freeze(self) -> None: + pass + + @abc.abstractmethod + def raw_match(self, path: str) -> bool: + """Perform a raw match against path""" + + +class AbstractRoute(abc.ABC): + def __init__( + self, + method: str, + handler: Union[Handler, Type[AbstractView]], + *, + expect_handler: Optional[_ExpectHandler] = None, + resource: Optional[AbstractResource] = None, + ) -> None: + + if expect_handler is None: + expect_handler = _default_expect_handler + + assert asyncio.iscoroutinefunction( + expect_handler + ), f"Coroutine is expected, got {expect_handler!r}" + + method = method.upper() + if not HTTP_METHOD_RE.match(method): + raise ValueError(f"{method} is not allowed HTTP method") + + assert callable(handler), handler + if asyncio.iscoroutinefunction(handler): + pass + elif inspect.isgeneratorfunction(handler): + warnings.warn( + "Bare generators are deprecated, " "use @coroutine wrapper", + DeprecationWarning, + ) + elif isinstance(handler, type) and issubclass(handler, AbstractView): + pass + else: + warnings.warn( + "Bare functions are deprecated, " "use async ones", DeprecationWarning + ) + + @wraps(handler) + async def handler_wrapper(request: Request) -> StreamResponse: + result = old_handler(request) + if asyncio.iscoroutine(result): + return await result + return result # type: ignore[return-value] + + old_handler = handler + handler = handler_wrapper + + self._method = method + self._handler = handler + self._expect_handler = expect_handler + self._resource = resource + + @property + def method(self) -> str: + return self._method + + @property + def handler(self) -> Handler: + return self._handler + + @property + @abc.abstractmethod + def name(self) -> Optional[str]: + """Optional route's name, always equals to resource's name.""" + + @property + def resource(self) -> Optional[AbstractResource]: + return self._resource + + @abc.abstractmethod + def get_info(self) -> _InfoDict: + """Return a dict with additional info useful for introspection""" + + @abc.abstractmethod # pragma: no branch + def url_for(self, *args: str, **kwargs: str) -> URL: + """Construct url for route with additional params.""" + + async def handle_expect_header(self, request: Request) -> None: + await self._expect_handler(request) + + +class UrlMappingMatchInfo(BaseDict, AbstractMatchInfo): + def __init__(self, match_dict: Dict[str, str], route: AbstractRoute): + super().__init__(match_dict) + self._route = route + self._apps = [] # type: List[Application] + self._current_app = None # type: Optional[Application] + self._frozen = False + + @property + def handler(self) -> Handler: + return self._route.handler + + @property + def route(self) -> AbstractRoute: + return self._route + + @property + def expect_handler(self) -> _ExpectHandler: + return self._route.handle_expect_header + + @property + def http_exception(self) -> Optional[HTTPException]: + return None + + def get_info(self) -> _InfoDict: # type: ignore[override] + return self._route.get_info() + + @property + def apps(self) -> Tuple["Application", ...]: + return tuple(self._apps) + + def add_app(self, app: "Application") -> None: + if self._frozen: + raise RuntimeError("Cannot change apps stack after .freeze() call") + if self._current_app is None: + self._current_app = app + self._apps.insert(0, app) + + @property + def current_app(self) -> "Application": + app = self._current_app + assert app is not None + return app + + @contextmanager + def set_current_app(self, app: "Application") -> Generator[None, None, None]: + if DEBUG: # pragma: no cover + if app not in self._apps: + raise RuntimeError( + "Expected one of the following apps {!r}, got {!r}".format( + self._apps, app + ) + ) + prev = self._current_app + self._current_app = app + try: + yield + finally: + self._current_app = prev + + def freeze(self) -> None: + self._frozen = True + + def __repr__(self) -> str: + return f"<MatchInfo {super().__repr__()}: {self._route}>" + + +class MatchInfoError(UrlMappingMatchInfo): + def __init__(self, http_exception: HTTPException) -> None: + self._exception = http_exception + super().__init__({}, SystemRoute(self._exception)) + + @property + def http_exception(self) -> HTTPException: + return self._exception + + def __repr__(self) -> str: + return "<MatchInfoError {}: {}>".format( + self._exception.status, self._exception.reason + ) + + +async def _default_expect_handler(request: Request) -> None: + """Default handler for Expect header. + + Just send "100 Continue" to client. + raise HTTPExpectationFailed if value of header is not "100-continue" + """ + expect = request.headers.get(hdrs.EXPECT, "") + if request.version == HttpVersion11: + if expect.lower() == "100-continue": + await request.writer.write(b"HTTP/1.1 100 Continue\r\n\r\n") + else: + raise HTTPExpectationFailed(text="Unknown Expect: %s" % expect) + + +class Resource(AbstractResource): + def __init__(self, *, name: Optional[str] = None) -> None: + super().__init__(name=name) + self._routes = [] # type: List[ResourceRoute] + + def add_route( + self, + method: str, + handler: Union[Type[AbstractView], Handler], + *, + expect_handler: Optional[_ExpectHandler] = None, + ) -> "ResourceRoute": + + for route_obj in self._routes: + if route_obj.method == method or route_obj.method == hdrs.METH_ANY: + raise RuntimeError( + "Added route will never be executed, " + "method {route.method} is already " + "registered".format(route=route_obj) + ) + + route_obj = ResourceRoute(method, handler, self, expect_handler=expect_handler) + self.register_route(route_obj) + return route_obj + + def register_route(self, route: "ResourceRoute") -> None: + assert isinstance( + route, ResourceRoute + ), f"Instance of Route class is required, got {route!r}" + self._routes.append(route) + + async def resolve(self, request: Request) -> _Resolve: + allowed_methods = set() # type: Set[str] + + match_dict = self._match(request.rel_url.raw_path) + if match_dict is None: + return None, allowed_methods + + for route_obj in self._routes: + route_method = route_obj.method + allowed_methods.add(route_method) + + if route_method == request.method or route_method == hdrs.METH_ANY: + return (UrlMappingMatchInfo(match_dict, route_obj), allowed_methods) + else: + return None, allowed_methods + + @abc.abstractmethod + def _match(self, path: str) -> Optional[Dict[str, str]]: + pass # pragma: no cover + + def __len__(self) -> int: + return len(self._routes) + + def __iter__(self) -> Iterator[AbstractRoute]: + return iter(self._routes) + + # TODO: implement all abstract methods + + +class PlainResource(Resource): + def __init__(self, path: str, *, name: Optional[str] = None) -> None: + super().__init__(name=name) + assert not path or path.startswith("/") + self._path = path + + @property + def canonical(self) -> str: + return self._path + + def freeze(self) -> None: + if not self._path: + self._path = "/" + + def add_prefix(self, prefix: str) -> None: + assert prefix.startswith("/") + assert not prefix.endswith("/") + assert len(prefix) > 1 + self._path = prefix + self._path + + def _match(self, path: str) -> Optional[Dict[str, str]]: + # string comparison is about 10 times faster than regexp matching + if self._path == path: + return {} + else: + return None + + def raw_match(self, path: str) -> bool: + return self._path == path + + def get_info(self) -> _InfoDict: + return {"path": self._path} + + def url_for(self) -> URL: # type: ignore[override] + return URL.build(path=self._path, encoded=True) + + def __repr__(self) -> str: + name = "'" + self.name + "' " if self.name is not None else "" + return f"<PlainResource {name} {self._path}>" + + +class DynamicResource(Resource): + + DYN = re.compile(r"\{(?P<var>[_a-zA-Z][_a-zA-Z0-9]*)\}") + DYN_WITH_RE = re.compile(r"\{(?P<var>[_a-zA-Z][_a-zA-Z0-9]*):(?P<re>.+)\}") + GOOD = r"[^{}/]+" + + def __init__(self, path: str, *, name: Optional[str] = None) -> None: + super().__init__(name=name) + pattern = "" + formatter = "" + for part in ROUTE_RE.split(path): + match = self.DYN.fullmatch(part) + if match: + pattern += "(?P<{}>{})".format(match.group("var"), self.GOOD) + formatter += "{" + match.group("var") + "}" + continue + + match = self.DYN_WITH_RE.fullmatch(part) + if match: + pattern += "(?P<{var}>{re})".format(**match.groupdict()) + formatter += "{" + match.group("var") + "}" + continue + + if "{" in part or "}" in part: + raise ValueError(f"Invalid path '{path}'['{part}']") + + part = _requote_path(part) + formatter += part + pattern += re.escape(part) + + try: + compiled = re.compile(pattern) + except re.error as exc: + raise ValueError(f"Bad pattern '{pattern}': {exc}") from None + assert compiled.pattern.startswith(PATH_SEP) + assert formatter.startswith("/") + self._pattern = compiled + self._formatter = formatter + + @property + def canonical(self) -> str: + return self._formatter + + def add_prefix(self, prefix: str) -> None: + assert prefix.startswith("/") + assert not prefix.endswith("/") + assert len(prefix) > 1 + self._pattern = re.compile(re.escape(prefix) + self._pattern.pattern) + self._formatter = prefix + self._formatter + + def _match(self, path: str) -> Optional[Dict[str, str]]: + match = self._pattern.fullmatch(path) + if match is None: + return None + else: + return { + key: _unquote_path(value) for key, value in match.groupdict().items() + } + + def raw_match(self, path: str) -> bool: + return self._formatter == path + + def get_info(self) -> _InfoDict: + return {"formatter": self._formatter, "pattern": self._pattern} + + def url_for(self, **parts: str) -> URL: + url = self._formatter.format_map({k: _quote_path(v) for k, v in parts.items()}) + return URL.build(path=url, encoded=True) + + def __repr__(self) -> str: + name = "'" + self.name + "' " if self.name is not None else "" + return "<DynamicResource {name} {formatter}>".format( + name=name, formatter=self._formatter + ) + + +class PrefixResource(AbstractResource): + def __init__(self, prefix: str, *, name: Optional[str] = None) -> None: + assert not prefix or prefix.startswith("/"), prefix + assert prefix in ("", "/") or not prefix.endswith("/"), prefix + super().__init__(name=name) + self._prefix = _requote_path(prefix) + self._prefix2 = self._prefix + "/" + + @property + def canonical(self) -> str: + return self._prefix + + def add_prefix(self, prefix: str) -> None: + assert prefix.startswith("/") + assert not prefix.endswith("/") + assert len(prefix) > 1 + self._prefix = prefix + self._prefix + self._prefix2 = self._prefix + "/" + + def raw_match(self, prefix: str) -> bool: + return False + + # TODO: impl missing abstract methods + + +class StaticResource(PrefixResource): + VERSION_KEY = "v" + + def __init__( + self, + prefix: str, + directory: PathLike, + *, + name: Optional[str] = None, + expect_handler: Optional[_ExpectHandler] = None, + chunk_size: int = 256 * 1024, + show_index: bool = False, + follow_symlinks: bool = False, + append_version: bool = False, + ) -> None: + super().__init__(prefix, name=name) + try: + directory = Path(directory) + if str(directory).startswith("~"): + directory = Path(os.path.expanduser(str(directory))) + directory = directory.resolve() + if not directory.is_dir(): + raise ValueError("Not a directory") + except (FileNotFoundError, ValueError) as error: + raise ValueError(f"No directory exists at '{directory}'") from error + self._directory = directory + self._show_index = show_index + self._chunk_size = chunk_size + self._follow_symlinks = follow_symlinks + self._expect_handler = expect_handler + self._append_version = append_version + + self._routes = { + "GET": ResourceRoute( + "GET", self._handle, self, expect_handler=expect_handler + ), + "HEAD": ResourceRoute( + "HEAD", self._handle, self, expect_handler=expect_handler + ), + } + + def url_for( # type: ignore[override] + self, + *, + filename: Union[str, Path], + append_version: Optional[bool] = None, + ) -> URL: + if append_version is None: + append_version = self._append_version + if isinstance(filename, Path): + filename = str(filename) + filename = filename.lstrip("/") + + url = URL.build(path=self._prefix, encoded=True) + # filename is not encoded + if YARL_VERSION < (1, 6): + url = url / filename.replace("%", "%25") + else: + url = url / filename + + if append_version: + try: + filepath = self._directory.joinpath(filename).resolve() + if not self._follow_symlinks: + filepath.relative_to(self._directory) + except (ValueError, FileNotFoundError): + # ValueError for case when path point to symlink + # with follow_symlinks is False + return url # relatively safe + if filepath.is_file(): + # TODO cache file content + # with file watcher for cache invalidation + with filepath.open("rb") as f: + file_bytes = f.read() + h = self._get_file_hash(file_bytes) + url = url.with_query({self.VERSION_KEY: h}) + return url + return url + + @staticmethod + def _get_file_hash(byte_array: bytes) -> str: + m = hashlib.sha256() # todo sha256 can be configurable param + m.update(byte_array) + b64 = base64.urlsafe_b64encode(m.digest()) + return b64.decode("ascii") + + def get_info(self) -> _InfoDict: + return { + "directory": self._directory, + "prefix": self._prefix, + "routes": self._routes, + } + + def set_options_route(self, handler: Handler) -> None: + if "OPTIONS" in self._routes: + raise RuntimeError("OPTIONS route was set already") + self._routes["OPTIONS"] = ResourceRoute( + "OPTIONS", handler, self, expect_handler=self._expect_handler + ) + + async def resolve(self, request: Request) -> _Resolve: + path = request.rel_url.raw_path + method = request.method + allowed_methods = set(self._routes) + if not path.startswith(self._prefix2) and path != self._prefix: + return None, set() + + if method not in allowed_methods: + return None, allowed_methods + + match_dict = {"filename": _unquote_path(path[len(self._prefix) + 1 :])} + return (UrlMappingMatchInfo(match_dict, self._routes[method]), allowed_methods) + + def __len__(self) -> int: + return len(self._routes) + + def __iter__(self) -> Iterator[AbstractRoute]: + return iter(self._routes.values()) + + async def _handle(self, request: Request) -> StreamResponse: + rel_url = request.match_info["filename"] + try: + filename = Path(rel_url) + if filename.anchor: + # rel_url is an absolute name like + # /static/\\machine_name\c$ or /static/D:\path + # where the static dir is totally different + raise HTTPForbidden() + filepath = self._directory.joinpath(filename).resolve() + if not self._follow_symlinks: + filepath.relative_to(self._directory) + except (ValueError, FileNotFoundError) as error: + # relatively safe + raise HTTPNotFound() from error + except HTTPForbidden: + raise + except Exception as error: + # perm error or other kind! + request.app.logger.exception(error) + raise HTTPNotFound() from error + + # on opening a dir, load its contents if allowed + if filepath.is_dir(): + if self._show_index: + try: + return Response( + text=self._directory_as_html(filepath), content_type="text/html" + ) + except PermissionError: + raise HTTPForbidden() + else: + raise HTTPForbidden() + elif filepath.is_file(): + return FileResponse(filepath, chunk_size=self._chunk_size) + else: + raise HTTPNotFound + + def _directory_as_html(self, filepath: Path) -> str: + # returns directory's index as html + + # sanity check + assert filepath.is_dir() + + relative_path_to_dir = filepath.relative_to(self._directory).as_posix() + index_of = f"Index of /{relative_path_to_dir}" + h1 = f"<h1>{index_of}</h1>" + + index_list = [] + dir_index = filepath.iterdir() + for _file in sorted(dir_index): + # show file url as relative to static path + rel_path = _file.relative_to(self._directory).as_posix() + file_url = self._prefix + "/" + rel_path + + # if file is a directory, add '/' to the end of the name + if _file.is_dir(): + file_name = f"{_file.name}/" + else: + file_name = _file.name + + index_list.append( + '<li><a href="{url}">{name}</a></li>'.format( + url=file_url, name=file_name + ) + ) + ul = "<ul>\n{}\n</ul>".format("\n".join(index_list)) + body = f"<body>\n{h1}\n{ul}\n</body>" + + head_str = f"<head>\n<title>{index_of}</title>\n</head>" + html = f"<html>\n{head_str}\n{body}\n</html>" + + return html + + def __repr__(self) -> str: + name = "'" + self.name + "'" if self.name is not None else "" + return "<StaticResource {name} {path} -> {directory!r}>".format( + name=name, path=self._prefix, directory=self._directory + ) + + +class PrefixedSubAppResource(PrefixResource): + def __init__(self, prefix: str, app: "Application") -> None: + super().__init__(prefix) + self._app = app + for resource in app.router.resources(): + resource.add_prefix(prefix) + + def add_prefix(self, prefix: str) -> None: + super().add_prefix(prefix) + for resource in self._app.router.resources(): + resource.add_prefix(prefix) + + def url_for(self, *args: str, **kwargs: str) -> URL: + raise RuntimeError(".url_for() is not supported " "by sub-application root") + + def get_info(self) -> _InfoDict: + return {"app": self._app, "prefix": self._prefix} + + async def resolve(self, request: Request) -> _Resolve: + if ( + not request.url.raw_path.startswith(self._prefix2) + and request.url.raw_path != self._prefix + ): + return None, set() + match_info = await self._app.router.resolve(request) + match_info.add_app(self._app) + if isinstance(match_info.http_exception, HTTPMethodNotAllowed): + methods = match_info.http_exception.allowed_methods + else: + methods = set() + return match_info, methods + + def __len__(self) -> int: + return len(self._app.router.routes()) + + def __iter__(self) -> Iterator[AbstractRoute]: + return iter(self._app.router.routes()) + + def __repr__(self) -> str: + return "<PrefixedSubAppResource {prefix} -> {app!r}>".format( + prefix=self._prefix, app=self._app + ) + + +class AbstractRuleMatching(abc.ABC): + @abc.abstractmethod # pragma: no branch + async def match(self, request: Request) -> bool: + """Return bool if the request satisfies the criteria""" + + @abc.abstractmethod # pragma: no branch + def get_info(self) -> _InfoDict: + """Return a dict with additional info useful for introspection""" + + @property + @abc.abstractmethod # pragma: no branch + def canonical(self) -> str: + """Return a str""" + + +class Domain(AbstractRuleMatching): + re_part = re.compile(r"(?!-)[a-z\d-]{1,63}(?<!-)") + + def __init__(self, domain: str) -> None: + super().__init__() + self._domain = self.validation(domain) + + @property + def canonical(self) -> str: + return self._domain + + def validation(self, domain: str) -> str: + if not isinstance(domain, str): + raise TypeError("Domain must be str") + domain = domain.rstrip(".").lower() + if not domain: + raise ValueError("Domain cannot be empty") + elif "://" in domain: + raise ValueError("Scheme not supported") + url = URL("http://" + domain) + assert url.raw_host is not None + if not all(self.re_part.fullmatch(x) for x in url.raw_host.split(".")): + raise ValueError("Domain not valid") + if url.port == 80: + return url.raw_host + return f"{url.raw_host}:{url.port}" + + async def match(self, request: Request) -> bool: + host = request.headers.get(hdrs.HOST) + if not host: + return False + return self.match_domain(host) + + def match_domain(self, host: str) -> bool: + return host.lower() == self._domain + + def get_info(self) -> _InfoDict: + return {"domain": self._domain} + + +class MaskDomain(Domain): + re_part = re.compile(r"(?!-)[a-z\d\*-]{1,63}(?<!-)") + + def __init__(self, domain: str) -> None: + super().__init__(domain) + mask = self._domain.replace(".", r"\.").replace("*", ".*") + self._mask = re.compile(mask) + + @property + def canonical(self) -> str: + return self._mask.pattern + + def match_domain(self, host: str) -> bool: + return self._mask.fullmatch(host) is not None + + +class MatchedSubAppResource(PrefixedSubAppResource): + def __init__(self, rule: AbstractRuleMatching, app: "Application") -> None: + AbstractResource.__init__(self) + self._prefix = "" + self._app = app + self._rule = rule + + @property + def canonical(self) -> str: + return self._rule.canonical + + def get_info(self) -> _InfoDict: + return {"app": self._app, "rule": self._rule} + + async def resolve(self, request: Request) -> _Resolve: + if not await self._rule.match(request): + return None, set() + match_info = await self._app.router.resolve(request) + match_info.add_app(self._app) + if isinstance(match_info.http_exception, HTTPMethodNotAllowed): + methods = match_info.http_exception.allowed_methods + else: + methods = set() + return match_info, methods + + def __repr__(self) -> str: + return "<MatchedSubAppResource -> {app!r}>" "".format(app=self._app) + + +class ResourceRoute(AbstractRoute): + """A route with resource""" + + def __init__( + self, + method: str, + handler: Union[Handler, Type[AbstractView]], + resource: AbstractResource, + *, + expect_handler: Optional[_ExpectHandler] = None, + ) -> None: + super().__init__( + method, handler, expect_handler=expect_handler, resource=resource + ) + + def __repr__(self) -> str: + return "<ResourceRoute [{method}] {resource} -> {handler!r}".format( + method=self.method, resource=self._resource, handler=self.handler + ) + + @property + def name(self) -> Optional[str]: + if self._resource is None: + return None + return self._resource.name + + def url_for(self, *args: str, **kwargs: str) -> URL: + """Construct url for route with additional params.""" + assert self._resource is not None + return self._resource.url_for(*args, **kwargs) + + def get_info(self) -> _InfoDict: + assert self._resource is not None + return self._resource.get_info() + + +class SystemRoute(AbstractRoute): + def __init__(self, http_exception: HTTPException) -> None: + super().__init__(hdrs.METH_ANY, self._handle) + self._http_exception = http_exception + + def url_for(self, *args: str, **kwargs: str) -> URL: + raise RuntimeError(".url_for() is not allowed for SystemRoute") + + @property + def name(self) -> Optional[str]: + return None + + def get_info(self) -> _InfoDict: + return {"http_exception": self._http_exception} + + async def _handle(self, request: Request) -> StreamResponse: + raise self._http_exception + + @property + def status(self) -> int: + return self._http_exception.status + + @property + def reason(self) -> str: + return self._http_exception.reason + + def __repr__(self) -> str: + return "<SystemRoute {self.status}: {self.reason}>".format(self=self) + + +class View(AbstractView): + async def _iter(self) -> StreamResponse: + if self.request.method not in hdrs.METH_ALL: + self._raise_allowed_methods() + method: Callable[[], Awaitable[StreamResponse]] = getattr( + self, self.request.method.lower(), None + ) + if method is None: + self._raise_allowed_methods() + resp = await method() + return resp + + def __await__(self) -> Generator[Any, None, StreamResponse]: + return self._iter().__await__() + + def _raise_allowed_methods(self) -> None: + allowed_methods = {m for m in hdrs.METH_ALL if hasattr(self, m.lower())} + raise HTTPMethodNotAllowed(self.request.method, allowed_methods) + + +class ResourcesView(Sized, Iterable[AbstractResource], Container[AbstractResource]): + def __init__(self, resources: List[AbstractResource]) -> None: + self._resources = resources + + def __len__(self) -> int: + return len(self._resources) + + def __iter__(self) -> Iterator[AbstractResource]: + yield from self._resources + + def __contains__(self, resource: object) -> bool: + return resource in self._resources + + +class RoutesView(Sized, Iterable[AbstractRoute], Container[AbstractRoute]): + def __init__(self, resources: List[AbstractResource]): + self._routes = [] # type: List[AbstractRoute] + for resource in resources: + for route in resource: + self._routes.append(route) + + def __len__(self) -> int: + return len(self._routes) + + def __iter__(self) -> Iterator[AbstractRoute]: + yield from self._routes + + def __contains__(self, route: object) -> bool: + return route in self._routes + + +class UrlDispatcher(AbstractRouter, Mapping[str, AbstractResource]): + + NAME_SPLIT_RE = re.compile(r"[.:-]") + + def __init__(self) -> None: + super().__init__() + self._resources = [] # type: List[AbstractResource] + self._named_resources = {} # type: Dict[str, AbstractResource] + + async def resolve(self, request: Request) -> UrlMappingMatchInfo: + method = request.method + allowed_methods = set() # type: Set[str] + + for resource in self._resources: + match_dict, allowed = await resource.resolve(request) + if match_dict is not None: + return match_dict + else: + allowed_methods |= allowed + + if allowed_methods: + return MatchInfoError(HTTPMethodNotAllowed(method, allowed_methods)) + else: + return MatchInfoError(HTTPNotFound()) + + def __iter__(self) -> Iterator[str]: + return iter(self._named_resources) + + def __len__(self) -> int: + return len(self._named_resources) + + def __contains__(self, resource: object) -> bool: + return resource in self._named_resources + + def __getitem__(self, name: str) -> AbstractResource: + return self._named_resources[name] + + def resources(self) -> ResourcesView: + return ResourcesView(self._resources) + + def routes(self) -> RoutesView: + return RoutesView(self._resources) + + def named_resources(self) -> Mapping[str, AbstractResource]: + return MappingProxyType(self._named_resources) + + def register_resource(self, resource: AbstractResource) -> None: + assert isinstance( + resource, AbstractResource + ), f"Instance of AbstractResource class is required, got {resource!r}" + if self.frozen: + raise RuntimeError("Cannot register a resource into frozen router.") + + name = resource.name + + if name is not None: + parts = self.NAME_SPLIT_RE.split(name) + for part in parts: + if keyword.iskeyword(part): + raise ValueError( + f"Incorrect route name {name!r}, " + "python keywords cannot be used " + "for route name" + ) + if not part.isidentifier(): + raise ValueError( + "Incorrect route name {!r}, " + "the name should be a sequence of " + "python identifiers separated " + "by dash, dot or column".format(name) + ) + if name in self._named_resources: + raise ValueError( + "Duplicate {!r}, " + "already handled by {!r}".format(name, self._named_resources[name]) + ) + self._named_resources[name] = resource + self._resources.append(resource) + + def add_resource(self, path: str, *, name: Optional[str] = None) -> Resource: + if path and not path.startswith("/"): + raise ValueError("path should be started with / or be empty") + # Reuse last added resource if path and name are the same + if self._resources: + resource = self._resources[-1] + if resource.name == name and resource.raw_match(path): + return cast(Resource, resource) + if not ("{" in path or "}" in path or ROUTE_RE.search(path)): + resource = PlainResource(_requote_path(path), name=name) + self.register_resource(resource) + return resource + resource = DynamicResource(path, name=name) + self.register_resource(resource) + return resource + + def add_route( + self, + method: str, + path: str, + handler: Union[Handler, Type[AbstractView]], + *, + name: Optional[str] = None, + expect_handler: Optional[_ExpectHandler] = None, + ) -> AbstractRoute: + resource = self.add_resource(path, name=name) + return resource.add_route(method, handler, expect_handler=expect_handler) + + def add_static( + self, + prefix: str, + path: PathLike, + *, + name: Optional[str] = None, + expect_handler: Optional[_ExpectHandler] = None, + chunk_size: int = 256 * 1024, + show_index: bool = False, + follow_symlinks: bool = False, + append_version: bool = False, + ) -> AbstractResource: + """Add static files view. + + prefix - url prefix + path - folder with files + + """ + assert prefix.startswith("/") + if prefix.endswith("/"): + prefix = prefix[:-1] + resource = StaticResource( + prefix, + path, + name=name, + expect_handler=expect_handler, + chunk_size=chunk_size, + show_index=show_index, + follow_symlinks=follow_symlinks, + append_version=append_version, + ) + self.register_resource(resource) + return resource + + def add_head(self, path: str, handler: Handler, **kwargs: Any) -> AbstractRoute: + """Shortcut for add_route with method HEAD.""" + return self.add_route(hdrs.METH_HEAD, path, handler, **kwargs) + + def add_options(self, path: str, handler: Handler, **kwargs: Any) -> AbstractRoute: + """Shortcut for add_route with method OPTIONS.""" + return self.add_route(hdrs.METH_OPTIONS, path, handler, **kwargs) + + def add_get( + self, + path: str, + handler: Handler, + *, + name: Optional[str] = None, + allow_head: bool = True, + **kwargs: Any, + ) -> AbstractRoute: + """Shortcut for add_route with method GET. + + If allow_head is true, another + route is added allowing head requests to the same endpoint. + """ + resource = self.add_resource(path, name=name) + if allow_head: + resource.add_route(hdrs.METH_HEAD, handler, **kwargs) + return resource.add_route(hdrs.METH_GET, handler, **kwargs) + + def add_post(self, path: str, handler: Handler, **kwargs: Any) -> AbstractRoute: + """Shortcut for add_route with method POST.""" + return self.add_route(hdrs.METH_POST, path, handler, **kwargs) + + def add_put(self, path: str, handler: Handler, **kwargs: Any) -> AbstractRoute: + """Shortcut for add_route with method PUT.""" + return self.add_route(hdrs.METH_PUT, path, handler, **kwargs) + + def add_patch(self, path: str, handler: Handler, **kwargs: Any) -> AbstractRoute: + """Shortcut for add_route with method PATCH.""" + return self.add_route(hdrs.METH_PATCH, path, handler, **kwargs) + + def add_delete(self, path: str, handler: Handler, **kwargs: Any) -> AbstractRoute: + """Shortcut for add_route with method DELETE.""" + return self.add_route(hdrs.METH_DELETE, path, handler, **kwargs) + + def add_view( + self, path: str, handler: Type[AbstractView], **kwargs: Any + ) -> AbstractRoute: + """Shortcut for add_route with ANY methods for a class-based view.""" + return self.add_route(hdrs.METH_ANY, path, handler, **kwargs) + + def freeze(self) -> None: + super().freeze() + for resource in self._resources: + resource.freeze() + + def add_routes(self, routes: Iterable[AbstractRouteDef]) -> List[AbstractRoute]: + """Append routes to route table. + + Parameter should be a sequence of RouteDef objects. + + Returns a list of registered AbstractRoute instances. + """ + registered_routes = [] + for route_def in routes: + registered_routes.extend(route_def.register(self)) + return registered_routes + + +def _quote_path(value: str) -> str: + if YARL_VERSION < (1, 6): + value = value.replace("%", "%25") + return URL.build(path=value, encoded=False).raw_path + + +def _unquote_path(value: str) -> str: + return URL.build(path=value, encoded=True).path + + +def _requote_path(value: str) -> str: + # Quote non-ascii characters and other characters which must be quoted, + # but preserve existing %-sequences. + result = _quote_path(value) + if "%" in value: + result = result.replace("%25", "%") + return result diff --git a/contrib/python/aiohttp/aiohttp/web_ws.py b/contrib/python/aiohttp/aiohttp/web_ws.py new file mode 100644 index 0000000000..16b0a1747c --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/web_ws.py @@ -0,0 +1,487 @@ +import asyncio +import base64 +import binascii +import hashlib +import json +from typing import Any, Iterable, Optional, Tuple, cast + +import async_timeout +import attr +from multidict import CIMultiDict + +from . import hdrs +from .abc import AbstractStreamWriter +from .helpers import call_later, set_result +from .http import ( + WS_CLOSED_MESSAGE, + WS_CLOSING_MESSAGE, + WS_KEY, + WebSocketError, + WebSocketReader, + WebSocketWriter, + WSCloseCode, + WSMessage, + WSMsgType as WSMsgType, + ws_ext_gen, + ws_ext_parse, +) +from .log import ws_logger +from .streams import EofStream, FlowControlDataQueue +from .typedefs import Final, JSONDecoder, JSONEncoder +from .web_exceptions import HTTPBadRequest, HTTPException +from .web_request import BaseRequest +from .web_response import StreamResponse + +__all__ = ( + "WebSocketResponse", + "WebSocketReady", + "WSMsgType", +) + +THRESHOLD_CONNLOST_ACCESS: Final[int] = 5 + + +@attr.s(auto_attribs=True, frozen=True, slots=True) +class WebSocketReady: + ok: bool + protocol: Optional[str] + + def __bool__(self) -> bool: + return self.ok + + +class WebSocketResponse(StreamResponse): + + _length_check = False + + def __init__( + self, + *, + timeout: float = 10.0, + receive_timeout: Optional[float] = None, + autoclose: bool = True, + autoping: bool = True, + heartbeat: Optional[float] = None, + protocols: Iterable[str] = (), + compress: bool = True, + max_msg_size: int = 4 * 1024 * 1024, + ) -> None: + super().__init__(status=101) + self._protocols = protocols + self._ws_protocol = None # type: Optional[str] + self._writer = None # type: Optional[WebSocketWriter] + self._reader = None # type: Optional[FlowControlDataQueue[WSMessage]] + self._closed = False + self._closing = False + self._conn_lost = 0 + self._close_code = None # type: Optional[int] + self._loop = None # type: Optional[asyncio.AbstractEventLoop] + self._waiting = None # type: Optional[asyncio.Future[bool]] + self._exception = None # type: Optional[BaseException] + self._timeout = timeout + self._receive_timeout = receive_timeout + self._autoclose = autoclose + self._autoping = autoping + self._heartbeat = heartbeat + self._heartbeat_cb: Optional[asyncio.TimerHandle] = None + if heartbeat is not None: + self._pong_heartbeat = heartbeat / 2.0 + self._pong_response_cb: Optional[asyncio.TimerHandle] = None + self._compress = compress + self._max_msg_size = max_msg_size + + def _cancel_heartbeat(self) -> None: + if self._pong_response_cb is not None: + self._pong_response_cb.cancel() + self._pong_response_cb = None + + if self._heartbeat_cb is not None: + self._heartbeat_cb.cancel() + self._heartbeat_cb = None + + def _reset_heartbeat(self) -> None: + self._cancel_heartbeat() + + if self._heartbeat is not None: + assert self._loop is not None + self._heartbeat_cb = call_later( + self._send_heartbeat, self._heartbeat, self._loop + ) + + def _send_heartbeat(self) -> None: + if self._heartbeat is not None and not self._closed: + assert self._loop is not None + # fire-and-forget a task is not perfect but maybe ok for + # sending ping. Otherwise we need a long-living heartbeat + # task in the class. + self._loop.create_task(self._writer.ping()) # type: ignore[union-attr] + + if self._pong_response_cb is not None: + self._pong_response_cb.cancel() + self._pong_response_cb = call_later( + self._pong_not_received, self._pong_heartbeat, self._loop + ) + + def _pong_not_received(self) -> None: + if self._req is not None and self._req.transport is not None: + self._closed = True + self._close_code = WSCloseCode.ABNORMAL_CLOSURE + self._exception = asyncio.TimeoutError() + self._req.transport.close() + + async def prepare(self, request: BaseRequest) -> AbstractStreamWriter: + # make pre-check to don't hide it by do_handshake() exceptions + if self._payload_writer is not None: + return self._payload_writer + + protocol, writer = self._pre_start(request) + payload_writer = await super().prepare(request) + assert payload_writer is not None + self._post_start(request, protocol, writer) + await payload_writer.drain() + return payload_writer + + def _handshake( + self, request: BaseRequest + ) -> Tuple["CIMultiDict[str]", str, bool, bool]: + headers = request.headers + if "websocket" != headers.get(hdrs.UPGRADE, "").lower().strip(): + raise HTTPBadRequest( + text=( + "No WebSocket UPGRADE hdr: {}\n Can " + '"Upgrade" only to "WebSocket".' + ).format(headers.get(hdrs.UPGRADE)) + ) + + if "upgrade" not in headers.get(hdrs.CONNECTION, "").lower(): + raise HTTPBadRequest( + text="No CONNECTION upgrade hdr: {}".format( + headers.get(hdrs.CONNECTION) + ) + ) + + # find common sub-protocol between client and server + protocol = None + if hdrs.SEC_WEBSOCKET_PROTOCOL in headers: + req_protocols = [ + str(proto.strip()) + for proto in headers[hdrs.SEC_WEBSOCKET_PROTOCOL].split(",") + ] + + for proto in req_protocols: + if proto in self._protocols: + protocol = proto + break + else: + # No overlap found: Return no protocol as per spec + ws_logger.warning( + "Client protocols %r don’t overlap server-known ones %r", + req_protocols, + self._protocols, + ) + + # check supported version + version = headers.get(hdrs.SEC_WEBSOCKET_VERSION, "") + if version not in ("13", "8", "7"): + raise HTTPBadRequest(text=f"Unsupported version: {version}") + + # check client handshake for validity + key = headers.get(hdrs.SEC_WEBSOCKET_KEY) + try: + if not key or len(base64.b64decode(key)) != 16: + raise HTTPBadRequest(text=f"Handshake error: {key!r}") + except binascii.Error: + raise HTTPBadRequest(text=f"Handshake error: {key!r}") from None + + accept_val = base64.b64encode( + hashlib.sha1(key.encode() + WS_KEY).digest() + ).decode() + response_headers = CIMultiDict( # type: ignore[var-annotated] + { + hdrs.UPGRADE: "websocket", # type: ignore[arg-type] + hdrs.CONNECTION: "upgrade", + hdrs.SEC_WEBSOCKET_ACCEPT: accept_val, + } + ) + + notakeover = False + compress = 0 + if self._compress: + extensions = headers.get(hdrs.SEC_WEBSOCKET_EXTENSIONS) + # Server side always get return with no exception. + # If something happened, just drop compress extension + compress, notakeover = ws_ext_parse(extensions, isserver=True) + if compress: + enabledext = ws_ext_gen( + compress=compress, isserver=True, server_notakeover=notakeover + ) + response_headers[hdrs.SEC_WEBSOCKET_EXTENSIONS] = enabledext + + if protocol: + response_headers[hdrs.SEC_WEBSOCKET_PROTOCOL] = protocol + return ( + response_headers, + protocol, + compress, + notakeover, + ) # type: ignore[return-value] + + def _pre_start(self, request: BaseRequest) -> Tuple[str, WebSocketWriter]: + self._loop = request._loop + + headers, protocol, compress, notakeover = self._handshake(request) + + self.set_status(101) + self.headers.update(headers) + self.force_close() + self._compress = compress + transport = request._protocol.transport + assert transport is not None + writer = WebSocketWriter( + request._protocol, transport, compress=compress, notakeover=notakeover + ) + + return protocol, writer + + def _post_start( + self, request: BaseRequest, protocol: str, writer: WebSocketWriter + ) -> None: + self._ws_protocol = protocol + self._writer = writer + + self._reset_heartbeat() + + loop = self._loop + assert loop is not None + self._reader = FlowControlDataQueue(request._protocol, 2 ** 16, loop=loop) + request.protocol.set_parser( + WebSocketReader(self._reader, self._max_msg_size, compress=self._compress) + ) + # disable HTTP keepalive for WebSocket + request.protocol.keep_alive(False) + + def can_prepare(self, request: BaseRequest) -> WebSocketReady: + if self._writer is not None: + raise RuntimeError("Already started") + try: + _, protocol, _, _ = self._handshake(request) + except HTTPException: + return WebSocketReady(False, None) + else: + return WebSocketReady(True, protocol) + + @property + def closed(self) -> bool: + return self._closed + + @property + def close_code(self) -> Optional[int]: + return self._close_code + + @property + def ws_protocol(self) -> Optional[str]: + return self._ws_protocol + + @property + def compress(self) -> bool: + return self._compress + + def exception(self) -> Optional[BaseException]: + return self._exception + + async def ping(self, message: bytes = b"") -> None: + if self._writer is None: + raise RuntimeError("Call .prepare() first") + await self._writer.ping(message) + + async def pong(self, message: bytes = b"") -> None: + # unsolicited pong + if self._writer is None: + raise RuntimeError("Call .prepare() first") + await self._writer.pong(message) + + async def send_str(self, data: str, compress: Optional[bool] = None) -> None: + if self._writer is None: + raise RuntimeError("Call .prepare() first") + if not isinstance(data, str): + raise TypeError("data argument must be str (%r)" % type(data)) + await self._writer.send(data, binary=False, compress=compress) + + async def send_bytes(self, data: bytes, compress: Optional[bool] = None) -> None: + if self._writer is None: + raise RuntimeError("Call .prepare() first") + if not isinstance(data, (bytes, bytearray, memoryview)): + raise TypeError("data argument must be byte-ish (%r)" % type(data)) + await self._writer.send(data, binary=True, compress=compress) + + async def send_json( + self, + data: Any, + compress: Optional[bool] = None, + *, + dumps: JSONEncoder = json.dumps, + ) -> None: + await self.send_str(dumps(data), compress=compress) + + async def write_eof(self) -> None: # type: ignore[override] + if self._eof_sent: + return + if self._payload_writer is None: + raise RuntimeError("Response has not been started") + + await self.close() + self._eof_sent = True + + async def close(self, *, code: int = WSCloseCode.OK, message: bytes = b"") -> bool: + if self._writer is None: + raise RuntimeError("Call .prepare() first") + + self._cancel_heartbeat() + reader = self._reader + assert reader is not None + + # we need to break `receive()` cycle first, + # `close()` may be called from different task + if self._waiting is not None and not self._closed: + reader.feed_data(WS_CLOSING_MESSAGE, 0) + await self._waiting + + if not self._closed: + self._closed = True + try: + await self._writer.close(code, message) + writer = self._payload_writer + assert writer is not None + await writer.drain() + except (asyncio.CancelledError, asyncio.TimeoutError): + self._close_code = WSCloseCode.ABNORMAL_CLOSURE + raise + except Exception as exc: + self._close_code = WSCloseCode.ABNORMAL_CLOSURE + self._exception = exc + return True + + if self._closing: + return True + + reader = self._reader + assert reader is not None + try: + async with async_timeout.timeout(self._timeout): + msg = await reader.read() + except asyncio.CancelledError: + self._close_code = WSCloseCode.ABNORMAL_CLOSURE + raise + except Exception as exc: + self._close_code = WSCloseCode.ABNORMAL_CLOSURE + self._exception = exc + return True + + if msg.type == WSMsgType.CLOSE: + self._close_code = msg.data + return True + + self._close_code = WSCloseCode.ABNORMAL_CLOSURE + self._exception = asyncio.TimeoutError() + return True + else: + return False + + async def receive(self, timeout: Optional[float] = None) -> WSMessage: + if self._reader is None: + raise RuntimeError("Call .prepare() first") + + loop = self._loop + assert loop is not None + while True: + if self._waiting is not None: + raise RuntimeError("Concurrent call to receive() is not allowed") + + if self._closed: + self._conn_lost += 1 + if self._conn_lost >= THRESHOLD_CONNLOST_ACCESS: + raise RuntimeError("WebSocket connection is closed.") + return WS_CLOSED_MESSAGE + elif self._closing: + return WS_CLOSING_MESSAGE + + try: + self._waiting = loop.create_future() + try: + async with async_timeout.timeout(timeout or self._receive_timeout): + msg = await self._reader.read() + self._reset_heartbeat() + finally: + waiter = self._waiting + set_result(waiter, True) + self._waiting = None + except (asyncio.CancelledError, asyncio.TimeoutError): + self._close_code = WSCloseCode.ABNORMAL_CLOSURE + raise + except EofStream: + self._close_code = WSCloseCode.OK + await self.close() + return WSMessage(WSMsgType.CLOSED, None, None) + except WebSocketError as exc: + self._close_code = exc.code + await self.close(code=exc.code) + return WSMessage(WSMsgType.ERROR, exc, None) + except Exception as exc: + self._exception = exc + self._closing = True + self._close_code = WSCloseCode.ABNORMAL_CLOSURE + await self.close() + return WSMessage(WSMsgType.ERROR, exc, None) + + if msg.type == WSMsgType.CLOSE: + self._closing = True + self._close_code = msg.data + if not self._closed and self._autoclose: + await self.close() + elif msg.type == WSMsgType.CLOSING: + self._closing = True + elif msg.type == WSMsgType.PING and self._autoping: + await self.pong(msg.data) + continue + elif msg.type == WSMsgType.PONG and self._autoping: + continue + + return msg + + async def receive_str(self, *, timeout: Optional[float] = None) -> str: + msg = await self.receive(timeout) + if msg.type != WSMsgType.TEXT: + raise TypeError( + "Received message {}:{!r} is not WSMsgType.TEXT".format( + msg.type, msg.data + ) + ) + return cast(str, msg.data) + + async def receive_bytes(self, *, timeout: Optional[float] = None) -> bytes: + msg = await self.receive(timeout) + if msg.type != WSMsgType.BINARY: + raise TypeError(f"Received message {msg.type}:{msg.data!r} is not bytes") + return cast(bytes, msg.data) + + async def receive_json( + self, *, loads: JSONDecoder = json.loads, timeout: Optional[float] = None + ) -> Any: + data = await self.receive_str(timeout=timeout) + return loads(data) + + async def write(self, data: bytes) -> None: + raise RuntimeError("Cannot call .write() for websocket") + + def __aiter__(self) -> "WebSocketResponse": + return self + + async def __anext__(self) -> WSMessage: + msg = await self.receive() + if msg.type in (WSMsgType.CLOSE, WSMsgType.CLOSING, WSMsgType.CLOSED): + raise StopAsyncIteration + return msg + + def _cancel(self, exc: BaseException) -> None: + if self._reader is not None: + self._reader.set_exception(exc) diff --git a/contrib/python/aiohttp/aiohttp/worker.py b/contrib/python/aiohttp/aiohttp/worker.py new file mode 100644 index 0000000000..08945bcb4b --- /dev/null +++ b/contrib/python/aiohttp/aiohttp/worker.py @@ -0,0 +1,269 @@ +"""Async gunicorn worker for aiohttp.web""" + +import asyncio +import os +import re +import signal +import sys +from types import FrameType +from typing import Any, Awaitable, Callable, Optional, Union # noqa + +from gunicorn.config import AccessLogFormat as GunicornAccessLogFormat +from gunicorn.workers import base + +from aiohttp import web + +from .helpers import set_result +from .web_app import Application +from .web_log import AccessLogger + +try: + import ssl + + SSLContext = ssl.SSLContext +except ImportError: # pragma: no cover + ssl = None # type: ignore[assignment] + SSLContext = object # type: ignore[misc,assignment] + + +__all__ = ("GunicornWebWorker", "GunicornUVLoopWebWorker", "GunicornTokioWebWorker") + + +class GunicornWebWorker(base.Worker): # type: ignore[misc,no-any-unimported] + + DEFAULT_AIOHTTP_LOG_FORMAT = AccessLogger.LOG_FORMAT + DEFAULT_GUNICORN_LOG_FORMAT = GunicornAccessLogFormat.default + + def __init__(self, *args: Any, **kw: Any) -> None: # pragma: no cover + super().__init__(*args, **kw) + + self._task = None # type: Optional[asyncio.Task[None]] + self.exit_code = 0 + self._notify_waiter = None # type: Optional[asyncio.Future[bool]] + + def init_process(self) -> None: + # create new event_loop after fork + asyncio.get_event_loop().close() + + self.loop = asyncio.new_event_loop() + asyncio.set_event_loop(self.loop) + + super().init_process() + + def run(self) -> None: + self._task = self.loop.create_task(self._run()) + + try: # ignore all finalization problems + self.loop.run_until_complete(self._task) + except Exception: + self.log.exception("Exception in gunicorn worker") + self.loop.run_until_complete(self.loop.shutdown_asyncgens()) + self.loop.close() + + sys.exit(self.exit_code) + + async def _run(self) -> None: + runner = None + if isinstance(self.wsgi, Application): + app = self.wsgi + elif asyncio.iscoroutinefunction(self.wsgi): + wsgi = await self.wsgi() + if isinstance(wsgi, web.AppRunner): + runner = wsgi + app = runner.app + else: + app = wsgi + else: + raise RuntimeError( + "wsgi app should be either Application or " + "async function returning Application, got {}".format(self.wsgi) + ) + + if runner is None: + access_log = self.log.access_log if self.cfg.accesslog else None + runner = web.AppRunner( + app, + logger=self.log, + keepalive_timeout=self.cfg.keepalive, + access_log=access_log, + access_log_format=self._get_valid_log_format( + self.cfg.access_log_format + ), + ) + await runner.setup() + + ctx = self._create_ssl_context(self.cfg) if self.cfg.is_ssl else None + + runner = runner + assert runner is not None + server = runner.server + assert server is not None + for sock in self.sockets: + site = web.SockSite( + runner, + sock, + ssl_context=ctx, + shutdown_timeout=self.cfg.graceful_timeout / 100 * 95, + ) + await site.start() + + # If our parent changed then we shut down. + pid = os.getpid() + try: + while self.alive: # type: ignore[has-type] + self.notify() + + cnt = server.requests_count + if self.cfg.max_requests and cnt > self.cfg.max_requests: + self.alive = False + self.log.info("Max requests, shutting down: %s", self) + + elif pid == os.getpid() and self.ppid != os.getppid(): + self.alive = False + self.log.info("Parent changed, shutting down: %s", self) + else: + await self._wait_next_notify() + except BaseException: + pass + + await runner.cleanup() + + def _wait_next_notify(self) -> "asyncio.Future[bool]": + self._notify_waiter_done() + + loop = self.loop + assert loop is not None + self._notify_waiter = waiter = loop.create_future() + self.loop.call_later(1.0, self._notify_waiter_done, waiter) + + return waiter + + def _notify_waiter_done( + self, waiter: Optional["asyncio.Future[bool]"] = None + ) -> None: + if waiter is None: + waiter = self._notify_waiter + if waiter is not None: + set_result(waiter, True) + + if waiter is self._notify_waiter: + self._notify_waiter = None + + def init_signals(self) -> None: + # Set up signals through the event loop API. + + self.loop.add_signal_handler( + signal.SIGQUIT, self.handle_quit, signal.SIGQUIT, None + ) + + self.loop.add_signal_handler( + signal.SIGTERM, self.handle_exit, signal.SIGTERM, None + ) + + self.loop.add_signal_handler( + signal.SIGINT, self.handle_quit, signal.SIGINT, None + ) + + self.loop.add_signal_handler( + signal.SIGWINCH, self.handle_winch, signal.SIGWINCH, None + ) + + self.loop.add_signal_handler( + signal.SIGUSR1, self.handle_usr1, signal.SIGUSR1, None + ) + + self.loop.add_signal_handler( + signal.SIGABRT, self.handle_abort, signal.SIGABRT, None + ) + + # Don't let SIGTERM and SIGUSR1 disturb active requests + # by interrupting system calls + signal.siginterrupt(signal.SIGTERM, False) + signal.siginterrupt(signal.SIGUSR1, False) + # Reset signals so Gunicorn doesn't swallow subprocess return codes + # See: https://github.com/aio-libs/aiohttp/issues/6130 + if sys.version_info < (3, 8): + # Starting from Python 3.8, + # the default child watcher is ThreadedChildWatcher. + # The watcher doesn't depend on SIGCHLD signal, + # there is no need to reset it. + signal.signal(signal.SIGCHLD, signal.SIG_DFL) + + def handle_quit(self, sig: int, frame: FrameType) -> None: + self.alive = False + + # worker_int callback + self.cfg.worker_int(self) + + # wakeup closing process + self._notify_waiter_done() + + def handle_abort(self, sig: int, frame: FrameType) -> None: + self.alive = False + self.exit_code = 1 + self.cfg.worker_abort(self) + sys.exit(1) + + @staticmethod + def _create_ssl_context(cfg: Any) -> "SSLContext": + """Creates SSLContext instance for usage in asyncio.create_server. + + See ssl.SSLSocket.__init__ for more details. + """ + if ssl is None: # pragma: no cover + raise RuntimeError("SSL is not supported.") + + ctx = ssl.SSLContext(cfg.ssl_version) + ctx.load_cert_chain(cfg.certfile, cfg.keyfile) + ctx.verify_mode = cfg.cert_reqs + if cfg.ca_certs: + ctx.load_verify_locations(cfg.ca_certs) + if cfg.ciphers: + ctx.set_ciphers(cfg.ciphers) + return ctx + + def _get_valid_log_format(self, source_format: str) -> str: + if source_format == self.DEFAULT_GUNICORN_LOG_FORMAT: + return self.DEFAULT_AIOHTTP_LOG_FORMAT + elif re.search(r"%\([^\)]+\)", source_format): + raise ValueError( + "Gunicorn's style options in form of `%(name)s` are not " + "supported for the log formatting. Please use aiohttp's " + "format specification to configure access log formatting: " + "http://docs.aiohttp.org/en/stable/logging.html" + "#format-specification" + ) + else: + return source_format + + +class GunicornUVLoopWebWorker(GunicornWebWorker): + def init_process(self) -> None: + import uvloop + + # Close any existing event loop before setting a + # new policy. + asyncio.get_event_loop().close() + + # Setup uvloop policy, so that every + # asyncio.get_event_loop() will create an instance + # of uvloop event loop. + asyncio.set_event_loop_policy(uvloop.EventLoopPolicy()) + + super().init_process() + + +class GunicornTokioWebWorker(GunicornWebWorker): + def init_process(self) -> None: # pragma: no cover + import tokio + + # Close any existing event loop before setting a + # new policy. + asyncio.get_event_loop().close() + + # Setup tokio policy, so that every + # asyncio.get_event_loop() will create an instance + # of tokio event loop. + asyncio.set_event_loop_policy(tokio.EventLoopPolicy()) + + super().init_process() diff --git a/contrib/python/aiohttp/ya.make b/contrib/python/aiohttp/ya.make new file mode 100644 index 0000000000..9fd0c0df74 --- /dev/null +++ b/contrib/python/aiohttp/ya.make @@ -0,0 +1,101 @@ +# Generated by devtools/yamaker (pypi). + +PY3_LIBRARY() + +VERSION(3.8.1) + +LICENSE(Apache-2.0) + +PEERDIR( + contrib/python/aiosignal + contrib/python/async-timeout + contrib/python/attrs + contrib/python/charset-normalizer + contrib/python/frozenlist + contrib/python/multidict + contrib/python/yarl + contrib/restricted/llhttp +) + +ADDINCL( + contrib/restricted/llhttp/include + contrib/python/aiohttp/aiohttp + FOR cython contrib/python/aiohttp +) + +NO_COMPILER_WARNINGS() + +NO_LINT() + +NO_CHECK_IMPORTS( + aiohttp.pytest_plugin + aiohttp.worker +) + +SRCS( + aiohttp/_find_header.c +) + +PY_SRCS( + TOP_LEVEL + aiohttp/__init__.py + aiohttp/_helpers.pyi + aiohttp/abc.py + aiohttp/base_protocol.py + aiohttp/client.py + aiohttp/client_exceptions.py + aiohttp/client_proto.py + aiohttp/client_reqrep.py + aiohttp/client_ws.py + aiohttp/connector.py + aiohttp/cookiejar.py + aiohttp/formdata.py + aiohttp/hdrs.py + aiohttp/helpers.py + aiohttp/http.py + aiohttp/http_exceptions.py + aiohttp/http_parser.py + aiohttp/http_websocket.py + aiohttp/http_writer.py + aiohttp/locks.py + aiohttp/log.py + aiohttp/multipart.py + aiohttp/payload.py + aiohttp/payload_streamer.py + aiohttp/pytest_plugin.py + aiohttp/resolver.py + aiohttp/streams.py + aiohttp/tcp_helpers.py + aiohttp/test_utils.py + aiohttp/tracing.py + aiohttp/typedefs.py + aiohttp/web.py + aiohttp/web_app.py + aiohttp/web_exceptions.py + aiohttp/web_fileresponse.py + aiohttp/web_log.py + aiohttp/web_middlewares.py + aiohttp/web_protocol.py + aiohttp/web_request.py + aiohttp/web_response.py + aiohttp/web_routedef.py + aiohttp/web_runner.py + aiohttp/web_server.py + aiohttp/web_urldispatcher.py + aiohttp/web_ws.py + aiohttp/worker.py + CYTHON_C + aiohttp/_helpers.pyx + aiohttp/_http_parser.pyx + aiohttp/_http_writer.pyx + aiohttp/_websocket.pyx +) + +RESOURCE_FILES( + PREFIX contrib/python/aiohttp/ + .dist-info/METADATA + .dist-info/top_level.txt + aiohttp/py.typed +) + +END() diff --git a/contrib/python/aiosignal/.dist-info/METADATA b/contrib/python/aiosignal/.dist-info/METADATA new file mode 100644 index 0000000000..fc964525f0 --- /dev/null +++ b/contrib/python/aiosignal/.dist-info/METADATA @@ -0,0 +1,128 @@ +Metadata-Version: 2.1 +Name: aiosignal +Version: 1.3.1 +Summary: aiosignal: a list of registered asynchronous callbacks +Home-page: https://github.com/aio-libs/aiosignal +Maintainer: aiohttp team <team@aiohttp.org> +Maintainer-email: team@aiohttp.org +License: Apache 2.0 +Project-URL: Chat: Gitter, https://gitter.im/aio-libs/Lobby +Project-URL: CI: GitHub Actions, https://github.com/aio-libs/aiosignal/actions +Project-URL: Coverage: codecov, https://codecov.io/github/aio-libs/aiosignal +Project-URL: Docs: RTD, https://docs.aiosignal.org +Project-URL: GitHub: issues, https://github.com/aio-libs/aiosignal/issues +Project-URL: GitHub: repo, https://github.com/aio-libs/aiosignal +Classifier: License :: OSI Approved :: Apache Software License +Classifier: Intended Audience :: Developers +Classifier: Programming Language :: Python +Classifier: Programming Language :: Python :: 3 +Classifier: Programming Language :: Python :: 3 :: Only +Classifier: Programming Language :: Python :: 3.7 +Classifier: Programming Language :: Python :: 3.8 +Classifier: Programming Language :: Python :: 3.9 +Classifier: Programming Language :: Python :: 3.10 +Classifier: Programming Language :: Python :: 3.11 +Classifier: Development Status :: 5 - Production/Stable +Classifier: Operating System :: POSIX +Classifier: Operating System :: MacOS :: MacOS X +Classifier: Operating System :: Microsoft :: Windows +Classifier: Framework :: AsyncIO +Requires-Python: >=3.7 +Description-Content-Type: text/x-rst +License-File: LICENSE +Requires-Dist: frozenlist (>=1.1.0) + +========= +aiosignal +========= + +.. image:: https://github.com/aio-libs/aiosignal/workflows/CI/badge.svg + :target: https://github.com/aio-libs/aiosignal/actions?query=workflow%3ACI + :alt: GitHub status for master branch + +.. image:: https://codecov.io/gh/aio-libs/aiosignal/branch/master/graph/badge.svg + :target: https://codecov.io/gh/aio-libs/aiosignal + :alt: codecov.io status for master branch + +.. image:: https://badge.fury.io/py/aiosignal.svg + :target: https://pypi.org/project/aiosignal + :alt: Latest PyPI package version + +.. image:: https://readthedocs.org/projects/aiosignal/badge/?version=latest + :target: https://aiosignal.readthedocs.io/ + :alt: Latest Read The Docs + +.. image:: https://img.shields.io/discourse/topics?server=https%3A%2F%2Faio-libs.discourse.group%2F + :target: https://aio-libs.discourse.group/ + :alt: Discourse group for io-libs + +.. image:: https://badges.gitter.im/Join%20Chat.svg + :target: https://gitter.im/aio-libs/Lobby + :alt: Chat on Gitter + +Introduction +============ + +A project to manage callbacks in `asyncio` projects. + +``Signal`` is a list of registered asynchronous callbacks. + +The signal's life-cycle has two stages: after creation its content +could be filled by using standard list operations: ``sig.append()`` +etc. + +After you call ``sig.freeze()`` the signal is *frozen*: adding, removing +and dropping callbacks is forbidden. + +The only available operation is calling the previously registered +callbacks by using ``await sig.send(data)``. + +For concrete usage examples see the `Signals +<https://docs.aiohttp.org/en/stable/web_advanced.html#aiohttp-web-signals> +section of the `Web Server Advanced +<https://docs.aiohttp.org/en/stable/web_advanced.html>` chapter of the `aiohttp +documentation`_. + + +Installation +------------ + +:: + + $ pip install aiosignal + +The library requires Python 3.6 or newer. + + +Documentation +============= + +https://aiosignal.readthedocs.io/ + +Communication channels +====================== + +*gitter chat* https://gitter.im/aio-libs/Lobby + +Requirements +============ + +- Python >= 3.6 +- frozenlist >= 1.0.0 + +License +======= + +``aiosignal`` is offered under the Apache 2 license. + +Source code +=========== + +The project is hosted on GitHub_ + +Please file an issue in the `bug tracker +<https://github.com/aio-libs/aiosignal/issues>`_ if you have found a bug +or have some suggestions to improve the library. + +.. _GitHub: https://github.com/aio-libs/aiosignal +.. _aiohttp documentation: https://docs.aiohttp.org/ diff --git a/contrib/python/aiosignal/.dist-info/top_level.txt b/contrib/python/aiosignal/.dist-info/top_level.txt new file mode 100644 index 0000000000..ac6df3afe7 --- /dev/null +++ b/contrib/python/aiosignal/.dist-info/top_level.txt @@ -0,0 +1 @@ +aiosignal diff --git a/contrib/python/aiosignal/LICENSE b/contrib/python/aiosignal/LICENSE new file mode 100644 index 0000000000..7082a2d5b9 --- /dev/null +++ b/contrib/python/aiosignal/LICENSE @@ -0,0 +1,201 @@ +Apache License + Version 2.0, January 2004 + http://www.apache.org/licenses/ + + TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION + + 1. Definitions. + + "License" shall mean the terms and conditions for use, reproduction, + and distribution as defined by Sections 1 through 9 of this document. + + "Licensor" shall mean the copyright owner or entity authorized by + the copyright owner that is granting the License. + + "Legal Entity" shall mean the union of the acting entity and all + other entities that control, are controlled by, or are under common + control with that entity. For the purposes of this definition, + "control" means (i) the power, direct or indirect, to cause the + direction or management of such entity, whether by contract or + otherwise, or (ii) ownership of fifty percent (50%) or more of the + outstanding shares, or (iii) beneficial ownership of such entity. + + "You" (or "Your") shall mean an individual or Legal Entity + exercising permissions granted by this License. + + "Source" form shall mean the preferred form for making modifications, + including but not limited to software source code, documentation + source, and configuration files. + + "Object" form shall mean any form resulting from mechanical + transformation or translation of a Source form, including but + not limited to compiled object code, generated documentation, + and conversions to other media types. + + "Work" shall mean the work of authorship, whether in Source or + Object form, made available under the License, as indicated by a + copyright notice that is included in or attached to the work + (an example is provided in the Appendix below). + + "Derivative Works" shall mean any work, whether in Source or Object + form, that is based on (or derived from) the Work and for which the + editorial revisions, annotations, elaborations, or other modifications + represent, as a whole, an original work of authorship. For the purposes + of this License, Derivative Works shall not include works that remain + separable from, or merely link (or bind by name) to the interfaces of, + the Work and Derivative Works thereof. + + "Contribution" shall mean any work of authorship, including + the original version of the Work and any modifications or additions + to that Work or Derivative Works thereof, that is intentionally + submitted to Licensor for inclusion in the Work by the copyright owner + or by an individual or Legal Entity authorized to submit on behalf of + the copyright owner. For the purposes of this definition, "submitted" + means any form of electronic, verbal, or written communication sent + to the Licensor or its representatives, including but not limited to + communication on electronic mailing lists, source code control systems, + and issue tracking systems that are managed by, or on behalf of, the + Licensor for the purpose of discussing and improving the Work, but + excluding communication that is conspicuously marked or otherwise + designated in writing by the copyright owner as "Not a Contribution." + + "Contributor" shall mean Licensor and any individual or Legal Entity + on behalf of whom a Contribution has been received by Licensor and + subsequently incorporated within the Work. + + 2. Grant of Copyright License. Subject to the terms and conditions of + this License, each Contributor hereby grants to You a perpetual, + worldwide, non-exclusive, no-charge, royalty-free, irrevocable + copyright license to reproduce, prepare Derivative Works of, + publicly display, publicly perform, sublicense, and distribute the + Work and such Derivative Works in Source or Object form. + + 3. Grant of Patent License. Subject to the terms and conditions of + this License, each Contributor hereby grants to You a perpetual, + worldwide, non-exclusive, no-charge, royalty-free, irrevocable + (except as stated in this section) patent license to make, have made, + use, offer to sell, sell, import, and otherwise transfer the Work, + where such license applies only to those patent claims licensable + by such Contributor that are necessarily infringed by their + Contribution(s) alone or by combination of their Contribution(s) + with the Work to which such Contribution(s) was submitted. If You + institute patent litigation against any entity (including a + cross-claim or counterclaim in a lawsuit) alleging that the Work + or a Contribution incorporated within the Work constitutes direct + or contributory patent infringement, then any patent licenses + granted to You under this License for that Work shall terminate + as of the date such litigation is filed. + + 4. Redistribution. You may reproduce and distribute copies of the + Work or Derivative Works thereof in any medium, with or without + modifications, and in Source or Object form, provided that You + meet the following conditions: + + (a) You must give any other recipients of the Work or + Derivative Works a copy of this License; and + + (b) You must cause any modified files to carry prominent notices + stating that You changed the files; and + + (c) You must retain, in the Source form of any Derivative Works + that You distribute, all copyright, patent, trademark, and + attribution notices from the Source form of the Work, + excluding those notices that do not pertain to any part of + the Derivative Works; and + + (d) If the Work includes a "NOTICE" text file as part of its + distribution, then any Derivative Works that You distribute must + include a readable copy of the attribution notices contained + within such NOTICE file, excluding those notices that do not + pertain to any part of the Derivative Works, in at least one + of the following places: within a NOTICE text file distributed + as part of the Derivative Works; within the Source form or + documentation, if provided along with the Derivative Works; or, + within a display generated by the Derivative Works, if and + wherever such third-party notices normally appear. The contents + of the NOTICE file are for informational purposes only and + do not modify the License. You may add Your own attribution + notices within Derivative Works that You distribute, alongside + or as an addendum to the NOTICE text from the Work, provided + that such additional attribution notices cannot be construed + as modifying the License. + + You may add Your own copyright statement to Your modifications and + may provide additional or different license terms and conditions + for use, reproduction, or distribution of Your modifications, or + for any such Derivative Works as a whole, provided Your use, + reproduction, and distribution of the Work otherwise complies with + the conditions stated in this License. + + 5. Submission of Contributions. Unless You explicitly state otherwise, + any Contribution intentionally submitted for inclusion in the Work + by You to the Licensor shall be under the terms and conditions of + this License, without any additional terms or conditions. + Notwithstanding the above, nothing herein shall supersede or modify + the terms of any separate license agreement you may have executed + with Licensor regarding such Contributions. + + 6. Trademarks. This License does not grant permission to use the trade + names, trademarks, service marks, or product names of the Licensor, + except as required for reasonable and customary use in describing the + origin of the Work and reproducing the content of the NOTICE file. + + 7. Disclaimer of Warranty. Unless required by applicable law or + agreed to in writing, Licensor provides the Work (and each + Contributor provides its Contributions) on an "AS IS" BASIS, + WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or + implied, including, without limitation, any warranties or conditions + of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A + PARTICULAR PURPOSE. You are solely responsible for determining the + appropriateness of using or redistributing the Work and assume any + risks associated with Your exercise of permissions under this License. + + 8. Limitation of Liability. In no event and under no legal theory, + whether in tort (including negligence), contract, or otherwise, + unless required by applicable law (such as deliberate and grossly + negligent acts) or agreed to in writing, shall any Contributor be + liable to You for damages, including any direct, indirect, special, + incidental, or consequential damages of any character arising as a + result of this License or out of the use or inability to use the + Work (including but not limited to damages for loss of goodwill, + work stoppage, computer failure or malfunction, or any and all + other commercial damages or losses), even if such Contributor + has been advised of the possibility of such damages. + + 9. Accepting Warranty or Additional Liability. While redistributing + the Work or Derivative Works thereof, You may choose to offer, + and charge a fee for, acceptance of support, warranty, indemnity, + or other liability obligations and/or rights consistent with this + License. However, in accepting such obligations, You may act only + on Your own behalf and on Your sole responsibility, not on behalf + of any other Contributor, and only if You agree to indemnify, + defend, and hold each Contributor harmless for any liability + incurred by, or claims asserted against, such Contributor by reason + of your accepting any such warranty or additional liability. + + END OF TERMS AND CONDITIONS + + APPENDIX: How to apply the Apache License to your work. + + To apply the Apache License to your work, attach the following + boilerplate notice, with the fields enclosed by brackets "{}" + replaced with your own identifying information. (Don't include + the brackets!) The text should be enclosed in the appropriate + comment syntax for the file format. We also recommend that a + file or class name and description of purpose be included on the + same "printed page" as the copyright notice for easier + identification within third-party archives. + + Copyright 2013-2019 Nikolay Kim and Andrew Svetlov + + Licensed under the Apache License, Version 2.0 (the "License"); + you may not use this file except in compliance with the License. + You may obtain a copy of the License at + + http://www.apache.org/licenses/LICENSE-2.0 + + Unless required by applicable law or agreed to in writing, software + distributed under the License is distributed on an "AS IS" BASIS, + WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + See the License for the specific language governing permissions and + limitations under the License. diff --git a/contrib/python/aiosignal/README.rst b/contrib/python/aiosignal/README.rst new file mode 100644 index 0000000000..d21fc96bd6 --- /dev/null +++ b/contrib/python/aiosignal/README.rst @@ -0,0 +1,94 @@ +========= +aiosignal +========= + +.. image:: https://github.com/aio-libs/aiosignal/workflows/CI/badge.svg + :target: https://github.com/aio-libs/aiosignal/actions?query=workflow%3ACI + :alt: GitHub status for master branch + +.. image:: https://codecov.io/gh/aio-libs/aiosignal/branch/master/graph/badge.svg + :target: https://codecov.io/gh/aio-libs/aiosignal + :alt: codecov.io status for master branch + +.. image:: https://badge.fury.io/py/aiosignal.svg + :target: https://pypi.org/project/aiosignal + :alt: Latest PyPI package version + +.. image:: https://readthedocs.org/projects/aiosignal/badge/?version=latest + :target: https://aiosignal.readthedocs.io/ + :alt: Latest Read The Docs + +.. image:: https://img.shields.io/discourse/topics?server=https%3A%2F%2Faio-libs.discourse.group%2F + :target: https://aio-libs.discourse.group/ + :alt: Discourse group for io-libs + +.. image:: https://badges.gitter.im/Join%20Chat.svg + :target: https://gitter.im/aio-libs/Lobby + :alt: Chat on Gitter + +Introduction +============ + +A project to manage callbacks in `asyncio` projects. + +``Signal`` is a list of registered asynchronous callbacks. + +The signal's life-cycle has two stages: after creation its content +could be filled by using standard list operations: ``sig.append()`` +etc. + +After you call ``sig.freeze()`` the signal is *frozen*: adding, removing +and dropping callbacks is forbidden. + +The only available operation is calling the previously registered +callbacks by using ``await sig.send(data)``. + +For concrete usage examples see the `Signals +<https://docs.aiohttp.org/en/stable/web_advanced.html#aiohttp-web-signals> +section of the `Web Server Advanced +<https://docs.aiohttp.org/en/stable/web_advanced.html>` chapter of the `aiohttp +documentation`_. + + +Installation +------------ + +:: + + $ pip install aiosignal + +The library requires Python 3.6 or newer. + + +Documentation +============= + +https://aiosignal.readthedocs.io/ + +Communication channels +====================== + +*gitter chat* https://gitter.im/aio-libs/Lobby + +Requirements +============ + +- Python >= 3.6 +- frozenlist >= 1.0.0 + +License +======= + +``aiosignal`` is offered under the Apache 2 license. + +Source code +=========== + +The project is hosted on GitHub_ + +Please file an issue in the `bug tracker +<https://github.com/aio-libs/aiosignal/issues>`_ if you have found a bug +or have some suggestions to improve the library. + +.. _GitHub: https://github.com/aio-libs/aiosignal +.. _aiohttp documentation: https://docs.aiohttp.org/ diff --git a/contrib/python/aiosignal/aiosignal/__init__.py b/contrib/python/aiosignal/aiosignal/__init__.py new file mode 100644 index 0000000000..3d288e6ede --- /dev/null +++ b/contrib/python/aiosignal/aiosignal/__init__.py @@ -0,0 +1,36 @@ +from frozenlist import FrozenList + +__version__ = "1.3.1" + +__all__ = ("Signal",) + + +class Signal(FrozenList): + """Coroutine-based signal implementation. + + To connect a callback to a signal, use any list method. + + Signals are fired using the send() coroutine, which takes named + arguments. + """ + + __slots__ = ("_owner",) + + def __init__(self, owner): + super().__init__() + self._owner = owner + + def __repr__(self): + return "<Signal owner={}, frozen={}, {!r}>".format( + self._owner, self.frozen, list(self) + ) + + async def send(self, *args, **kwargs): + """ + Sends data to all registered receivers. + """ + if not self.frozen: + raise RuntimeError("Cannot send non-frozen signal.") + + for receiver in self: + await receiver(*args, **kwargs) # type: ignore diff --git a/ydb/public/sdk/python3/ydb/_grpc/__init__.py b/contrib/python/aiosignal/aiosignal/py.typed index e69de29bb2..e69de29bb2 100644 --- a/ydb/public/sdk/python3/ydb/_grpc/__init__.py +++ b/contrib/python/aiosignal/aiosignal/py.typed diff --git a/ydb/public/sdk/python3/ydb/_grpc/grpcwrapper/__init__.py b/contrib/python/aiosignal/tests/conftest.py index e69de29bb2..e69de29bb2 100644 --- a/ydb/public/sdk/python3/ydb/_grpc/grpcwrapper/__init__.py +++ b/contrib/python/aiosignal/tests/conftest.py diff --git a/contrib/python/aiosignal/tests/test_signals.py b/contrib/python/aiosignal/tests/test_signals.py new file mode 100644 index 0000000000..a040a78ae5 --- /dev/null +++ b/contrib/python/aiosignal/tests/test_signals.py @@ -0,0 +1,160 @@ +import re +from unittest import mock + +import pytest + +from aiosignal import Signal + + +class Owner: + def __repr__(self) -> str: + return "<Owner 0xdeadbeef>" + + +@pytest.fixture +def owner() -> Owner: + return Owner() + + +@pytest.mark.asyncio +async def test_add_signal_handler_not_a_callable(owner: Owner) -> None: + callback = True + signal = Signal(owner) + signal.append(callback) + signal.freeze() + with pytest.raises(TypeError): + await signal.send() + + +@pytest.mark.asyncio +async def test_function_signal_dispatch_kwargs(owner: Owner) -> None: + signal = Signal(owner) + kwargs = {"foo": 1, "bar": 2} + + callback_mock = mock.Mock() + + async def callback(**kwargs): + callback_mock(**kwargs) + + signal.append(callback) + signal.freeze() + + await signal.send(**kwargs) + callback_mock.assert_called_once_with(**kwargs) + + +@pytest.mark.asyncio +async def test_function_signal_dispatch_args_kwargs(owner: Owner) -> None: + signal = Signal(owner) + args = {"a", "b"} + kwargs = {"foo": 1, "bar": 2} + + callback_mock = mock.Mock() + + async def callback(*args, **kwargs): + callback_mock(*args, **kwargs) + + signal.append(callback) + signal.freeze() + + await signal.send(*args, **kwargs) + callback_mock.assert_called_once_with(*args, **kwargs) + + +@pytest.mark.asyncio +async def test_non_coroutine(owner: Owner) -> None: + signal = Signal(owner) + kwargs = {"foo": 1, "bar": 2} + + callback = mock.Mock() + + signal.append(callback) + signal.freeze() + + with pytest.raises(TypeError): + await signal.send(**kwargs) + + +def test_setitem(owner: Owner) -> None: + signal = Signal(owner) + m1 = mock.Mock() + signal.append(m1) + assert signal[0] is m1 + m2 = mock.Mock() + signal[0] = m2 + assert signal[0] is m2 + + +def test_delitem(owner: Owner) -> None: + signal = Signal(owner) + m1 = mock.Mock() + signal.append(m1) + assert len(signal) == 1 + del signal[0] + assert len(signal) == 0 + + +def test_cannot_append_to_frozen_signal(owner: Owner) -> None: + signal = Signal(owner) + m1 = mock.Mock() + m2 = mock.Mock() + signal.append(m1) + signal.freeze() + with pytest.raises(RuntimeError): + signal.append(m2) + + assert list(signal) == [m1] + + +def test_cannot_setitem_in_frozen_signal(owner: Owner) -> None: + signal = Signal(owner) + m1 = mock.Mock() + m2 = mock.Mock() + signal.append(m1) + signal.freeze() + with pytest.raises(RuntimeError): + signal[0] = m2 + + assert list(signal) == [m1] + + +def test_cannot_delitem_in_frozen_signal(owner: Owner) -> None: + signal = Signal(owner) + m1 = mock.Mock() + signal.append(m1) + signal.freeze() + with pytest.raises(RuntimeError): + del signal[0] + + assert list(signal) == [m1] + + +@pytest.mark.asyncio +async def test_cannot_send_non_frozen_signal(owner: Owner) -> None: + signal = Signal(owner) + + callback_mock = mock.Mock() + + async def callback(**kwargs): + callback_mock(**kwargs) + + signal.append(callback) + + with pytest.raises(RuntimeError): + await signal.send() + + assert not callback_mock.called + + +def test_repr(owner: Owner) -> None: + signal = Signal(owner) + + signal.append(mock.Mock(__repr__=lambda *a: "<callback>")) + + assert ( + re.match( + r"<Signal owner=<Owner 0xdeadbeef>, frozen=False, " r"\[<callback>\]>", + repr(signal), + ) + is not None + ) diff --git a/contrib/python/aiosignal/tests/ya.make b/contrib/python/aiosignal/tests/ya.make new file mode 100644 index 0000000000..edba214080 --- /dev/null +++ b/contrib/python/aiosignal/tests/ya.make @@ -0,0 +1,14 @@ +PY3TEST() + +PEERDIR( + contrib/python/aiosignal +) + +TEST_SRCS( + conftest.py + test_signals.py +) + +NO_LINT() + +END() diff --git a/contrib/python/aiosignal/ya.make b/contrib/python/aiosignal/ya.make new file mode 100644 index 0000000000..6094942c66 --- /dev/null +++ b/contrib/python/aiosignal/ya.make @@ -0,0 +1,32 @@ +# Generated by devtools/yamaker (pypi). + +PY3_LIBRARY() + +VERSION(1.3.1) + +LICENSE(Apache-2.0) + +PEERDIR( + contrib/python/frozenlist +) + +NO_LINT() + +PY_SRCS( + TOP_LEVEL + aiosignal/__init__.py + aiosignal/__init__.pyi +) + +RESOURCE_FILES( + PREFIX contrib/python/aiosignal/ + .dist-info/METADATA + .dist-info/top_level.txt + aiosignal/py.typed +) + +END() + +RECURSE_FOR_TESTS( + tests +) diff --git a/contrib/python/async-timeout/.dist-info/METADATA b/contrib/python/async-timeout/.dist-info/METADATA new file mode 100644 index 0000000000..d8dd6d12d6 --- /dev/null +++ b/contrib/python/async-timeout/.dist-info/METADATA @@ -0,0 +1,131 @@ +Metadata-Version: 2.1 +Name: async-timeout +Version: 4.0.3 +Summary: Timeout context manager for asyncio programs +Home-page: https://github.com/aio-libs/async-timeout +Author: Andrew Svetlov <andrew.svetlov@gmail.com> +Author-email: andrew.svetlov@gmail.com +License: Apache 2 +Project-URL: Chat: Gitter, https://gitter.im/aio-libs/Lobby +Project-URL: CI: GitHub Actions, https://github.com/aio-libs/async-timeout/actions +Project-URL: Coverage: codecov, https://codecov.io/github/aio-libs/async-timeout +Project-URL: GitHub: issues, https://github.com/aio-libs/async-timeout/issues +Project-URL: GitHub: repo, https://github.com/aio-libs/async-timeout +Classifier: Development Status :: 5 - Production/Stable +Classifier: Topic :: Software Development :: Libraries +Classifier: Framework :: AsyncIO +Classifier: Intended Audience :: Developers +Classifier: License :: OSI Approved :: Apache Software License +Classifier: Programming Language :: Python +Classifier: Programming Language :: Python :: 3 +Classifier: Programming Language :: Python :: 3 :: Only +Classifier: Programming Language :: Python :: 3.7 +Classifier: Programming Language :: Python :: 3.8 +Classifier: Programming Language :: Python :: 3.9 +Classifier: Programming Language :: Python :: 3.10 +Classifier: Programming Language :: Python :: 3.11 +Requires-Python: >=3.7 +Description-Content-Type: text/x-rst +License-File: LICENSE +Requires-Dist: typing-extensions >=3.6.5 ; python_version < "3.8" + +async-timeout +============= +.. image:: https://travis-ci.com/aio-libs/async-timeout.svg?branch=master + :target: https://travis-ci.com/aio-libs/async-timeout +.. image:: https://codecov.io/gh/aio-libs/async-timeout/branch/master/graph/badge.svg + :target: https://codecov.io/gh/aio-libs/async-timeout +.. image:: https://img.shields.io/pypi/v/async-timeout.svg + :target: https://pypi.python.org/pypi/async-timeout +.. image:: https://badges.gitter.im/Join%20Chat.svg + :target: https://gitter.im/aio-libs/Lobby + :alt: Chat on Gitter + +asyncio-compatible timeout context manager. + + +Usage example +------------- + + +The context manager is useful in cases when you want to apply timeout +logic around block of code or in cases when ``asyncio.wait_for()`` is +not suitable. Also it's much faster than ``asyncio.wait_for()`` +because ``timeout`` doesn't create a new task. + +The ``timeout(delay, *, loop=None)`` call returns a context manager +that cancels a block on *timeout* expiring:: + + from async_timeout import timeout + async with timeout(1.5): + await inner() + +1. If ``inner()`` is executed faster than in ``1.5`` seconds nothing + happens. +2. Otherwise ``inner()`` is cancelled internally by sending + ``asyncio.CancelledError`` into but ``asyncio.TimeoutError`` is + raised outside of context manager scope. + +*timeout* parameter could be ``None`` for skipping timeout functionality. + + +Alternatively, ``timeout_at(when)`` can be used for scheduling +at the absolute time:: + + loop = asyncio.get_event_loop() + now = loop.time() + + async with timeout_at(now + 1.5): + await inner() + + +Please note: it is not POSIX time but a time with +undefined starting base, e.g. the time of the system power on. + + +Context manager has ``.expired`` property for check if timeout happens +exactly in context manager:: + + async with timeout(1.5) as cm: + await inner() + print(cm.expired) + +The property is ``True`` if ``inner()`` execution is cancelled by +timeout context manager. + +If ``inner()`` call explicitly raises ``TimeoutError`` ``cm.expired`` +is ``False``. + +The scheduled deadline time is available as ``.deadline`` property:: + + async with timeout(1.5) as cm: + cm.deadline + +Not finished yet timeout can be rescheduled by ``shift_by()`` +or ``shift_to()`` methods:: + + async with timeout(1.5) as cm: + cm.shift(1) # add another second on waiting + cm.update(loop.time() + 5) # reschedule to now+5 seconds + +Rescheduling is forbidden if the timeout is expired or after exit from ``async with`` +code block. + + +Installation +------------ + +:: + + $ pip install async-timeout + +The library is Python 3 only! + + + +Authors and License +------------------- + +The module is written by Andrew Svetlov. + +It's *Apache 2* licensed and freely available. diff --git a/contrib/python/async-timeout/.dist-info/top_level.txt b/contrib/python/async-timeout/.dist-info/top_level.txt new file mode 100644 index 0000000000..ad29955ef9 --- /dev/null +++ b/contrib/python/async-timeout/.dist-info/top_level.txt @@ -0,0 +1 @@ +async_timeout diff --git a/contrib/python/async-timeout/LICENSE b/contrib/python/async-timeout/LICENSE new file mode 100644 index 0000000000..033c86b7a4 --- /dev/null +++ b/contrib/python/async-timeout/LICENSE @@ -0,0 +1,13 @@ +Copyright 2016-2020 aio-libs collaboration. + +Licensed under the Apache License, Version 2.0 (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + + http://www.apache.org/licenses/LICENSE-2.0 + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. diff --git a/contrib/python/async-timeout/README.rst b/contrib/python/async-timeout/README.rst new file mode 100644 index 0000000000..5ed02e4e93 --- /dev/null +++ b/contrib/python/async-timeout/README.rst @@ -0,0 +1,100 @@ +async-timeout +============= +.. image:: https://travis-ci.com/aio-libs/async-timeout.svg?branch=master + :target: https://travis-ci.com/aio-libs/async-timeout +.. image:: https://codecov.io/gh/aio-libs/async-timeout/branch/master/graph/badge.svg + :target: https://codecov.io/gh/aio-libs/async-timeout +.. image:: https://img.shields.io/pypi/v/async-timeout.svg + :target: https://pypi.python.org/pypi/async-timeout +.. image:: https://badges.gitter.im/Join%20Chat.svg + :target: https://gitter.im/aio-libs/Lobby + :alt: Chat on Gitter + +asyncio-compatible timeout context manager. + + +Usage example +------------- + + +The context manager is useful in cases when you want to apply timeout +logic around block of code or in cases when ``asyncio.wait_for()`` is +not suitable. Also it's much faster than ``asyncio.wait_for()`` +because ``timeout`` doesn't create a new task. + +The ``timeout(delay, *, loop=None)`` call returns a context manager +that cancels a block on *timeout* expiring:: + + from async_timeout import timeout + async with timeout(1.5): + await inner() + +1. If ``inner()`` is executed faster than in ``1.5`` seconds nothing + happens. +2. Otherwise ``inner()`` is cancelled internally by sending + ``asyncio.CancelledError`` into but ``asyncio.TimeoutError`` is + raised outside of context manager scope. + +*timeout* parameter could be ``None`` for skipping timeout functionality. + + +Alternatively, ``timeout_at(when)`` can be used for scheduling +at the absolute time:: + + loop = asyncio.get_event_loop() + now = loop.time() + + async with timeout_at(now + 1.5): + await inner() + + +Please note: it is not POSIX time but a time with +undefined starting base, e.g. the time of the system power on. + + +Context manager has ``.expired`` property for check if timeout happens +exactly in context manager:: + + async with timeout(1.5) as cm: + await inner() + print(cm.expired) + +The property is ``True`` if ``inner()`` execution is cancelled by +timeout context manager. + +If ``inner()`` call explicitly raises ``TimeoutError`` ``cm.expired`` +is ``False``. + +The scheduled deadline time is available as ``.deadline`` property:: + + async with timeout(1.5) as cm: + cm.deadline + +Not finished yet timeout can be rescheduled by ``shift_by()`` +or ``shift_to()`` methods:: + + async with timeout(1.5) as cm: + cm.shift(1) # add another second on waiting + cm.update(loop.time() + 5) # reschedule to now+5 seconds + +Rescheduling is forbidden if the timeout is expired or after exit from ``async with`` +code block. + + +Installation +------------ + +:: + + $ pip install async-timeout + +The library is Python 3 only! + + + +Authors and License +------------------- + +The module is written by Andrew Svetlov. + +It's *Apache 2* licensed and freely available. diff --git a/contrib/python/async-timeout/async_timeout/__init__.py b/contrib/python/async-timeout/async_timeout/__init__.py new file mode 100644 index 0000000000..1ffb069fce --- /dev/null +++ b/contrib/python/async-timeout/async_timeout/__init__.py @@ -0,0 +1,239 @@ +import asyncio +import enum +import sys +import warnings +from types import TracebackType +from typing import Optional, Type + + +if sys.version_info >= (3, 8): + from typing import final +else: + from typing_extensions import final + + +if sys.version_info >= (3, 11): + + def _uncancel_task(task: "asyncio.Task[object]") -> None: + task.uncancel() + +else: + + def _uncancel_task(task: "asyncio.Task[object]") -> None: + pass + + +__version__ = "4.0.3" + + +__all__ = ("timeout", "timeout_at", "Timeout") + + +def timeout(delay: Optional[float]) -> "Timeout": + """timeout context manager. + + Useful in cases when you want to apply timeout logic around block + of code or in cases when asyncio.wait_for is not suitable. For example: + + >>> async with timeout(0.001): + ... async with aiohttp.get('https://github.com') as r: + ... await r.text() + + + delay - value in seconds or None to disable timeout logic + """ + loop = asyncio.get_running_loop() + if delay is not None: + deadline = loop.time() + delay # type: Optional[float] + else: + deadline = None + return Timeout(deadline, loop) + + +def timeout_at(deadline: Optional[float]) -> "Timeout": + """Schedule the timeout at absolute time. + + deadline argument points on the time in the same clock system + as loop.time(). + + Please note: it is not POSIX time but a time with + undefined starting base, e.g. the time of the system power on. + + >>> async with timeout_at(loop.time() + 10): + ... async with aiohttp.get('https://github.com') as r: + ... await r.text() + + + """ + loop = asyncio.get_running_loop() + return Timeout(deadline, loop) + + +class _State(enum.Enum): + INIT = "INIT" + ENTER = "ENTER" + TIMEOUT = "TIMEOUT" + EXIT = "EXIT" + + +@final +class Timeout: + # Internal class, please don't instantiate it directly + # Use timeout() and timeout_at() public factories instead. + # + # Implementation note: `async with timeout()` is preferred + # over `with timeout()`. + # While technically the Timeout class implementation + # doesn't need to be async at all, + # the `async with` statement explicitly points that + # the context manager should be used from async function context. + # + # This design allows to avoid many silly misusages. + # + # TimeoutError is raised immediately when scheduled + # if the deadline is passed. + # The purpose is to time out as soon as possible + # without waiting for the next await expression. + + __slots__ = ("_deadline", "_loop", "_state", "_timeout_handler", "_task") + + def __init__( + self, deadline: Optional[float], loop: asyncio.AbstractEventLoop + ) -> None: + self._loop = loop + self._state = _State.INIT + + self._task: Optional["asyncio.Task[object]"] = None + self._timeout_handler = None # type: Optional[asyncio.Handle] + if deadline is None: + self._deadline = None # type: Optional[float] + else: + self.update(deadline) + + def __enter__(self) -> "Timeout": + warnings.warn( + "with timeout() is deprecated, use async with timeout() instead", + DeprecationWarning, + stacklevel=2, + ) + self._do_enter() + return self + + def __exit__( + self, + exc_type: Optional[Type[BaseException]], + exc_val: Optional[BaseException], + exc_tb: Optional[TracebackType], + ) -> Optional[bool]: + self._do_exit(exc_type) + return None + + async def __aenter__(self) -> "Timeout": + self._do_enter() + return self + + async def __aexit__( + self, + exc_type: Optional[Type[BaseException]], + exc_val: Optional[BaseException], + exc_tb: Optional[TracebackType], + ) -> Optional[bool]: + self._do_exit(exc_type) + return None + + @property + def expired(self) -> bool: + """Is timeout expired during execution?""" + return self._state == _State.TIMEOUT + + @property + def deadline(self) -> Optional[float]: + return self._deadline + + def reject(self) -> None: + """Reject scheduled timeout if any.""" + # cancel is maybe better name but + # task.cancel() raises CancelledError in asyncio world. + if self._state not in (_State.INIT, _State.ENTER): + raise RuntimeError(f"invalid state {self._state.value}") + self._reject() + + def _reject(self) -> None: + self._task = None + if self._timeout_handler is not None: + self._timeout_handler.cancel() + self._timeout_handler = None + + def shift(self, delay: float) -> None: + """Advance timeout on delay seconds. + + The delay can be negative. + + Raise RuntimeError if shift is called when deadline is not scheduled + """ + deadline = self._deadline + if deadline is None: + raise RuntimeError("cannot shift timeout if deadline is not scheduled") + self.update(deadline + delay) + + def update(self, deadline: float) -> None: + """Set deadline to absolute value. + + deadline argument points on the time in the same clock system + as loop.time(). + + If new deadline is in the past the timeout is raised immediately. + + Please note: it is not POSIX time but a time with + undefined starting base, e.g. the time of the system power on. + """ + if self._state == _State.EXIT: + raise RuntimeError("cannot reschedule after exit from context manager") + if self._state == _State.TIMEOUT: + raise RuntimeError("cannot reschedule expired timeout") + if self._timeout_handler is not None: + self._timeout_handler.cancel() + self._deadline = deadline + if self._state != _State.INIT: + self._reschedule() + + def _reschedule(self) -> None: + assert self._state == _State.ENTER + deadline = self._deadline + if deadline is None: + return + + now = self._loop.time() + if self._timeout_handler is not None: + self._timeout_handler.cancel() + + self._task = asyncio.current_task() + if deadline <= now: + self._timeout_handler = self._loop.call_soon(self._on_timeout) + else: + self._timeout_handler = self._loop.call_at(deadline, self._on_timeout) + + def _do_enter(self) -> None: + if self._state != _State.INIT: + raise RuntimeError(f"invalid state {self._state.value}") + self._state = _State.ENTER + self._reschedule() + + def _do_exit(self, exc_type: Optional[Type[BaseException]]) -> None: + if exc_type is asyncio.CancelledError and self._state == _State.TIMEOUT: + assert self._task is not None + _uncancel_task(self._task) + self._timeout_handler = None + self._task = None + raise asyncio.TimeoutError + # timeout has not expired + self._state = _State.EXIT + self._reject() + return None + + def _on_timeout(self) -> None: + assert self._task is not None + self._task.cancel() + self._state = _State.TIMEOUT + # drop the reference early + self._timeout_handler = None diff --git a/contrib/python/async-timeout/async_timeout/py.typed b/contrib/python/async-timeout/async_timeout/py.typed new file mode 100644 index 0000000000..3b94f91573 --- /dev/null +++ b/contrib/python/async-timeout/async_timeout/py.typed @@ -0,0 +1 @@ +Placeholder diff --git a/contrib/python/async-timeout/ya.make b/contrib/python/async-timeout/ya.make new file mode 100644 index 0000000000..6bc2c940a1 --- /dev/null +++ b/contrib/python/async-timeout/ya.make @@ -0,0 +1,23 @@ +# Generated by devtools/yamaker (pypi). + +PY3_LIBRARY() + +VERSION(4.0.3) + +LICENSE(Apache-2.0) + +NO_LINT() + +PY_SRCS( + TOP_LEVEL + async_timeout/__init__.py +) + +RESOURCE_FILES( + PREFIX contrib/python/async-timeout/ + .dist-info/METADATA + .dist-info/top_level.txt + async_timeout/py.typed +) + +END() diff --git a/contrib/python/frozenlist/.dist-info/METADATA b/contrib/python/frozenlist/.dist-info/METADATA new file mode 100644 index 0000000000..0c9ca3b06f --- /dev/null +++ b/contrib/python/frozenlist/.dist-info/METADATA @@ -0,0 +1,150 @@ +Metadata-Version: 2.1 +Name: frozenlist +Version: 1.4.0 +Summary: A list-like structure which implements collections.abc.MutableSequence +Home-page: https://github.com/aio-libs/frozenlist +Maintainer: aiohttp team <team@aiohttp.org> +Maintainer-email: team@aiohttp.org +License: Apache 2 +Project-URL: Chat: Matrix, https://matrix.to/#/#aio-libs:matrix.org +Project-URL: Chat: Matrix Space, https://matrix.to/#/#aio-libs-space:matrix.org +Project-URL: CI: Github Actions, https://github.com/aio-libs/frozenlist/actions +Project-URL: Code of Conduct, https://github.com/aio-libs/.github/blob/master/CODE_OF_CONDUCT.md +Project-URL: Coverage: codecov, https://codecov.io/github/aio-libs/frozenlist +Project-URL: Docs: Changelog, https://github.com/aio-libs/frozenlist/blob/master/CHANGES.rst#changelog +Project-URL: Docs: RTD, https://frozenlist.aio-libs.org +Project-URL: GitHub: issues, https://github.com/aio-libs/frozenlist/issues +Project-URL: GitHub: repo, https://github.com/aio-libs/frozenlist +Classifier: License :: OSI Approved :: Apache Software License +Classifier: Intended Audience :: Developers +Classifier: Programming Language :: Python +Classifier: Programming Language :: Python :: 3 +Classifier: Programming Language :: Python :: 3.8 +Classifier: Programming Language :: Python :: 3.9 +Classifier: Programming Language :: Python :: 3.10 +Classifier: Programming Language :: Python :: 3.11 +Classifier: Development Status :: 5 - Production/Stable +Classifier: Operating System :: POSIX +Classifier: Operating System :: MacOS :: MacOS X +Classifier: Operating System :: Microsoft :: Windows +Requires-Python: >=3.8 +Description-Content-Type: text/x-rst +License-File: LICENSE + +========== +frozenlist +========== + +.. image:: https://github.com/aio-libs/frozenlist/workflows/CI/badge.svg + :target: https://github.com/aio-libs/frozenlist/actions + :alt: GitHub status for master branch + +.. image:: https://codecov.io/gh/aio-libs/frozenlist/branch/master/graph/badge.svg + :target: https://codecov.io/gh/aio-libs/frozenlist + :alt: codecov.io status for master branch + +.. image:: https://img.shields.io/pypi/v/frozenlist.svg?logo=Python&logoColor=white + :target: https://pypi.org/project/frozenlist + :alt: frozenlist @ PyPI + +.. image:: https://readthedocs.org/projects/frozenlist/badge/?version=latest + :target: https://frozenlist.aio-libs.org + :alt: Read The Docs build status badge + +.. image:: https://img.shields.io/matrix/aio-libs:matrix.org?label=Discuss%20on%20Matrix%20at%20%23aio-libs%3Amatrix.org&logo=matrix&server_fqdn=matrix.org&style=flat + :target: https://matrix.to/#/%23aio-libs:matrix.org + :alt: Matrix Room — #aio-libs:matrix.org + +.. image:: https://img.shields.io/matrix/aio-libs-space:matrix.org?label=Discuss%20on%20Matrix%20at%20%23aio-libs-space%3Amatrix.org&logo=matrix&server_fqdn=matrix.org&style=flat + :target: https://matrix.to/#/%23aio-libs-space:matrix.org + :alt: Matrix Space — #aio-libs-space:matrix.org + +Introduction +============ + +``frozenlist.FrozenList`` is a list-like structure which implements +``collections.abc.MutableSequence``. The list is *mutable* until ``FrozenList.freeze`` +is called, after which list modifications raise ``RuntimeError``: + + +>>> from frozenlist import FrozenList +>>> fl = FrozenList([17, 42]) +>>> fl.append('spam') +>>> fl.append('Vikings') +>>> fl +<FrozenList(frozen=False, [17, 42, 'spam', 'Vikings'])> +>>> fl.freeze() +>>> fl +<FrozenList(frozen=True, [17, 42, 'spam', 'Vikings'])> +>>> fl.frozen +True +>>> fl.append("Monty") +Traceback (most recent call last): + File "<stdin>", line 1, in <module> + File "frozenlist/_frozenlist.pyx", line 97, in frozenlist._frozenlist.FrozenList.append + self._check_frozen() + File "frozenlist/_frozenlist.pyx", line 19, in frozenlist._frozenlist.FrozenList._check_frozen + raise RuntimeError("Cannot modify frozen list.") +RuntimeError: Cannot modify frozen list. + + +FrozenList is also hashable, but only when frozen. Otherwise it also throws a RuntimeError: + + +>>> fl = FrozenList([17, 42, 'spam']) +>>> hash(fl) +Traceback (most recent call last): + File "<stdin>", line 1, in <module> + File "frozenlist/_frozenlist.pyx", line 111, in frozenlist._frozenlist.FrozenList.__hash__ + raise RuntimeError("Cannot hash unfrozen list.") +RuntimeError: Cannot hash unfrozen list. +>>> fl.freeze() +>>> hash(fl) +3713081631934410656 +>>> dictionary = {fl: 'Vikings'} # frozen fl can be a dict key +>>> dictionary +{<FrozenList(frozen=True, [1, 2])>: 'Vikings'} + + +Installation +------------ + +:: + + $ pip install frozenlist + +The library requires Python 3.8 or newer. + + +Documentation +============= + +https://frozenlist.aio-libs.org + +Communication channels +====================== + +We have a *Matrix Space* `#aio-libs-space:matrix.org +<https://matrix.to/#/%23aio-libs-space:matrix.org>`_ which is +also accessible via Gitter. + +Requirements +============ + +- Python >= 3.8 + +License +======= + +``frozenlist`` is offered under the Apache 2 license. + +Source code +=========== + +The project is hosted on GitHub_ + +Please file an issue in the `bug tracker +<https://github.com/aio-libs/frozenlist/issues>`_ if you have found a bug +or have some suggestions to improve the library. + +.. _GitHub: https://github.com/aio-libs/frozenlist diff --git a/contrib/python/frozenlist/.dist-info/top_level.txt b/contrib/python/frozenlist/.dist-info/top_level.txt new file mode 100644 index 0000000000..52f13fc459 --- /dev/null +++ b/contrib/python/frozenlist/.dist-info/top_level.txt @@ -0,0 +1 @@ +frozenlist diff --git a/contrib/python/frozenlist/LICENSE b/contrib/python/frozenlist/LICENSE new file mode 100644 index 0000000000..7082a2d5b9 --- /dev/null +++ b/contrib/python/frozenlist/LICENSE @@ -0,0 +1,201 @@ +Apache License + Version 2.0, January 2004 + http://www.apache.org/licenses/ + + TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION + + 1. Definitions. + + "License" shall mean the terms and conditions for use, reproduction, + and distribution as defined by Sections 1 through 9 of this document. + + "Licensor" shall mean the copyright owner or entity authorized by + the copyright owner that is granting the License. + + "Legal Entity" shall mean the union of the acting entity and all + other entities that control, are controlled by, or are under common + control with that entity. For the purposes of this definition, + "control" means (i) the power, direct or indirect, to cause the + direction or management of such entity, whether by contract or + otherwise, or (ii) ownership of fifty percent (50%) or more of the + outstanding shares, or (iii) beneficial ownership of such entity. + + "You" (or "Your") shall mean an individual or Legal Entity + exercising permissions granted by this License. + + "Source" form shall mean the preferred form for making modifications, + including but not limited to software source code, documentation + source, and configuration files. + + "Object" form shall mean any form resulting from mechanical + transformation or translation of a Source form, including but + not limited to compiled object code, generated documentation, + and conversions to other media types. + + "Work" shall mean the work of authorship, whether in Source or + Object form, made available under the License, as indicated by a + copyright notice that is included in or attached to the work + (an example is provided in the Appendix below). + + "Derivative Works" shall mean any work, whether in Source or Object + form, that is based on (or derived from) the Work and for which the + editorial revisions, annotations, elaborations, or other modifications + represent, as a whole, an original work of authorship. For the purposes + of this License, Derivative Works shall not include works that remain + separable from, or merely link (or bind by name) to the interfaces of, + the Work and Derivative Works thereof. + + "Contribution" shall mean any work of authorship, including + the original version of the Work and any modifications or additions + to that Work or Derivative Works thereof, that is intentionally + submitted to Licensor for inclusion in the Work by the copyright owner + or by an individual or Legal Entity authorized to submit on behalf of + the copyright owner. For the purposes of this definition, "submitted" + means any form of electronic, verbal, or written communication sent + to the Licensor or its representatives, including but not limited to + communication on electronic mailing lists, source code control systems, + and issue tracking systems that are managed by, or on behalf of, the + Licensor for the purpose of discussing and improving the Work, but + excluding communication that is conspicuously marked or otherwise + designated in writing by the copyright owner as "Not a Contribution." + + "Contributor" shall mean Licensor and any individual or Legal Entity + on behalf of whom a Contribution has been received by Licensor and + subsequently incorporated within the Work. + + 2. Grant of Copyright License. Subject to the terms and conditions of + this License, each Contributor hereby grants to You a perpetual, + worldwide, non-exclusive, no-charge, royalty-free, irrevocable + copyright license to reproduce, prepare Derivative Works of, + publicly display, publicly perform, sublicense, and distribute the + Work and such Derivative Works in Source or Object form. + + 3. Grant of Patent License. Subject to the terms and conditions of + this License, each Contributor hereby grants to You a perpetual, + worldwide, non-exclusive, no-charge, royalty-free, irrevocable + (except as stated in this section) patent license to make, have made, + use, offer to sell, sell, import, and otherwise transfer the Work, + where such license applies only to those patent claims licensable + by such Contributor that are necessarily infringed by their + Contribution(s) alone or by combination of their Contribution(s) + with the Work to which such Contribution(s) was submitted. If You + institute patent litigation against any entity (including a + cross-claim or counterclaim in a lawsuit) alleging that the Work + or a Contribution incorporated within the Work constitutes direct + or contributory patent infringement, then any patent licenses + granted to You under this License for that Work shall terminate + as of the date such litigation is filed. + + 4. Redistribution. You may reproduce and distribute copies of the + Work or Derivative Works thereof in any medium, with or without + modifications, and in Source or Object form, provided that You + meet the following conditions: + + (a) You must give any other recipients of the Work or + Derivative Works a copy of this License; and + + (b) You must cause any modified files to carry prominent notices + stating that You changed the files; and + + (c) You must retain, in the Source form of any Derivative Works + that You distribute, all copyright, patent, trademark, and + attribution notices from the Source form of the Work, + excluding those notices that do not pertain to any part of + the Derivative Works; and + + (d) If the Work includes a "NOTICE" text file as part of its + distribution, then any Derivative Works that You distribute must + include a readable copy of the attribution notices contained + within such NOTICE file, excluding those notices that do not + pertain to any part of the Derivative Works, in at least one + of the following places: within a NOTICE text file distributed + as part of the Derivative Works; within the Source form or + documentation, if provided along with the Derivative Works; or, + within a display generated by the Derivative Works, if and + wherever such third-party notices normally appear. The contents + of the NOTICE file are for informational purposes only and + do not modify the License. You may add Your own attribution + notices within Derivative Works that You distribute, alongside + or as an addendum to the NOTICE text from the Work, provided + that such additional attribution notices cannot be construed + as modifying the License. + + You may add Your own copyright statement to Your modifications and + may provide additional or different license terms and conditions + for use, reproduction, or distribution of Your modifications, or + for any such Derivative Works as a whole, provided Your use, + reproduction, and distribution of the Work otherwise complies with + the conditions stated in this License. + + 5. Submission of Contributions. Unless You explicitly state otherwise, + any Contribution intentionally submitted for inclusion in the Work + by You to the Licensor shall be under the terms and conditions of + this License, without any additional terms or conditions. + Notwithstanding the above, nothing herein shall supersede or modify + the terms of any separate license agreement you may have executed + with Licensor regarding such Contributions. + + 6. Trademarks. This License does not grant permission to use the trade + names, trademarks, service marks, or product names of the Licensor, + except as required for reasonable and customary use in describing the + origin of the Work and reproducing the content of the NOTICE file. + + 7. Disclaimer of Warranty. Unless required by applicable law or + agreed to in writing, Licensor provides the Work (and each + Contributor provides its Contributions) on an "AS IS" BASIS, + WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or + implied, including, without limitation, any warranties or conditions + of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A + PARTICULAR PURPOSE. You are solely responsible for determining the + appropriateness of using or redistributing the Work and assume any + risks associated with Your exercise of permissions under this License. + + 8. Limitation of Liability. In no event and under no legal theory, + whether in tort (including negligence), contract, or otherwise, + unless required by applicable law (such as deliberate and grossly + negligent acts) or agreed to in writing, shall any Contributor be + liable to You for damages, including any direct, indirect, special, + incidental, or consequential damages of any character arising as a + result of this License or out of the use or inability to use the + Work (including but not limited to damages for loss of goodwill, + work stoppage, computer failure or malfunction, or any and all + other commercial damages or losses), even if such Contributor + has been advised of the possibility of such damages. + + 9. Accepting Warranty or Additional Liability. While redistributing + the Work or Derivative Works thereof, You may choose to offer, + and charge a fee for, acceptance of support, warranty, indemnity, + or other liability obligations and/or rights consistent with this + License. However, in accepting such obligations, You may act only + on Your own behalf and on Your sole responsibility, not on behalf + of any other Contributor, and only if You agree to indemnify, + defend, and hold each Contributor harmless for any liability + incurred by, or claims asserted against, such Contributor by reason + of your accepting any such warranty or additional liability. + + END OF TERMS AND CONDITIONS + + APPENDIX: How to apply the Apache License to your work. + + To apply the Apache License to your work, attach the following + boilerplate notice, with the fields enclosed by brackets "{}" + replaced with your own identifying information. (Don't include + the brackets!) The text should be enclosed in the appropriate + comment syntax for the file format. We also recommend that a + file or class name and description of purpose be included on the + same "printed page" as the copyright notice for easier + identification within third-party archives. + + Copyright 2013-2019 Nikolay Kim and Andrew Svetlov + + Licensed under the Apache License, Version 2.0 (the "License"); + you may not use this file except in compliance with the License. + You may obtain a copy of the License at + + http://www.apache.org/licenses/LICENSE-2.0 + + Unless required by applicable law or agreed to in writing, software + distributed under the License is distributed on an "AS IS" BASIS, + WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + See the License for the specific language governing permissions and + limitations under the License. diff --git a/contrib/python/frozenlist/README.rst b/contrib/python/frozenlist/README.rst new file mode 100644 index 0000000000..0e864cf28d --- /dev/null +++ b/contrib/python/frozenlist/README.rst @@ -0,0 +1,117 @@ +========== +frozenlist +========== + +.. image:: https://github.com/aio-libs/frozenlist/workflows/CI/badge.svg + :target: https://github.com/aio-libs/frozenlist/actions + :alt: GitHub status for master branch + +.. image:: https://codecov.io/gh/aio-libs/frozenlist/branch/master/graph/badge.svg + :target: https://codecov.io/gh/aio-libs/frozenlist + :alt: codecov.io status for master branch + +.. image:: https://img.shields.io/pypi/v/frozenlist.svg?logo=Python&logoColor=white + :target: https://pypi.org/project/frozenlist + :alt: frozenlist @ PyPI + +.. image:: https://readthedocs.org/projects/frozenlist/badge/?version=latest + :target: https://frozenlist.aio-libs.org + :alt: Read The Docs build status badge + +.. image:: https://img.shields.io/matrix/aio-libs:matrix.org?label=Discuss%20on%20Matrix%20at%20%23aio-libs%3Amatrix.org&logo=matrix&server_fqdn=matrix.org&style=flat + :target: https://matrix.to/#/%23aio-libs:matrix.org + :alt: Matrix Room — #aio-libs:matrix.org + +.. image:: https://img.shields.io/matrix/aio-libs-space:matrix.org?label=Discuss%20on%20Matrix%20at%20%23aio-libs-space%3Amatrix.org&logo=matrix&server_fqdn=matrix.org&style=flat + :target: https://matrix.to/#/%23aio-libs-space:matrix.org + :alt: Matrix Space — #aio-libs-space:matrix.org + +Introduction +============ + +``frozenlist.FrozenList`` is a list-like structure which implements +``collections.abc.MutableSequence``. The list is *mutable* until ``FrozenList.freeze`` +is called, after which list modifications raise ``RuntimeError``: + + +>>> from frozenlist import FrozenList +>>> fl = FrozenList([17, 42]) +>>> fl.append('spam') +>>> fl.append('Vikings') +>>> fl +<FrozenList(frozen=False, [17, 42, 'spam', 'Vikings'])> +>>> fl.freeze() +>>> fl +<FrozenList(frozen=True, [17, 42, 'spam', 'Vikings'])> +>>> fl.frozen +True +>>> fl.append("Monty") +Traceback (most recent call last): + File "<stdin>", line 1, in <module> + File "frozenlist/_frozenlist.pyx", line 97, in frozenlist._frozenlist.FrozenList.append + self._check_frozen() + File "frozenlist/_frozenlist.pyx", line 19, in frozenlist._frozenlist.FrozenList._check_frozen + raise RuntimeError("Cannot modify frozen list.") +RuntimeError: Cannot modify frozen list. + + +FrozenList is also hashable, but only when frozen. Otherwise it also throws a RuntimeError: + + +>>> fl = FrozenList([17, 42, 'spam']) +>>> hash(fl) +Traceback (most recent call last): + File "<stdin>", line 1, in <module> + File "frozenlist/_frozenlist.pyx", line 111, in frozenlist._frozenlist.FrozenList.__hash__ + raise RuntimeError("Cannot hash unfrozen list.") +RuntimeError: Cannot hash unfrozen list. +>>> fl.freeze() +>>> hash(fl) +3713081631934410656 +>>> dictionary = {fl: 'Vikings'} # frozen fl can be a dict key +>>> dictionary +{<FrozenList(frozen=True, [1, 2])>: 'Vikings'} + + +Installation +------------ + +:: + + $ pip install frozenlist + +The library requires Python 3.8 or newer. + + +Documentation +============= + +https://frozenlist.aio-libs.org + +Communication channels +====================== + +We have a *Matrix Space* `#aio-libs-space:matrix.org +<https://matrix.to/#/%23aio-libs-space:matrix.org>`_ which is +also accessible via Gitter. + +Requirements +============ + +- Python >= 3.8 + +License +======= + +``frozenlist`` is offered under the Apache 2 license. + +Source code +=========== + +The project is hosted on GitHub_ + +Please file an issue in the `bug tracker +<https://github.com/aio-libs/frozenlist/issues>`_ if you have found a bug +or have some suggestions to improve the library. + +.. _GitHub: https://github.com/aio-libs/frozenlist diff --git a/contrib/python/frozenlist/frozenlist/__init__.py b/contrib/python/frozenlist/frozenlist/__init__.py new file mode 100644 index 0000000000..152356588d --- /dev/null +++ b/contrib/python/frozenlist/frozenlist/__init__.py @@ -0,0 +1,95 @@ +import os +import sys +import types +from collections.abc import MutableSequence +from functools import total_ordering +from typing import Type + +__version__ = "1.4.0" + +__all__ = ("FrozenList", "PyFrozenList") # type: Tuple[str, ...] + + +NO_EXTENSIONS = bool(os.environ.get("FROZENLIST_NO_EXTENSIONS")) # type: bool + + +@total_ordering +class FrozenList(MutableSequence): + __slots__ = ("_frozen", "_items") + + if sys.version_info >= (3, 9): + __class_getitem__ = classmethod(types.GenericAlias) + else: + + @classmethod + def __class_getitem__(cls: Type["FrozenList"]) -> Type["FrozenList"]: + return cls + + def __init__(self, items=None): + self._frozen = False + if items is not None: + items = list(items) + else: + items = [] + self._items = items + + @property + def frozen(self): + return self._frozen + + def freeze(self): + self._frozen = True + + def __getitem__(self, index): + return self._items[index] + + def __setitem__(self, index, value): + if self._frozen: + raise RuntimeError("Cannot modify frozen list.") + self._items[index] = value + + def __delitem__(self, index): + if self._frozen: + raise RuntimeError("Cannot modify frozen list.") + del self._items[index] + + def __len__(self): + return self._items.__len__() + + def __iter__(self): + return self._items.__iter__() + + def __reversed__(self): + return self._items.__reversed__() + + def __eq__(self, other): + return list(self) == other + + def __le__(self, other): + return list(self) <= other + + def insert(self, pos, item): + if self._frozen: + raise RuntimeError("Cannot modify frozen list.") + self._items.insert(pos, item) + + def __repr__(self): + return f"<FrozenList(frozen={self._frozen}, {self._items!r})>" + + def __hash__(self): + if self._frozen: + return hash(tuple(self)) + else: + raise RuntimeError("Cannot hash unfrozen list.") + + +PyFrozenList = FrozenList + + +try: + from ._frozenlist import FrozenList as CFrozenList # type: ignore + + if not NO_EXTENSIONS: # pragma: no cover + FrozenList = CFrozenList # type: ignore +except ImportError: # pragma: no cover + pass diff --git a/contrib/python/frozenlist/frozenlist/_frozenlist.pyx b/contrib/python/frozenlist/frozenlist/_frozenlist.pyx new file mode 100644 index 0000000000..9ee846c1ae --- /dev/null +++ b/contrib/python/frozenlist/frozenlist/_frozenlist.pyx @@ -0,0 +1,123 @@ +import sys +import types +from collections.abc import MutableSequence + + +cdef class FrozenList: + + if sys.version_info >= (3, 9): + __class_getitem__ = classmethod(types.GenericAlias) + else: + @classmethod + def __class_getitem__(cls): + return cls + + cdef readonly bint frozen + cdef list _items + + def __init__(self, items=None): + self.frozen = False + if items is not None: + items = list(items) + else: + items = [] + self._items = items + + cdef object _check_frozen(self): + if self.frozen: + raise RuntimeError("Cannot modify frozen list.") + + cdef inline object _fast_len(self): + return len(self._items) + + def freeze(self): + self.frozen = True + + def __getitem__(self, index): + return self._items[index] + + def __setitem__(self, index, value): + self._check_frozen() + self._items[index] = value + + def __delitem__(self, index): + self._check_frozen() + del self._items[index] + + def __len__(self): + return self._fast_len() + + def __iter__(self): + return self._items.__iter__() + + def __reversed__(self): + return self._items.__reversed__() + + def __richcmp__(self, other, op): + if op == 0: # < + return list(self) < other + if op == 1: # <= + return list(self) <= other + if op == 2: # == + return list(self) == other + if op == 3: # != + return list(self) != other + if op == 4: # > + return list(self) > other + if op == 5: # => + return list(self) >= other + + def insert(self, pos, item): + self._check_frozen() + self._items.insert(pos, item) + + def __contains__(self, item): + return item in self._items + + def __iadd__(self, items): + self._check_frozen() + self._items += list(items) + return self + + def index(self, item): + return self._items.index(item) + + def remove(self, item): + self._check_frozen() + self._items.remove(item) + + def clear(self): + self._check_frozen() + self._items.clear() + + def extend(self, items): + self._check_frozen() + self._items += list(items) + + def reverse(self): + self._check_frozen() + self._items.reverse() + + def pop(self, index=-1): + self._check_frozen() + return self._items.pop(index) + + def append(self, item): + self._check_frozen() + return self._items.append(item) + + def count(self, item): + return self._items.count(item) + + def __repr__(self): + return '<FrozenList(frozen={}, {!r})>'.format(self.frozen, + self._items) + + def __hash__(self): + if self.frozen: + return hash(tuple(self._items)) + else: + raise RuntimeError("Cannot hash unfrozen list.") + + +MutableSequence.register(FrozenList) diff --git a/contrib/python/frozenlist/frozenlist/py.typed b/contrib/python/frozenlist/frozenlist/py.typed new file mode 100644 index 0000000000..f5642f79f2 --- /dev/null +++ b/contrib/python/frozenlist/frozenlist/py.typed @@ -0,0 +1 @@ +Marker diff --git a/ydb/public/sdk/python3/ydb/_topic_common/__init__.py b/contrib/python/frozenlist/tests/conftest.py index e69de29bb2..e69de29bb2 100644 --- a/ydb/public/sdk/python3/ydb/_topic_common/__init__.py +++ b/contrib/python/frozenlist/tests/conftest.py diff --git a/contrib/python/frozenlist/tests/test_frozenlist.py b/contrib/python/frozenlist/tests/test_frozenlist.py new file mode 100644 index 0000000000..f3b1a8a4df --- /dev/null +++ b/contrib/python/frozenlist/tests/test_frozenlist.py @@ -0,0 +1,246 @@ +from collections.abc import MutableSequence + +import pytest + +from frozenlist import FrozenList, PyFrozenList + + +class FrozenListMixin: + FrozenList = NotImplemented + + SKIP_METHODS = {"__abstractmethods__", "__slots__"} + + def test_subclass(self) -> None: + assert issubclass(self.FrozenList, MutableSequence) + + def test_iface(self) -> None: + for name in set(dir(MutableSequence)) - self.SKIP_METHODS: + if name.startswith("_") and not name.endswith("_"): + continue + assert hasattr(self.FrozenList, name) + + def test_ctor_default(self) -> None: + _list = self.FrozenList([]) + assert not _list.frozen + + def test_ctor(self) -> None: + _list = self.FrozenList([1]) + assert not _list.frozen + + def test_ctor_copy_list(self) -> None: + orig = [1] + _list = self.FrozenList(orig) + del _list[0] + assert _list != orig + + def test_freeze(self) -> None: + _list = self.FrozenList() + _list.freeze() + assert _list.frozen + + def test_repr(self) -> None: + _list = self.FrozenList([1]) + assert repr(_list) == "<FrozenList(frozen=False, [1])>" + _list.freeze() + assert repr(_list) == "<FrozenList(frozen=True, [1])>" + + def test_getitem(self) -> None: + _list = self.FrozenList([1, 2]) + assert _list[1] == 2 + + def test_setitem(self) -> None: + _list = self.FrozenList([1, 2]) + _list[1] = 3 + assert _list[1] == 3 + + def test_delitem(self) -> None: + _list = self.FrozenList([1, 2]) + del _list[0] + assert len(_list) == 1 + assert _list[0] == 2 + + def test_len(self) -> None: + _list = self.FrozenList([1]) + assert len(_list) == 1 + + def test_iter(self) -> None: + _list = self.FrozenList([1, 2]) + assert list(iter(_list)) == [1, 2] + + def test_reversed(self) -> None: + _list = self.FrozenList([1, 2]) + assert list(reversed(_list)) == [2, 1] + + def test_eq(self) -> None: + _list = self.FrozenList([1]) + assert _list == [1] + + def test_ne(self) -> None: + _list = self.FrozenList([1]) + assert _list != [2] + + def test_le(self) -> None: + _list = self.FrozenList([1]) + assert _list <= [1] + + def test_lt(self) -> None: + _list = self.FrozenList([1]) + assert _list <= [3] + + def test_ge(self) -> None: + _list = self.FrozenList([1]) + assert _list >= [1] + + def test_gt(self) -> None: + _list = self.FrozenList([2]) + assert _list > [1] + + def test_insert(self) -> None: + _list = self.FrozenList([2]) + _list.insert(0, 1) + assert _list == [1, 2] + + def test_frozen_setitem(self) -> None: + _list = self.FrozenList([1]) + _list.freeze() + with pytest.raises(RuntimeError): + _list[0] = 2 + + def test_frozen_delitem(self) -> None: + _list = self.FrozenList([1]) + _list.freeze() + with pytest.raises(RuntimeError): + del _list[0] + + def test_frozen_insert(self) -> None: + _list = self.FrozenList([1]) + _list.freeze() + with pytest.raises(RuntimeError): + _list.insert(0, 2) + + def test_contains(self) -> None: + _list = self.FrozenList([2]) + assert 2 in _list + + def test_iadd(self) -> None: + _list = self.FrozenList([1]) + _list += [2] + assert _list == [1, 2] + + def test_iadd_frozen(self) -> None: + _list = self.FrozenList([1]) + _list.freeze() + with pytest.raises(RuntimeError): + _list += [2] + assert _list == [1] + + def test_index(self) -> None: + _list = self.FrozenList([1]) + assert _list.index(1) == 0 + + def test_remove(self) -> None: + _list = self.FrozenList([1]) + _list.remove(1) + assert len(_list) == 0 + + def test_remove_frozen(self) -> None: + _list = self.FrozenList([1]) + _list.freeze() + with pytest.raises(RuntimeError): + _list.remove(1) + assert _list == [1] + + def test_clear(self) -> None: + _list = self.FrozenList([1]) + _list.clear() + assert len(_list) == 0 + + def test_clear_frozen(self) -> None: + _list = self.FrozenList([1]) + _list.freeze() + with pytest.raises(RuntimeError): + _list.clear() + assert _list == [1] + + def test_extend(self) -> None: + _list = self.FrozenList([1]) + _list.extend([2]) + assert _list == [1, 2] + + def test_extend_frozen(self) -> None: + _list = self.FrozenList([1]) + _list.freeze() + with pytest.raises(RuntimeError): + _list.extend([2]) + assert _list == [1] + + def test_reverse(self) -> None: + _list = self.FrozenList([1, 2]) + _list.reverse() + assert _list == [2, 1] + + def test_reverse_frozen(self) -> None: + _list = self.FrozenList([1, 2]) + _list.freeze() + with pytest.raises(RuntimeError): + _list.reverse() + assert _list == [1, 2] + + def test_pop(self) -> None: + _list = self.FrozenList([1, 2]) + assert _list.pop(0) == 1 + assert _list == [2] + + def test_pop_default(self) -> None: + _list = self.FrozenList([1, 2]) + assert _list.pop() == 2 + assert _list == [1] + + def test_pop_frozen(self) -> None: + _list = self.FrozenList([1, 2]) + _list.freeze() + with pytest.raises(RuntimeError): + _list.pop() + assert _list == [1, 2] + + def test_append(self) -> None: + _list = self.FrozenList([1, 2]) + _list.append(3) + assert _list == [1, 2, 3] + + def test_append_frozen(self) -> None: + _list = self.FrozenList([1, 2]) + _list.freeze() + with pytest.raises(RuntimeError): + _list.append(3) + assert _list == [1, 2] + + def test_hash(self) -> None: + _list = self.FrozenList([1, 2]) + with pytest.raises(RuntimeError): + hash(_list) + + def test_hash_frozen(self) -> None: + _list = self.FrozenList([1, 2]) + _list.freeze() + h = hash(_list) + assert h == hash((1, 2)) + + def test_dict_key(self) -> None: + _list = self.FrozenList([1, 2]) + with pytest.raises(RuntimeError): + {_list: "hello"} + _list.freeze() + {_list: "hello"} + + def test_count(self) -> None: + _list = self.FrozenList([1, 2]) + assert _list.count(1) == 1 + + +class TestFrozenList(FrozenListMixin): + FrozenList = FrozenList + + +class TestFrozenListPy(FrozenListMixin): + FrozenList = PyFrozenList diff --git a/contrib/python/frozenlist/tests/ya.make b/contrib/python/frozenlist/tests/ya.make new file mode 100644 index 0000000000..148532159a --- /dev/null +++ b/contrib/python/frozenlist/tests/ya.make @@ -0,0 +1,14 @@ +PY3TEST() + +PEERDIR( + contrib/python/frozenlist +) + +TEST_SRCS( + conftest.py + test_frozenlist.py +) + +NO_LINT() + +END() diff --git a/contrib/python/frozenlist/ya.make b/contrib/python/frozenlist/ya.make new file mode 100644 index 0000000000..4928a9f7da --- /dev/null +++ b/contrib/python/frozenlist/ya.make @@ -0,0 +1,32 @@ +# Generated by devtools/yamaker (pypi). + +PY3_LIBRARY() + +VERSION(1.4.0) + +LICENSE(Apache-2.0) + +NO_COMPILER_WARNINGS() + +NO_LINT() + +PY_SRCS( + TOP_LEVEL + frozenlist/__init__.py + frozenlist/__init__.pyi + CYTHON_CPP + frozenlist/_frozenlist.pyx +) + +RESOURCE_FILES( + PREFIX contrib/python/frozenlist/ + .dist-info/METADATA + .dist-info/top_level.txt + frozenlist/py.typed +) + +END() + +RECURSE_FOR_TESTS( + tests +) diff --git a/contrib/python/multidict/.dist-info/METADATA b/contrib/python/multidict/.dist-info/METADATA new file mode 100644 index 0000000000..55377fcd85 --- /dev/null +++ b/contrib/python/multidict/.dist-info/METADATA @@ -0,0 +1,130 @@ +Metadata-Version: 2.1 +Name: multidict +Version: 6.0.4 +Summary: multidict implementation +Home-page: https://github.com/aio-libs/multidict +Author: Andrew Svetlov +Author-email: andrew.svetlov@gmail.com +License: Apache 2 +Project-URL: Chat: Gitter, https://gitter.im/aio-libs/Lobby +Project-URL: CI: GitHub, https://github.com/aio-libs/multidict/actions +Project-URL: Coverage: codecov, https://codecov.io/github/aio-libs/multidict +Project-URL: Docs: RTD, https://multidict.readthedocs.io +Project-URL: GitHub: issues, https://github.com/aio-libs/multidict/issues +Project-URL: GitHub: repo, https://github.com/aio-libs/multidict +Classifier: License :: OSI Approved :: Apache Software License +Classifier: Intended Audience :: Developers +Classifier: Programming Language :: Python +Classifier: Programming Language :: Python :: 3 +Classifier: Programming Language :: Python :: 3.7 +Classifier: Programming Language :: Python :: 3.8 +Classifier: Programming Language :: Python :: 3.9 +Classifier: Programming Language :: Python :: 3.10 +Classifier: Programming Language :: Python :: 3.11 +Classifier: Development Status :: 5 - Production/Stable +Requires-Python: >=3.7 +License-File: LICENSE + +========= +multidict +========= + +.. image:: https://github.com/aio-libs/multidict/workflows/CI/badge.svg + :target: https://github.com/aio-libs/multidict/actions?query=workflow%3ACI + :alt: GitHub status for master branch + +.. image:: https://codecov.io/gh/aio-libs/multidict/branch/master/graph/badge.svg + :target: https://codecov.io/gh/aio-libs/multidict + :alt: Coverage metrics + +.. image:: https://img.shields.io/pypi/v/multidict.svg + :target: https://pypi.org/project/multidict + :alt: PyPI + +.. image:: https://readthedocs.org/projects/multidict/badge/?version=latest + :target: http://multidict.readthedocs.org/en/latest/?badge=latest + :alt: Documentationb + +.. image:: https://img.shields.io/pypi/pyversions/multidict.svg + :target: https://pypi.org/project/multidict + :alt: Python versions + +.. image:: https://badges.gitter.im/Join%20Chat.svg + :target: https://gitter.im/aio-libs/Lobby + :alt: Chat on Gitter + +Multidict is dict-like collection of *key-value pairs* where key +might occur more than once in the container. + +Introduction +------------ + +*HTTP Headers* and *URL query string* require specific data structure: +*multidict*. It behaves mostly like a regular ``dict`` but it may have +several *values* for the same *key* and *preserves insertion ordering*. + +The *key* is ``str`` (or ``istr`` for case-insensitive dictionaries). + +``multidict`` has four multidict classes: +``MultiDict``, ``MultiDictProxy``, ``CIMultiDict`` +and ``CIMultiDictProxy``. + +Immutable proxies (``MultiDictProxy`` and +``CIMultiDictProxy``) provide a dynamic view for the +proxied multidict, the view reflects underlying collection changes. They +implement the ``collections.abc.Mapping`` interface. + +Regular mutable (``MultiDict`` and ``CIMultiDict``) classes +implement ``collections.abc.MutableMapping`` and allows them to change +their own content. + + +*Case insensitive* (``CIMultiDict`` and +``CIMultiDictProxy``) assume the *keys* are case +insensitive, e.g.:: + + >>> dct = CIMultiDict(key='val') + >>> 'Key' in dct + True + >>> dct['Key'] + 'val' + +*Keys* should be ``str`` or ``istr`` instances. + +The library has optional C Extensions for speed. + + +License +------- + +Apache 2 + +Library Installation +-------------------- + +.. code-block:: bash + + $ pip install multidict + +The library is Python 3 only! + +PyPI contains binary wheels for Linux, Windows and MacOS. If you want to install +``multidict`` on another operating system (or *Alpine Linux* inside a Docker) the +tarball will be used to compile the library from source. It requires a C compiler and +Python headers to be installed. + +To skip the compilation, please use the `MULTIDICT_NO_EXTENSIONS` environment variable, +e.g.: + +.. code-block:: bash + + $ MULTIDICT_NO_EXTENSIONS=1 pip install multidict + +Please note, the pure Python (uncompiled) version is about 20-50 times slower depending on +the usage scenario!!! + + + +Changelog +--------- +See `RTD page <http://multidict.readthedocs.org/en/latest/changes.html>`_. diff --git a/contrib/python/multidict/.dist-info/top_level.txt b/contrib/python/multidict/.dist-info/top_level.txt new file mode 100644 index 0000000000..afcecdff08 --- /dev/null +++ b/contrib/python/multidict/.dist-info/top_level.txt @@ -0,0 +1 @@ +multidict diff --git a/contrib/python/multidict/LICENSE b/contrib/python/multidict/LICENSE new file mode 100644 index 0000000000..305eef6003 --- /dev/null +++ b/contrib/python/multidict/LICENSE @@ -0,0 +1,13 @@ + Copyright 2016-2021 Andrew Svetlov and aio-libs team + + Licensed under the Apache License, Version 2.0 (the "License"); + you may not use this file except in compliance with the License. + You may obtain a copy of the License at + + http://www.apache.org/licenses/LICENSE-2.0 + + Unless required by applicable law or agreed to in writing, software + distributed under the License is distributed on an "AS IS" BASIS, + WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + See the License for the specific language governing permissions and + limitations under the License. diff --git a/contrib/python/multidict/README.rst b/contrib/python/multidict/README.rst new file mode 100644 index 0000000000..0fb146c446 --- /dev/null +++ b/contrib/python/multidict/README.rst @@ -0,0 +1,103 @@ +========= +multidict +========= + +.. image:: https://github.com/aio-libs/multidict/workflows/CI/badge.svg + :target: https://github.com/aio-libs/multidict/actions?query=workflow%3ACI + :alt: GitHub status for master branch + +.. image:: https://codecov.io/gh/aio-libs/multidict/branch/master/graph/badge.svg + :target: https://codecov.io/gh/aio-libs/multidict + :alt: Coverage metrics + +.. image:: https://img.shields.io/pypi/v/multidict.svg + :target: https://pypi.org/project/multidict + :alt: PyPI + +.. image:: https://readthedocs.org/projects/multidict/badge/?version=latest + :target: http://multidict.readthedocs.org/en/latest/?badge=latest + :alt: Documentationb + +.. image:: https://img.shields.io/pypi/pyversions/multidict.svg + :target: https://pypi.org/project/multidict + :alt: Python versions + +.. image:: https://badges.gitter.im/Join%20Chat.svg + :target: https://gitter.im/aio-libs/Lobby + :alt: Chat on Gitter + +Multidict is dict-like collection of *key-value pairs* where key +might occur more than once in the container. + +Introduction +------------ + +*HTTP Headers* and *URL query string* require specific data structure: +*multidict*. It behaves mostly like a regular ``dict`` but it may have +several *values* for the same *key* and *preserves insertion ordering*. + +The *key* is ``str`` (or ``istr`` for case-insensitive dictionaries). + +``multidict`` has four multidict classes: +``MultiDict``, ``MultiDictProxy``, ``CIMultiDict`` +and ``CIMultiDictProxy``. + +Immutable proxies (``MultiDictProxy`` and +``CIMultiDictProxy``) provide a dynamic view for the +proxied multidict, the view reflects underlying collection changes. They +implement the ``collections.abc.Mapping`` interface. + +Regular mutable (``MultiDict`` and ``CIMultiDict``) classes +implement ``collections.abc.MutableMapping`` and allows them to change +their own content. + + +*Case insensitive* (``CIMultiDict`` and +``CIMultiDictProxy``) assume the *keys* are case +insensitive, e.g.:: + + >>> dct = CIMultiDict(key='val') + >>> 'Key' in dct + True + >>> dct['Key'] + 'val' + +*Keys* should be ``str`` or ``istr`` instances. + +The library has optional C Extensions for speed. + + +License +------- + +Apache 2 + +Library Installation +-------------------- + +.. code-block:: bash + + $ pip install multidict + +The library is Python 3 only! + +PyPI contains binary wheels for Linux, Windows and MacOS. If you want to install +``multidict`` on another operating system (or *Alpine Linux* inside a Docker) the +tarball will be used to compile the library from source. It requires a C compiler and +Python headers to be installed. + +To skip the compilation, please use the `MULTIDICT_NO_EXTENSIONS` environment variable, +e.g.: + +.. code-block:: bash + + $ MULTIDICT_NO_EXTENSIONS=1 pip install multidict + +Please note, the pure Python (uncompiled) version is about 20-50 times slower depending on +the usage scenario!!! + + + +Changelog +--------- +See `RTD page <http://multidict.readthedocs.org/en/latest/changes.html>`_. diff --git a/contrib/python/multidict/multidict/__init__.py b/contrib/python/multidict/multidict/__init__.py new file mode 100644 index 0000000000..d9ea722167 --- /dev/null +++ b/contrib/python/multidict/multidict/__init__.py @@ -0,0 +1,48 @@ +"""Multidict implementation. + +HTTP Headers and URL query string require specific data structure: +multidict. It behaves mostly like a dict but it can have +several values for the same key. +""" + +from ._abc import MultiMapping, MutableMultiMapping +from ._compat import USE_EXTENSIONS + +__all__ = ( + "MultiMapping", + "MutableMultiMapping", + "MultiDictProxy", + "CIMultiDictProxy", + "MultiDict", + "CIMultiDict", + "upstr", + "istr", + "getversion", +) + +__version__ = "6.0.4" + + +try: + if not USE_EXTENSIONS: + raise ImportError + from ._multidict import ( + CIMultiDict, + CIMultiDictProxy, + MultiDict, + MultiDictProxy, + getversion, + istr, + ) +except ImportError: # pragma: no cover + from ._multidict_py import ( + CIMultiDict, + CIMultiDictProxy, + MultiDict, + MultiDictProxy, + getversion, + istr, + ) + + +upstr = istr diff --git a/contrib/python/multidict/multidict/_abc.py b/contrib/python/multidict/multidict/_abc.py new file mode 100644 index 0000000000..0603cdd244 --- /dev/null +++ b/contrib/python/multidict/multidict/_abc.py @@ -0,0 +1,48 @@ +import abc +import sys +import types +from collections.abc import Mapping, MutableMapping + + +class _TypingMeta(abc.ABCMeta): + # A fake metaclass to satisfy typing deps in runtime + # basically MultiMapping[str] and other generic-like type instantiations + # are emulated. + # Note: real type hints are provided by __init__.pyi stub file + if sys.version_info >= (3, 9): + + def __getitem__(self, key): + return types.GenericAlias(self, key) + + else: + + def __getitem__(self, key): + return self + + +class MultiMapping(Mapping, metaclass=_TypingMeta): + @abc.abstractmethod + def getall(self, key, default=None): + raise KeyError + + @abc.abstractmethod + def getone(self, key, default=None): + raise KeyError + + +class MutableMultiMapping(MultiMapping, MutableMapping): + @abc.abstractmethod + def add(self, key, value): + raise NotImplementedError + + @abc.abstractmethod + def extend(self, *args, **kwargs): + raise NotImplementedError + + @abc.abstractmethod + def popone(self, key, default=None): + raise KeyError + + @abc.abstractmethod + def popall(self, key, default=None): + raise KeyError diff --git a/contrib/python/multidict/multidict/_compat.py b/contrib/python/multidict/multidict/_compat.py new file mode 100644 index 0000000000..d1ff392b25 --- /dev/null +++ b/contrib/python/multidict/multidict/_compat.py @@ -0,0 +1,14 @@ +import os +import platform + +NO_EXTENSIONS = bool(os.environ.get("MULTIDICT_NO_EXTENSIONS")) + +PYPY = platform.python_implementation() == "PyPy" + +USE_EXTENSIONS = not NO_EXTENSIONS and not PYPY + +if USE_EXTENSIONS: + try: + from . import _multidict # noqa + except ImportError: + USE_EXTENSIONS = False diff --git a/contrib/python/multidict/multidict/_multidict.c b/contrib/python/multidict/multidict/_multidict.c new file mode 100644 index 0000000000..1ba79df304 --- /dev/null +++ b/contrib/python/multidict/multidict/_multidict.c @@ -0,0 +1,1824 @@ +#include "Python.h" +#include "structmember.h" + +// Include order important +#include "_multilib/defs.h" +#include "_multilib/istr.h" +#include "_multilib/pair_list.h" +#include "_multilib/dict.h" +#include "_multilib/iter.h" +#include "_multilib/views.h" + +#ifndef _PyArg_UnpackKeywords +#define FASTCALL_OLD +#endif + + +static PyObject *collections_abc_mapping; +static PyObject *collections_abc_mut_mapping; +static PyObject *collections_abc_mut_multi_mapping; + +static PyTypeObject multidict_type; +static PyTypeObject cimultidict_type; +static PyTypeObject multidict_proxy_type; +static PyTypeObject cimultidict_proxy_type; + +static PyObject *repr_func; + +#define MultiDict_CheckExact(o) (Py_TYPE(o) == &multidict_type) +#define CIMultiDict_CheckExact(o) (Py_TYPE(o) == &cimultidict_type) +#define MultiDictProxy_CheckExact(o) (Py_TYPE(o) == &multidict_proxy_type) +#define CIMultiDictProxy_CheckExact(o) (Py_TYPE(o) == &cimultidict_proxy_type) + +/* Helper macro for something like isinstance(obj, Base) */ +#define _MultiDict_Check(o) \ + ((MultiDict_CheckExact(o)) || \ + (CIMultiDict_CheckExact(o)) || \ + (MultiDictProxy_CheckExact(o)) || \ + (CIMultiDictProxy_CheckExact(o))) + +/******************** Internal Methods ********************/ + +/* Forward declaration */ +static PyObject *multidict_items(MultiDictObject *self); + +static inline PyObject * +_multidict_getone(MultiDictObject *self, PyObject *key, PyObject *_default) +{ + PyObject *val = pair_list_get_one(&self->pairs, key); + + if (val == NULL && + PyErr_ExceptionMatches(PyExc_KeyError) && + _default != NULL) + { + PyErr_Clear(); + Py_INCREF(_default); + return _default; + } + + return val; +} + +static inline int +_multidict_eq(MultiDictObject *self, MultiDictObject *other) +{ + Py_ssize_t pos1 = 0, + pos2 = 0; + + Py_hash_t h1 = 0, + h2 = 0; + + PyObject *identity1 = NULL, + *identity2 = NULL, + *value1 = NULL, + *value2 = NULL; + + int cmp_identity = 0, + cmp_value = 0; + + if (self == other) { + return 1; + } + + if (pair_list_len(&self->pairs) != pair_list_len(&other->pairs)) { + return 0; + } + + while (_pair_list_next(&self->pairs, &pos1, &identity1, NULL, &value1, &h1) && + _pair_list_next(&other->pairs, &pos2, &identity2, NULL, &value2, &h2)) + { + if (h1 != h2) { + return 0; + } + cmp_identity = PyObject_RichCompareBool(identity1, identity2, Py_NE); + if (cmp_identity < 0) { + return -1; + } + cmp_value = PyObject_RichCompareBool(value1, value2, Py_NE); + if (cmp_value < 0) { + return -1; + } + if (cmp_identity || cmp_value) { + return 0; + } + } + + return 1; +} + +static inline int +_multidict_update_items(MultiDictObject *self, pair_list_t *pairs) +{ + return pair_list_update(&self->pairs, pairs); +} + +static inline int +_multidict_append_items(MultiDictObject *self, pair_list_t *pairs) +{ + PyObject *key = NULL, + *value = NULL; + + Py_ssize_t pos = 0; + + while (_pair_list_next(pairs, &pos, NULL, &key, &value, NULL)) { + if (pair_list_add(&self->pairs, key, value) < 0) { + return -1; + } + } + + return 0; +} + +static inline int +_multidict_append_items_seq(MultiDictObject *self, PyObject *arg, + const char *name) +{ + PyObject *key = NULL, + *value = NULL, + *item = NULL, + *iter = PyObject_GetIter(arg); + + if (iter == NULL) { + return -1; + } + + while ((item = PyIter_Next(iter)) != NULL) { + if (PyTuple_CheckExact(item)) { + if (PyTuple_GET_SIZE(item) != 2) { + goto invalid_type; + } + key = PyTuple_GET_ITEM(item, 0); + Py_INCREF(key); + value = PyTuple_GET_ITEM(item, 1); + Py_INCREF(value); + } + else if (PyList_CheckExact(item)) { + if (PyList_GET_SIZE(item) != 2) { + goto invalid_type; + } + key = PyList_GET_ITEM(item, 0); + Py_INCREF(key); + value = PyList_GET_ITEM(item, 1); + Py_INCREF(value); + } + else if (PySequence_Check(item)) { + if (PySequence_Size(item) != 2) { + goto invalid_type; + } + key = PySequence_GetItem(item, 0); + value = PySequence_GetItem(item, 1); + } else { + goto invalid_type; + } + + if (pair_list_add(&self->pairs, key, value) < 0) { + goto fail; + } + Py_CLEAR(key); + Py_CLEAR(value); + Py_CLEAR(item); + } + + Py_DECREF(iter); + + if (PyErr_Occurred()) { + return -1; + } + + return 0; +invalid_type: + PyErr_Format( + PyExc_TypeError, + "%s takes either dict or list of (key, value) pairs", + name, + NULL + ); + goto fail; +fail: + Py_XDECREF(key); + Py_XDECREF(value); + Py_XDECREF(item); + Py_DECREF(iter); + return -1; +} + +static inline int +_multidict_list_extend(PyObject *list, PyObject *target_list) +{ + PyObject *item = NULL, + *iter = PyObject_GetIter(target_list); + + if (iter == NULL) { + return -1; + } + + while ((item = PyIter_Next(iter)) != NULL) { + if (PyList_Append(list, item) < 0) { + Py_DECREF(item); + Py_DECREF(iter); + return -1; + } + Py_DECREF(item); + } + + Py_DECREF(iter); + + if (PyErr_Occurred()) { + return -1; + } + + return 0; +} + +static inline int +_multidict_extend_with_args(MultiDictObject *self, PyObject *arg, + PyObject *kwds, const char *name, int do_add) +{ + PyObject *arg_items = NULL, /* tracked by GC */ + *kwds_items = NULL; /* new reference */ + pair_list_t *pairs = NULL; + + int err = 0; + + if (kwds && !PyArg_ValidateKeywordArguments(kwds)) { + return -1; + } + + // TODO: mb can be refactored more clear + if (_MultiDict_Check(arg) && kwds == NULL) { + if (MultiDict_CheckExact(arg) || CIMultiDict_CheckExact(arg)) { + pairs = &((MultiDictObject*)arg)->pairs; + } else if (MultiDictProxy_CheckExact(arg) || CIMultiDictProxy_CheckExact(arg)) { + pairs = &((MultiDictProxyObject*)arg)->md->pairs; + } + + if (do_add) { + return _multidict_append_items(self, pairs); + } + + return _multidict_update_items(self, pairs); + } + + if (PyObject_HasAttrString(arg, "items")) { + if (_MultiDict_Check(arg)) { + arg_items = multidict_items((MultiDictObject*)arg); + } else { + arg_items = PyMapping_Items(arg); + } + if (arg_items == NULL) { + return -1; + } + } else { + arg_items = arg; + Py_INCREF(arg_items); + } + + if (kwds) { + PyObject *tmp = PySequence_List(arg_items); + Py_DECREF(arg_items); + arg_items = tmp; + if (arg_items == NULL) { + return -1; + } + + kwds_items = PyDict_Items(kwds); + if (kwds_items == NULL) { + Py_DECREF(arg_items); + return -1; + } + err = _multidict_list_extend(arg_items, kwds_items); + Py_DECREF(kwds_items); + if (err < 0) { + Py_DECREF(arg_items); + return -1; + } + } + + if (do_add) { + err = _multidict_append_items_seq(self, arg_items, name); + } else { + err = pair_list_update_from_seq(&self->pairs, arg_items); + } + + Py_DECREF(arg_items); + + return err; +} + +static inline int +_multidict_extend_with_kwds(MultiDictObject *self, PyObject *kwds, + const char *name, int do_add) +{ + PyObject *arg = NULL; + + int err = 0; + + if (!PyArg_ValidateKeywordArguments(kwds)) { + return -1; + } + + arg = PyDict_Items(kwds); + if (do_add) { + err = _multidict_append_items_seq(self, arg, name); + } else { + err = pair_list_update_from_seq(&self->pairs, arg); + } + + Py_DECREF(arg); + return err; +} + +static inline int +_multidict_extend(MultiDictObject *self, PyObject *args, PyObject *kwds, + const char *name, int do_add) +{ + PyObject *arg = NULL; + + if (args && PyObject_Length(args) > 1) { + PyErr_Format( + PyExc_TypeError, + "%s takes at most 1 positional argument (%zd given)", + name, PyObject_Length(args), NULL + ); + return -1; + } + + if (args && PyObject_Length(args) > 0) { + if (!PyArg_UnpackTuple(args, name, 0, 1, &arg)) { + return -1; + } + if (_multidict_extend_with_args(self, arg, kwds, name, do_add) < 0) { + return -1; + } + } else if (kwds && PyObject_Length(kwds) > 0) { + if (_multidict_extend_with_kwds(self, kwds, name, do_add) < 0) { + return -1; + } + } + + return 0; +} + +static inline PyObject * +_multidict_copy(MultiDictObject *self, PyTypeObject *multidict_tp_object) +{ + MultiDictObject *new_multidict = NULL; + + PyObject *arg_items = NULL, + *items = NULL; + + new_multidict = (MultiDictObject*)PyType_GenericNew( + multidict_tp_object, NULL, NULL); + if (new_multidict == NULL) { + return NULL; + } + + if (multidict_tp_object->tp_init( + (PyObject*)new_multidict, NULL, NULL) < 0) + { + return NULL; + } + + items = multidict_items(self); + if (items == NULL) { + goto fail; + } + + // TODO: "Implementation looks as slow as possible ..." + arg_items = PyTuple_New(1); + if (arg_items == NULL) { + goto fail; + } + + Py_INCREF(items); + PyTuple_SET_ITEM(arg_items, 0, items); + + if (_multidict_extend( + new_multidict, arg_items, NULL, "copy", 1) < 0) + { + goto fail; + } + + Py_DECREF(items); + Py_DECREF(arg_items); + + return (PyObject*)new_multidict; + +fail: + Py_XDECREF(items); + Py_XDECREF(arg_items); + + Py_DECREF(new_multidict); + + return NULL; +} + +static inline PyObject * +_multidict_proxy_copy(MultiDictProxyObject *self, PyTypeObject *type) +{ + PyObject *new_multidict = PyType_GenericNew(type, NULL, NULL); + if (new_multidict == NULL) { + goto fail; + } + if (type->tp_init(new_multidict, NULL, NULL) < 0) { + goto fail; + } + if (_multidict_extend_with_args( + (MultiDictObject*)new_multidict, (PyObject*)self, NULL, "copy", 1) < 0) + { + goto fail; + } + + return new_multidict; + +fail: + Py_XDECREF(new_multidict); + return NULL; +} + + +/******************** Base Methods ********************/ + +static inline PyObject * +multidict_getall(MultiDictObject *self, PyObject *const *args, + Py_ssize_t nargs, PyObject *kwnames) +{ + PyObject *list = NULL, + *key = NULL, + *_default = NULL; + + static const char * const _keywords[] = {"key", "default", NULL}; +#ifdef FASTCALL_OLD + static _PyArg_Parser _parser = {"O|O:getall", _keywords, 0}; + if (!_PyArg_ParseStackAndKeywords(args, nargs, kwnames, &_parser, + &key, &_default)) { + return NULL; + } +#else + static _PyArg_Parser _parser = {NULL, _keywords, "getall", 0}; + PyObject *argsbuf[2]; + Py_ssize_t noptargs = nargs + (kwnames ? PyTuple_GET_SIZE(kwnames) : 0) - 1; + args = _PyArg_UnpackKeywords(args, nargs, NULL, kwnames, + &_parser, 1, 2, 0, argsbuf); + if (!args) { + return NULL; + } + key = args[0]; + if (!noptargs) { + goto skip_optional_pos; + } + + _default = args[1]; +skip_optional_pos: +#endif + list = pair_list_get_all(&self->pairs, key); + + if (list == NULL && + PyErr_ExceptionMatches(PyExc_KeyError) && + _default != NULL) + { + PyErr_Clear(); + Py_INCREF(_default); + return _default; + } + + return list; +} + +static inline PyObject * +multidict_getone(MultiDictObject *self, PyObject *const *args, + Py_ssize_t nargs, PyObject *kwnames) +{ + PyObject *key = NULL, + *_default = NULL; + + static const char * const _keywords[] = {"key", "default", NULL}; +#ifdef FASTCALL_OLD + static _PyArg_Parser _parser = {"O|O:getone", _keywords, 0}; + if (!_PyArg_ParseStackAndKeywords(args, nargs, kwnames, &_parser, + &key, &_default)) { + return NULL; + } +#else + static _PyArg_Parser _parser = {NULL, _keywords, "getone", 0}; + PyObject *argsbuf[2]; + Py_ssize_t noptargs = nargs + (kwnames ? PyTuple_GET_SIZE(kwnames) : 0) - 1; + args = _PyArg_UnpackKeywords(args, nargs, NULL, kwnames, + &_parser, 1, 2, 0, argsbuf); + if (!args) { + return NULL; + } + key = args[0]; + if (!noptargs) { + goto skip_optional_pos; + } + + _default = args[1]; +skip_optional_pos: +#endif + return _multidict_getone(self, key, _default); +} + +static inline PyObject * +multidict_get(MultiDictObject *self, PyObject *const *args, + Py_ssize_t nargs, PyObject *kwnames) +{ + PyObject *key = NULL, + *_default = Py_None, + *ret; + + static const char * const _keywords[] = {"key", "default", NULL}; +#ifdef FASTCALL_OLD + static _PyArg_Parser _parser = {"O|O:get", _keywords, 0}; + if (!_PyArg_ParseStackAndKeywords(args, nargs, kwnames, &_parser, + &key, &_default)) { + return NULL; + } +#else + static _PyArg_Parser _parser = {NULL, _keywords, "get", 0}; + PyObject *argsbuf[2]; + Py_ssize_t noptargs = nargs + (kwnames ? PyTuple_GET_SIZE(kwnames) : 0) - 1; + args = _PyArg_UnpackKeywords(args, nargs, NULL, kwnames, + &_parser, 1, 2, 0, argsbuf); + if (!args) { + return NULL; + } + key = args[0]; + if (!noptargs) { + goto skip_optional_pos; + } + + _default = args[1]; +skip_optional_pos: +#endif + ret = _multidict_getone(self, key, _default); + return ret; +} + +static inline PyObject * +multidict_keys(MultiDictObject *self) +{ + return multidict_keysview_new((PyObject*)self); +} + +static inline PyObject * +multidict_items(MultiDictObject *self) +{ + return multidict_itemsview_new((PyObject*)self); +} + +static inline PyObject * +multidict_values(MultiDictObject *self) +{ + return multidict_valuesview_new((PyObject*)self); +} + +static inline PyObject * +multidict_reduce(MultiDictObject *self) +{ + PyObject *items = NULL, + *items_list = NULL, + *args = NULL, + *result = NULL; + + items = multidict_items(self); + if (items == NULL) { + goto ret; + } + + items_list = PySequence_List(items); + if (items_list == NULL) { + goto ret; + } + + args = PyTuple_Pack(1, items_list); + if (args == NULL) { + goto ret; + } + + result = PyTuple_Pack(2, Py_TYPE(self), args); + +ret: + Py_XDECREF(args); + Py_XDECREF(items_list); + Py_XDECREF(items); + + return result; +} + +static inline PyObject * +multidict_repr(PyObject *self) +{ + return PyObject_CallFunctionObjArgs( + repr_func, self, NULL); +} + +static inline Py_ssize_t +multidict_mp_len(MultiDictObject *self) +{ + return pair_list_len(&self->pairs); +} + +static inline PyObject * +multidict_mp_subscript(MultiDictObject *self, PyObject *key) +{ + return _multidict_getone(self, key, NULL); +} + +static inline int +multidict_mp_as_subscript(MultiDictObject *self, PyObject *key, PyObject *val) +{ + if (val == NULL) { + return pair_list_del(&self->pairs, key); + } else { + return pair_list_replace(&self->pairs, key, val); + } +} + +static inline int +multidict_sq_contains(MultiDictObject *self, PyObject *key) +{ + return pair_list_contains(&self->pairs, key); +} + +static inline PyObject * +multidict_tp_iter(MultiDictObject *self) +{ + return multidict_keys_iter_new(self); +} + +static inline PyObject * +multidict_tp_richcompare(PyObject *self, PyObject *other, int op) +{ + // TODO: refactoring me with love + + int cmp = 0; + + if (op != Py_EQ && op != Py_NE) { + Py_RETURN_NOTIMPLEMENTED; + } + + if (MultiDict_CheckExact(other) || CIMultiDict_CheckExact(other)) { + cmp = _multidict_eq( + (MultiDictObject*)self, + (MultiDictObject*)other + ); + if (cmp < 0) { + return NULL; + } + if (op == Py_NE) { + cmp = !cmp; + } + return PyBool_FromLong(cmp); + } + + if (MultiDictProxy_CheckExact(other) || CIMultiDictProxy_CheckExact(other)) { + cmp = _multidict_eq( + (MultiDictObject*)self, + ((MultiDictProxyObject*)other)->md + ); + if (cmp < 0) { + return NULL; + } + if (op == Py_NE) { + cmp = !cmp; + } + return PyBool_FromLong(cmp); + } + + cmp = PyObject_IsInstance(other, (PyObject*)collections_abc_mapping); + if (cmp < 0) { + return NULL; + } + + if (cmp) { + cmp = pair_list_eq_to_mapping(&((MultiDictObject*)self)->pairs, other); + if (cmp < 0) { + return NULL; + } + if (op == Py_NE) { + cmp = !cmp; + } + return PyBool_FromLong(cmp); + } + + Py_RETURN_NOTIMPLEMENTED; +} + +static inline void +multidict_tp_dealloc(MultiDictObject *self) +{ + PyObject_GC_UnTrack(self); + Py_TRASHCAN_SAFE_BEGIN(self); + if (self->weaklist != NULL) { + PyObject_ClearWeakRefs((PyObject *)self); + }; + pair_list_dealloc(&self->pairs); + Py_TYPE(self)->tp_free((PyObject *)self); + Py_TRASHCAN_SAFE_END(self); +} + +static inline int +multidict_tp_traverse(MultiDictObject *self, visitproc visit, void *arg) +{ + return pair_list_traverse(&self->pairs, visit, arg); +} + +static inline int +multidict_tp_clear(MultiDictObject *self) +{ + return pair_list_clear(&self->pairs); +} + +PyDoc_STRVAR(multidict_getall_doc, +"Return a list of all values matching the key."); + +PyDoc_STRVAR(multidict_getone_doc, +"Get first value matching the key."); + +PyDoc_STRVAR(multidict_get_doc, +"Get first value matching the key.\n\nThe method is alias for .getone()."); + +PyDoc_STRVAR(multidict_keys_doc, +"Return a new view of the dictionary's keys."); + +PyDoc_STRVAR(multidict_items_doc, +"Return a new view of the dictionary's items *(key, value) pairs)."); + +PyDoc_STRVAR(multidict_values_doc, +"Return a new view of the dictionary's values."); + +/******************** MultiDict ********************/ + +static inline int +multidict_tp_init(MultiDictObject *self, PyObject *args, PyObject *kwds) +{ + if (pair_list_init(&self->pairs) < 0) { + return -1; + } + if (_multidict_extend(self, args, kwds, "MultiDict", 1) < 0) { + return -1; + } + return 0; +} + +static inline PyObject * +multidict_add(MultiDictObject *self, PyObject *const *args, + Py_ssize_t nargs, PyObject *kwnames) +{ + PyObject *key = NULL, + *val = NULL; + + static const char * const _keywords[] = {"key", "value", NULL}; +#ifdef FASTCALL_OLD + static _PyArg_Parser _parser = {"OO:add", _keywords, 0}; + if (!_PyArg_ParseStackAndKeywords(args, nargs, kwnames, &_parser, + &key, &val)) { + return NULL; + } +#else + static _PyArg_Parser _parser = {NULL, _keywords, "add", 0}; + PyObject *argsbuf[2]; + + args = _PyArg_UnpackKeywords(args, nargs, NULL, kwnames, + &_parser, 2, 2, 0, argsbuf); + if (!args) { + return NULL; + } + key = args[0]; + val = args[1]; +#endif + if (pair_list_add(&self->pairs, key, val) < 0) { + return NULL; + } + + Py_RETURN_NONE; +} + +static inline PyObject * +multidict_copy(MultiDictObject *self) +{ + return _multidict_copy(self, &multidict_type); +} + +static inline PyObject * +multidict_extend(MultiDictObject *self, PyObject *args, PyObject *kwds) +{ + if (_multidict_extend(self, args, kwds, "extend", 1) < 0) { + return NULL; + } + + Py_RETURN_NONE; +} + +static inline PyObject * +multidict_clear(MultiDictObject *self) +{ + if (pair_list_clear(&self->pairs) < 0) { + return NULL; + } + + Py_RETURN_NONE; +} + +static inline PyObject * +multidict_setdefault(MultiDictObject *self, PyObject *const *args, + Py_ssize_t nargs, PyObject *kwnames) +{ + PyObject *key = NULL, + *_default = NULL; + + static const char * const _keywords[] = {"key", "default", NULL}; +#ifdef FASTCALL_OLD + static _PyArg_Parser _parser = {"O|O:setdefault", _keywords, 0}; + if (!_PyArg_ParseStackAndKeywords(args, nargs, kwnames, &_parser, + &key, &_default)) { + return NULL; + } +#else + static _PyArg_Parser _parser = {NULL, _keywords, "setdefault", 0}; + PyObject *argsbuf[3]; + Py_ssize_t noptargs = nargs + (kwnames ? PyTuple_GET_SIZE(kwnames) : 0) - 1; + + args = _PyArg_UnpackKeywords(args, nargs, NULL, kwnames, + &_parser, 1, 2, 0, argsbuf); + if (!args) { + return NULL; + } + key = args[0]; + if (!noptargs) { + goto skip_optional_pos; + } + _default = args[1]; + +skip_optional_pos: +#endif + return pair_list_set_default(&self->pairs, key, _default); +} + +static inline PyObject * +multidict_popone(MultiDictObject *self, PyObject *const *args, + Py_ssize_t nargs, PyObject *kwnames) +{ + PyObject *key = NULL, + *_default = NULL, + *ret_val = NULL; + + static const char * const _keywords[] = {"key", "default", NULL}; +#ifdef FASTCALL_OLD + static _PyArg_Parser _parser = {"O|O:popone", _keywords, 0}; + if (!_PyArg_ParseStackAndKeywords(args, nargs, kwnames, &_parser, + &key, &_default)) { + return NULL; + } +#else + static _PyArg_Parser _parser = {NULL, _keywords, "popone", 0}; + PyObject *argsbuf[3]; + Py_ssize_t noptargs = nargs + (kwnames ? PyTuple_GET_SIZE(kwnames) : 0) - 1; + + args = _PyArg_UnpackKeywords(args, nargs, NULL, kwnames, + &_parser, 1, 2, 0, argsbuf); + if (!args) { + return NULL; + } + key = args[0]; + if (!noptargs) { + goto skip_optional_pos; + } + _default = args[1]; + +skip_optional_pos: +#endif + ret_val = pair_list_pop_one(&self->pairs, key); + + if (ret_val == NULL && + PyErr_ExceptionMatches(PyExc_KeyError) && + _default != NULL) + { + PyErr_Clear(); + Py_INCREF(_default); + return _default; + } + + return ret_val; +} + +static inline PyObject * +multidict_pop(MultiDictObject *self, PyObject *const *args, + Py_ssize_t nargs, PyObject *kwnames) +{ + PyObject *key = NULL, + *_default = NULL, + *ret_val = NULL; + + static const char * const _keywords[] = {"key", "default", NULL}; +#ifdef FASTCALL_OLD + static _PyArg_Parser _parser = {"O|O:pop", _keywords, 0}; + if (!_PyArg_ParseStackAndKeywords(args, nargs, kwnames, &_parser, + &key, &_default)) { + return NULL; + } +#else + static _PyArg_Parser _parser = {NULL, _keywords, "pop", 0}; + PyObject *argsbuf[3]; + Py_ssize_t noptargs = nargs + (kwnames ? PyTuple_GET_SIZE(kwnames) : 0) - 1; + + args = _PyArg_UnpackKeywords(args, nargs, NULL, kwnames, + &_parser, 1, 2, 0, argsbuf); + if (!args) { + return NULL; + } + key = args[0]; + if (!noptargs) { + goto skip_optional_pos; + } + _default = args[1]; + +skip_optional_pos: +#endif + ret_val = pair_list_pop_one(&self->pairs, key); + + if (ret_val == NULL && + PyErr_ExceptionMatches(PyExc_KeyError) && + _default != NULL) + { + PyErr_Clear(); + Py_INCREF(_default); + return _default; + } + + return ret_val; +} + +static inline PyObject * +multidict_popall(MultiDictObject *self, PyObject *const *args, + Py_ssize_t nargs, PyObject *kwnames) +{ + PyObject *key = NULL, + *_default = NULL, + *ret_val = NULL; + + + static const char * const _keywords[] = {"key", "default", NULL}; +#ifdef FASTCALL_OLD + static _PyArg_Parser _parser = {"O|O:popall", _keywords, 0}; + if (!_PyArg_ParseStackAndKeywords(args, nargs, kwnames, &_parser, + &key, &_default)) { + return NULL; + } +#else + static _PyArg_Parser _parser = {NULL, _keywords, "popall", 0}; + PyObject *argsbuf[3]; + Py_ssize_t noptargs = nargs + (kwnames ? PyTuple_GET_SIZE(kwnames) : 0) - 1; + + args = _PyArg_UnpackKeywords(args, nargs, NULL, kwnames, + &_parser, 1, 2, 0, argsbuf); + if (!args) { + return NULL; + } + key = args[0]; + if (!noptargs) { + goto skip_optional_pos; + } + _default = args[1]; + +skip_optional_pos: +#endif + ret_val = pair_list_pop_all(&self->pairs, key); + + if (ret_val == NULL && + PyErr_ExceptionMatches(PyExc_KeyError) && + _default != NULL) + { + PyErr_Clear(); + Py_INCREF(_default); + return _default; + } + + return ret_val; +} + +static inline PyObject * +multidict_popitem(MultiDictObject *self) +{ + return pair_list_pop_item(&self->pairs); +} + +static inline PyObject * +multidict_update(MultiDictObject *self, PyObject *args, PyObject *kwds) +{ + if (_multidict_extend(self, args, kwds, "update", 0) < 0) { + return NULL; + } + Py_RETURN_NONE; +} + +PyDoc_STRVAR(multidict_add_doc, +"Add the key and value, not overwriting any previous value."); + +PyDoc_STRVAR(multidict_copy_doc, +"Return a copy of itself."); + +PyDoc_STRVAR(multdicit_method_extend_doc, +"Extend current MultiDict with more values.\n\ +This method must be used instead of update."); + +PyDoc_STRVAR(multidict_clear_doc, +"Remove all items from MultiDict"); + +PyDoc_STRVAR(multidict_setdefault_doc, +"Return value for key, set value to default if key is not present."); + +PyDoc_STRVAR(multidict_popone_doc, +"Remove the last occurrence of key and return the corresponding value.\n\n\ +If key is not found, default is returned if given, otherwise KeyError is \ +raised.\n"); + +PyDoc_STRVAR(multidict_pop_doc, +"Remove the last occurrence of key and return the corresponding value.\n\n\ +If key is not found, default is returned if given, otherwise KeyError is \ +raised.\n"); + +PyDoc_STRVAR(multidict_popall_doc, +"Remove all occurrences of key and return the list of corresponding values.\n\n\ +If key is not found, default is returned if given, otherwise KeyError is \ +raised.\n"); + +PyDoc_STRVAR(multidict_popitem_doc, +"Remove and return an arbitrary (key, value) pair."); + +PyDoc_STRVAR(multidict_update_doc, +"Update the dictionary from *other*, overwriting existing keys."); + + +#if PY_MAJOR_VERSION >= 3 && PY_MINOR_VERSION >= 9 +#define multidict_class_getitem Py_GenericAlias +#else +static inline PyObject * +multidict_class_getitem(PyObject *self, PyObject *arg) +{ + Py_INCREF(self); + return self; +} +#endif + + +PyDoc_STRVAR(sizeof__doc__, +"D.__sizeof__() -> size of D in memory, in bytes"); + +static inline PyObject * +_multidict_sizeof(MultiDictObject *self) +{ + Py_ssize_t size = sizeof(MultiDictObject); + if (self->pairs.pairs != self->pairs.buffer) { + size += (Py_ssize_t)sizeof(pair_t) * self->pairs.capacity; + } + return PyLong_FromSsize_t(size); +} + + +static PySequenceMethods multidict_sequence = { + .sq_contains = (objobjproc)multidict_sq_contains, +}; + +static PyMappingMethods multidict_mapping = { + .mp_length = (lenfunc)multidict_mp_len, + .mp_subscript = (binaryfunc)multidict_mp_subscript, + .mp_ass_subscript = (objobjargproc)multidict_mp_as_subscript, +}; + +static PyMethodDef multidict_methods[] = { + { + "getall", + (PyCFunction)multidict_getall, + METH_FASTCALL | METH_KEYWORDS, + multidict_getall_doc + }, + { + "getone", + (PyCFunction)multidict_getone, + METH_FASTCALL | METH_KEYWORDS, + multidict_getone_doc + }, + { + "get", + (PyCFunction)multidict_get, + METH_FASTCALL | METH_KEYWORDS, + multidict_get_doc + }, + { + "keys", + (PyCFunction)multidict_keys, + METH_NOARGS, + multidict_keys_doc + }, + { + "items", + (PyCFunction)multidict_items, + METH_NOARGS, + multidict_items_doc + }, + { + "values", + (PyCFunction)multidict_values, + METH_NOARGS, + multidict_values_doc + }, + { + "add", + (PyCFunction)multidict_add, + METH_FASTCALL | METH_KEYWORDS, + multidict_add_doc + }, + { + "copy", + (PyCFunction)multidict_copy, + METH_NOARGS, + multidict_copy_doc + }, + { + "extend", + (PyCFunction)multidict_extend, + METH_VARARGS | METH_KEYWORDS, + multdicit_method_extend_doc + }, + { + "clear", + (PyCFunction)multidict_clear, + METH_NOARGS, + multidict_clear_doc + }, + { + "setdefault", + (PyCFunction)multidict_setdefault, + METH_FASTCALL | METH_KEYWORDS, + multidict_setdefault_doc + }, + { + "popone", + (PyCFunction)multidict_popone, + METH_FASTCALL | METH_KEYWORDS, + multidict_popone_doc + }, + { + "pop", + (PyCFunction)multidict_pop, + METH_FASTCALL | METH_KEYWORDS, + multidict_pop_doc + }, + { + "popall", + (PyCFunction)multidict_popall, + METH_FASTCALL | METH_KEYWORDS, + multidict_popall_doc + }, + { + "popitem", + (PyCFunction)multidict_popitem, + METH_NOARGS, + multidict_popitem_doc + }, + { + "update", + (PyCFunction)multidict_update, + METH_VARARGS | METH_KEYWORDS, + multidict_update_doc + }, + { + "__reduce__", + (PyCFunction)multidict_reduce, + METH_NOARGS, + NULL, + }, + { + "__class_getitem__", + (PyCFunction)multidict_class_getitem, + METH_O | METH_CLASS, + NULL + }, + { + "__sizeof__", + (PyCFunction)_multidict_sizeof, + METH_NOARGS, + sizeof__doc__, + }, + { + NULL, + NULL + } /* sentinel */ +}; + + +PyDoc_STRVAR(MultDict_doc, +"Dictionary with the support for duplicate keys."); + + +static PyTypeObject multidict_type = { + PyVarObject_HEAD_INIT(NULL, 0) + "multidict._multidict.MultiDict", /* tp_name */ + sizeof(MultiDictObject), /* tp_basicsize */ + .tp_dealloc = (destructor)multidict_tp_dealloc, + .tp_repr = (reprfunc)multidict_repr, + .tp_as_sequence = &multidict_sequence, + .tp_as_mapping = &multidict_mapping, + .tp_flags = Py_TPFLAGS_DEFAULT | Py_TPFLAGS_BASETYPE | Py_TPFLAGS_HAVE_GC, + .tp_doc = MultDict_doc, + .tp_traverse = (traverseproc)multidict_tp_traverse, + .tp_clear = (inquiry)multidict_tp_clear, + .tp_richcompare = (richcmpfunc)multidict_tp_richcompare, + .tp_weaklistoffset = offsetof(MultiDictObject, weaklist), + .tp_iter = (getiterfunc)multidict_tp_iter, + .tp_methods = multidict_methods, + .tp_init = (initproc)multidict_tp_init, + .tp_alloc = PyType_GenericAlloc, + .tp_new = PyType_GenericNew, + .tp_free = PyObject_GC_Del, +}; + +/******************** CIMultiDict ********************/ + +static inline int +cimultidict_tp_init(MultiDictObject *self, PyObject *args, PyObject *kwds) +{ + if (ci_pair_list_init(&self->pairs) < 0) { + return -1; + } + if (_multidict_extend(self, args, kwds, "CIMultiDict", 1) < 0) { + return -1; + } + return 0; +} + +static inline PyObject * +cimultidict_copy(MultiDictObject *self) +{ + return _multidict_copy(self, &cimultidict_type); +} + +PyDoc_STRVAR(cimultidict_copy_doc, +"Return a copy of itself."); + +static PyMethodDef cimultidict_methods[] = { + { + "copy", + (PyCFunction)cimultidict_copy, + METH_NOARGS, + cimultidict_copy_doc + }, + { + NULL, + NULL + } /* sentinel */ +}; + +PyDoc_STRVAR(CIMultDict_doc, +"Dictionary with the support for duplicate case-insensitive keys."); + + +static PyTypeObject cimultidict_type = { + PyVarObject_HEAD_INIT(NULL, 0) + "multidict._multidict.CIMultiDict", /* tp_name */ + sizeof(MultiDictObject), /* tp_basicsize */ + .tp_dealloc = (destructor)multidict_tp_dealloc, + .tp_flags = Py_TPFLAGS_DEFAULT | Py_TPFLAGS_BASETYPE | Py_TPFLAGS_HAVE_GC, + .tp_doc = CIMultDict_doc, + .tp_traverse = (traverseproc)multidict_tp_traverse, + .tp_clear = (inquiry)multidict_tp_clear, + .tp_weaklistoffset = offsetof(MultiDictObject, weaklist), + .tp_methods = cimultidict_methods, + .tp_base = &multidict_type, + .tp_init = (initproc)cimultidict_tp_init, + .tp_alloc = PyType_GenericAlloc, + .tp_new = PyType_GenericNew, + .tp_free = PyObject_GC_Del, +}; + +/******************** MultiDictProxy ********************/ + +static inline int +multidict_proxy_tp_init(MultiDictProxyObject *self, PyObject *args, + PyObject *kwds) +{ + PyObject *arg = NULL; + MultiDictObject *md = NULL; + + if (!PyArg_UnpackTuple(args, "multidict._multidict.MultiDictProxy", + 0, 1, &arg)) + { + return -1; + } + if (arg == NULL) { + PyErr_Format( + PyExc_TypeError, + "__init__() missing 1 required positional argument: 'arg'" + ); + return -1; + } + if (!MultiDictProxy_CheckExact(arg) && + !CIMultiDict_CheckExact(arg) && + !MultiDict_CheckExact(arg)) + { + PyErr_Format( + PyExc_TypeError, + "ctor requires MultiDict or MultiDictProxy instance, " + "not <class '%s'>", + Py_TYPE(arg)->tp_name + ); + return -1; + } + + md = (MultiDictObject*)arg; + if (MultiDictProxy_CheckExact(arg)) { + md = ((MultiDictProxyObject*)arg)->md; + } + Py_INCREF(md); + self->md = md; + + return 0; +} + +static inline PyObject * +multidict_proxy_getall(MultiDictProxyObject *self, PyObject *const *args, + Py_ssize_t nargs, PyObject *kwnames) +{ + return multidict_getall(self->md, args, nargs, kwnames); +} + +static inline PyObject * +multidict_proxy_getone(MultiDictProxyObject *self, PyObject *const *args, + Py_ssize_t nargs, PyObject *kwnames) +{ + return multidict_getone(self->md, args, nargs, kwnames); +} + +static inline PyObject * +multidict_proxy_get(MultiDictProxyObject *self, PyObject *const *args, + Py_ssize_t nargs, PyObject *kwnames) +{ + return multidict_get(self->md, args, nargs, kwnames); +} + +static inline PyObject * +multidict_proxy_keys(MultiDictProxyObject *self) +{ + return multidict_keys(self->md); +} + +static inline PyObject * +multidict_proxy_items(MultiDictProxyObject *self) +{ + return multidict_items(self->md); +} + +static inline PyObject * +multidict_proxy_values(MultiDictProxyObject *self) +{ + return multidict_values(self->md); +} + +static inline PyObject * +multidict_proxy_copy(MultiDictProxyObject *self) +{ + return _multidict_proxy_copy(self, &multidict_type); +} + +static inline PyObject * +multidict_proxy_reduce(MultiDictProxyObject *self) +{ + PyErr_Format( + PyExc_TypeError, + "can't pickle %s objects", Py_TYPE(self)->tp_name + ); + + return NULL; +} + +static inline Py_ssize_t +multidict_proxy_mp_len(MultiDictProxyObject *self) +{ + return multidict_mp_len(self->md); +} + +static inline PyObject * +multidict_proxy_mp_subscript(MultiDictProxyObject *self, PyObject *key) +{ + return multidict_mp_subscript(self->md, key); +} + +static inline int +multidict_proxy_sq_contains(MultiDictProxyObject *self, PyObject *key) +{ + return multidict_sq_contains(self->md, key); +} + +static inline PyObject * +multidict_proxy_tp_iter(MultiDictProxyObject *self) +{ + return multidict_tp_iter(self->md); +} + +static inline PyObject * +multidict_proxy_tp_richcompare(MultiDictProxyObject *self, PyObject *other, + int op) +{ + return multidict_tp_richcompare((PyObject*)self->md, other, op); +} + +static inline void +multidict_proxy_tp_dealloc(MultiDictProxyObject *self) +{ + PyObject_GC_UnTrack(self); + if (self->weaklist != NULL) { + PyObject_ClearWeakRefs((PyObject *)self); + }; + Py_XDECREF(self->md); + Py_TYPE(self)->tp_free((PyObject *)self); +} + +static inline int +multidict_proxy_tp_traverse(MultiDictProxyObject *self, visitproc visit, + void *arg) +{ + Py_VISIT(self->md); + return 0; +} + +static inline int +multidict_proxy_tp_clear(MultiDictProxyObject *self) +{ + Py_CLEAR(self->md); + return 0; +} + +static PySequenceMethods multidict_proxy_sequence = { + .sq_contains = (objobjproc)multidict_proxy_sq_contains, +}; + +static PyMappingMethods multidict_proxy_mapping = { + .mp_length = (lenfunc)multidict_proxy_mp_len, + .mp_subscript = (binaryfunc)multidict_proxy_mp_subscript, +}; + +static PyMethodDef multidict_proxy_methods[] = { + { + "getall", + (PyCFunction)multidict_proxy_getall, + METH_FASTCALL | METH_KEYWORDS, + multidict_getall_doc + }, + { + "getone", + (PyCFunction)multidict_proxy_getone, + METH_FASTCALL | METH_KEYWORDS, + multidict_getone_doc + }, + { + "get", + (PyCFunction)multidict_proxy_get, + METH_FASTCALL | METH_KEYWORDS, + multidict_get_doc + }, + { + "keys", + (PyCFunction)multidict_proxy_keys, + METH_NOARGS, + multidict_keys_doc + }, + { + "items", + (PyCFunction)multidict_proxy_items, + METH_NOARGS, + multidict_items_doc + }, + { + "values", + (PyCFunction)multidict_proxy_values, + METH_NOARGS, + multidict_values_doc + }, + { + "copy", + (PyCFunction)multidict_proxy_copy, + METH_NOARGS, + multidict_copy_doc + }, + { + "__reduce__", + (PyCFunction)multidict_proxy_reduce, + METH_NOARGS, + NULL + }, + { + "__class_getitem__", + (PyCFunction)multidict_class_getitem, + METH_O | METH_CLASS, + NULL + }, + { + NULL, + NULL + } /* sentinel */ +}; + + +PyDoc_STRVAR(MultDictProxy_doc, +"Read-only proxy for MultiDict instance."); + + +static PyTypeObject multidict_proxy_type = { + PyVarObject_HEAD_INIT(NULL, 0) + "multidict._multidict.MultiDictProxy", /* tp_name */ + sizeof(MultiDictProxyObject), /* tp_basicsize */ + .tp_dealloc = (destructor)multidict_proxy_tp_dealloc, + .tp_repr = (reprfunc)multidict_repr, + .tp_as_sequence = &multidict_proxy_sequence, + .tp_as_mapping = &multidict_proxy_mapping, + .tp_flags = Py_TPFLAGS_DEFAULT | Py_TPFLAGS_BASETYPE | Py_TPFLAGS_HAVE_GC, + .tp_doc = MultDictProxy_doc, + .tp_traverse = (traverseproc)multidict_proxy_tp_traverse, + .tp_clear = (inquiry)multidict_proxy_tp_clear, + .tp_richcompare = (richcmpfunc)multidict_proxy_tp_richcompare, + .tp_weaklistoffset = offsetof(MultiDictProxyObject, weaklist), + .tp_iter = (getiterfunc)multidict_proxy_tp_iter, + .tp_methods = multidict_proxy_methods, + .tp_init = (initproc)multidict_proxy_tp_init, + .tp_alloc = PyType_GenericAlloc, + .tp_new = PyType_GenericNew, + .tp_free = PyObject_GC_Del, +}; + +/******************** CIMultiDictProxy ********************/ + +static inline int +cimultidict_proxy_tp_init(MultiDictProxyObject *self, PyObject *args, + PyObject *kwds) +{ + PyObject *arg = NULL; + MultiDictObject *md = NULL; + + if (!PyArg_UnpackTuple(args, "multidict._multidict.CIMultiDictProxy", + 1, 1, &arg)) + { + return -1; + } + if (arg == NULL) { + PyErr_Format( + PyExc_TypeError, + "__init__() missing 1 required positional argument: 'arg'" + ); + return -1; + } + if (!CIMultiDictProxy_CheckExact(arg) && !CIMultiDict_CheckExact(arg)) { + PyErr_Format( + PyExc_TypeError, + "ctor requires CIMultiDict or CIMultiDictProxy instance, " + "not <class '%s'>", + Py_TYPE(arg)->tp_name + ); + return -1; + } + + md = (MultiDictObject*)arg; + if (CIMultiDictProxy_CheckExact(arg)) { + md = ((MultiDictProxyObject*)arg)->md; + } + Py_INCREF(md); + self->md = md; + + return 0; +} + +static inline PyObject * +cimultidict_proxy_copy(MultiDictProxyObject *self) +{ + return _multidict_proxy_copy(self, &cimultidict_type); +} + + +PyDoc_STRVAR(CIMultDictProxy_doc, +"Read-only proxy for CIMultiDict instance."); + +PyDoc_STRVAR(cimultidict_proxy_copy_doc, +"Return copy of itself"); + +static PyMethodDef cimultidict_proxy_methods[] = { + { + "copy", + (PyCFunction)cimultidict_proxy_copy, + METH_NOARGS, + cimultidict_proxy_copy_doc + }, + { + NULL, + NULL + } /* sentinel */ +}; + +static PyTypeObject cimultidict_proxy_type = { + PyVarObject_HEAD_INIT(NULL, 0) + "multidict._multidict.CIMultiDictProxy", /* tp_name */ + sizeof(MultiDictProxyObject), /* tp_basicsize */ + .tp_dealloc = (destructor)multidict_proxy_tp_dealloc, + .tp_flags = Py_TPFLAGS_DEFAULT | Py_TPFLAGS_BASETYPE | Py_TPFLAGS_HAVE_GC, + .tp_doc = CIMultDictProxy_doc, + .tp_traverse = (traverseproc)multidict_proxy_tp_traverse, + .tp_clear = (inquiry)multidict_proxy_tp_clear, + .tp_richcompare = (richcmpfunc)multidict_proxy_tp_richcompare, + .tp_weaklistoffset = offsetof(MultiDictProxyObject, weaklist), + .tp_methods = cimultidict_proxy_methods, + .tp_base = &multidict_proxy_type, + .tp_init = (initproc)cimultidict_proxy_tp_init, + .tp_alloc = PyType_GenericAlloc, + .tp_new = PyType_GenericNew, + .tp_free = PyObject_GC_Del, +}; + +/******************** Other functions ********************/ + +static inline PyObject * +getversion(PyObject *self, PyObject *md) +{ + pair_list_t *pairs = NULL; + if (MultiDict_CheckExact(md) || CIMultiDict_CheckExact(md)) { + pairs = &((MultiDictObject*)md)->pairs; + } else if (MultiDictProxy_CheckExact(md) || CIMultiDictProxy_CheckExact(md)) { + pairs = &((MultiDictProxyObject*)md)->md->pairs; + } else { + PyErr_Format(PyExc_TypeError, "unexpected type"); + return NULL; + } + return PyLong_FromUnsignedLong(pair_list_version(pairs)); +} + +/******************** Module ********************/ + +static inline void +module_free(void *m) +{ + Py_CLEAR(collections_abc_mapping); + Py_CLEAR(collections_abc_mut_mapping); + Py_CLEAR(collections_abc_mut_multi_mapping); +} + +static PyMethodDef multidict_module_methods[] = { + { + "getversion", + (PyCFunction)getversion, + METH_O + }, + { + NULL, + NULL + } /* sentinel */ +}; + +static PyModuleDef multidict_module = { + PyModuleDef_HEAD_INIT, /* m_base */ + "_multidict", /* m_name */ + .m_size = -1, + .m_methods = multidict_module_methods, + .m_free = (freefunc)module_free, +}; + +PyMODINIT_FUNC +PyInit__multidict() +{ + PyObject *module = NULL, + *reg_func_call_result = NULL; + +#define WITH_MOD(NAME) \ + Py_CLEAR(module); \ + module = PyImport_ImportModule(NAME); \ + if (module == NULL) { \ + goto fail; \ + } + +#define GET_MOD_ATTR(VAR, NAME) \ + VAR = PyObject_GetAttrString(module, NAME); \ + if (VAR == NULL) { \ + goto fail; \ + } + + if (multidict_views_init() < 0) { + goto fail; + } + + if (multidict_iter_init() < 0) { + goto fail; + } + + if (istr_init() < 0) { + goto fail; + } + + if (PyType_Ready(&multidict_type) < 0 || + PyType_Ready(&cimultidict_type) < 0 || + PyType_Ready(&multidict_proxy_type) < 0 || + PyType_Ready(&cimultidict_proxy_type) < 0) + { + goto fail; + } + + WITH_MOD("collections.abc"); + GET_MOD_ATTR(collections_abc_mapping, "Mapping"); + + WITH_MOD("multidict._abc"); + GET_MOD_ATTR(collections_abc_mut_mapping, "MultiMapping"); + + WITH_MOD("multidict._abc"); + GET_MOD_ATTR(collections_abc_mut_multi_mapping, "MutableMultiMapping"); + + WITH_MOD("multidict._multidict_base"); + GET_MOD_ATTR(repr_func, "_mdrepr"); + + /* Register in _abc mappings (CI)MultiDict and (CI)MultiDictProxy */ + reg_func_call_result = PyObject_CallMethod( + collections_abc_mut_mapping, + "register", "O", + (PyObject*)&multidict_proxy_type + ); + if (reg_func_call_result == NULL) { + goto fail; + } + Py_DECREF(reg_func_call_result); + + reg_func_call_result = PyObject_CallMethod( + collections_abc_mut_mapping, + "register", "O", + (PyObject*)&cimultidict_proxy_type + ); + if (reg_func_call_result == NULL) { + goto fail; + } + Py_DECREF(reg_func_call_result); + + reg_func_call_result = PyObject_CallMethod( + collections_abc_mut_multi_mapping, + "register", "O", + (PyObject*)&multidict_type + ); + if (reg_func_call_result == NULL) { + goto fail; + } + Py_DECREF(reg_func_call_result); + + reg_func_call_result = PyObject_CallMethod( + collections_abc_mut_multi_mapping, + "register", "O", + (PyObject*)&cimultidict_type + ); + if (reg_func_call_result == NULL) { + goto fail; + } + Py_DECREF(reg_func_call_result); + + /* Instantiate this module */ + module = PyModule_Create(&multidict_module); + + Py_INCREF(&istr_type); + if (PyModule_AddObject( + module, "istr", (PyObject*)&istr_type) < 0) + { + goto fail; + } + + Py_INCREF(&multidict_type); + if (PyModule_AddObject( + module, "MultiDict", (PyObject*)&multidict_type) < 0) + { + goto fail; + } + + Py_INCREF(&cimultidict_type); + if (PyModule_AddObject( + module, "CIMultiDict", (PyObject*)&cimultidict_type) < 0) + { + goto fail; + } + + Py_INCREF(&multidict_proxy_type); + if (PyModule_AddObject( + module, "MultiDictProxy", (PyObject*)&multidict_proxy_type) < 0) + { + goto fail; + } + + Py_INCREF(&cimultidict_proxy_type); + if (PyModule_AddObject( + module, "CIMultiDictProxy", (PyObject*)&cimultidict_proxy_type) < 0) + { + goto fail; + } + + return module; + +fail: + Py_XDECREF(collections_abc_mapping); + Py_XDECREF(collections_abc_mut_mapping); + Py_XDECREF(collections_abc_mut_multi_mapping); + + return NULL; + +#undef WITH_MOD +#undef GET_MOD_ATTR +} diff --git a/contrib/python/multidict/multidict/_multidict_base.py b/contrib/python/multidict/multidict/_multidict_base.py new file mode 100644 index 0000000000..394466548c --- /dev/null +++ b/contrib/python/multidict/multidict/_multidict_base.py @@ -0,0 +1,144 @@ +from collections.abc import ItemsView, Iterable, KeysView, Set, ValuesView + + +def _abc_itemsview_register(view_cls): + ItemsView.register(view_cls) + + +def _abc_keysview_register(view_cls): + KeysView.register(view_cls) + + +def _abc_valuesview_register(view_cls): + ValuesView.register(view_cls) + + +def _viewbaseset_richcmp(view, other, op): + if op == 0: # < + if not isinstance(other, Set): + return NotImplemented + return len(view) < len(other) and view <= other + elif op == 1: # <= + if not isinstance(other, Set): + return NotImplemented + if len(view) > len(other): + return False + for elem in view: + if elem not in other: + return False + return True + elif op == 2: # == + if not isinstance(other, Set): + return NotImplemented + return len(view) == len(other) and view <= other + elif op == 3: # != + return not view == other + elif op == 4: # > + if not isinstance(other, Set): + return NotImplemented + return len(view) > len(other) and view >= other + elif op == 5: # >= + if not isinstance(other, Set): + return NotImplemented + if len(view) < len(other): + return False + for elem in other: + if elem not in view: + return False + return True + + +def _viewbaseset_and(view, other): + if not isinstance(other, Iterable): + return NotImplemented + if isinstance(view, Set): + view = set(iter(view)) + if isinstance(other, Set): + other = set(iter(other)) + if not isinstance(other, Set): + other = set(iter(other)) + return view & other + + +def _viewbaseset_or(view, other): + if not isinstance(other, Iterable): + return NotImplemented + if isinstance(view, Set): + view = set(iter(view)) + if isinstance(other, Set): + other = set(iter(other)) + if not isinstance(other, Set): + other = set(iter(other)) + return view | other + + +def _viewbaseset_sub(view, other): + if not isinstance(other, Iterable): + return NotImplemented + if isinstance(view, Set): + view = set(iter(view)) + if isinstance(other, Set): + other = set(iter(other)) + if not isinstance(other, Set): + other = set(iter(other)) + return view - other + + +def _viewbaseset_xor(view, other): + if not isinstance(other, Iterable): + return NotImplemented + if isinstance(view, Set): + view = set(iter(view)) + if isinstance(other, Set): + other = set(iter(other)) + if not isinstance(other, Set): + other = set(iter(other)) + return view ^ other + + +def _itemsview_isdisjoint(view, other): + "Return True if two sets have a null intersection." + for v in other: + if v in view: + return False + return True + + +def _itemsview_repr(view): + lst = [] + for k, v in view: + lst.append("{!r}: {!r}".format(k, v)) + body = ", ".join(lst) + return "{}({})".format(view.__class__.__name__, body) + + +def _keysview_isdisjoint(view, other): + "Return True if two sets have a null intersection." + for k in other: + if k in view: + return False + return True + + +def _keysview_repr(view): + lst = [] + for k in view: + lst.append("{!r}".format(k)) + body = ", ".join(lst) + return "{}({})".format(view.__class__.__name__, body) + + +def _valuesview_repr(view): + lst = [] + for v in view: + lst.append("{!r}".format(v)) + body = ", ".join(lst) + return "{}({})".format(view.__class__.__name__, body) + + +def _mdrepr(md): + lst = [] + for k, v in md.items(): + lst.append("'{}': {!r}".format(k, v)) + body = ", ".join(lst) + return "<{}({})>".format(md.__class__.__name__, body) diff --git a/contrib/python/multidict/multidict/_multidict_py.py b/contrib/python/multidict/multidict/_multidict_py.py new file mode 100644 index 0000000000..cdbc328903 --- /dev/null +++ b/contrib/python/multidict/multidict/_multidict_py.py @@ -0,0 +1,526 @@ +import sys +import types +from array import array +from collections import abc + +from ._abc import MultiMapping, MutableMultiMapping + +_marker = object() + +if sys.version_info >= (3, 9): + GenericAlias = types.GenericAlias +else: + def GenericAlias(cls): + return cls + + +class istr(str): + + """Case insensitive str.""" + + __is_istr__ = True + + +upstr = istr # for relaxing backward compatibility problems + + +def getversion(md): + if not isinstance(md, _Base): + raise TypeError("Parameter should be multidict or proxy") + return md._impl._version + + +_version = array("Q", [0]) + + +class _Impl: + __slots__ = ("_items", "_version") + + def __init__(self): + self._items = [] + self.incr_version() + + def incr_version(self): + global _version + v = _version + v[0] += 1 + self._version = v[0] + + if sys.implementation.name != "pypy": + + def __sizeof__(self): + return object.__sizeof__(self) + sys.getsizeof(self._items) + + +class _Base: + def _title(self, key): + return key + + def getall(self, key, default=_marker): + """Return a list of all values matching the key.""" + identity = self._title(key) + res = [v for i, k, v in self._impl._items if i == identity] + if res: + return res + if not res and default is not _marker: + return default + raise KeyError("Key not found: %r" % key) + + def getone(self, key, default=_marker): + """Get first value matching the key. + + Raises KeyError if the key is not found and no default is provided. + """ + identity = self._title(key) + for i, k, v in self._impl._items: + if i == identity: + return v + if default is not _marker: + return default + raise KeyError("Key not found: %r" % key) + + # Mapping interface # + + def __getitem__(self, key): + return self.getone(key) + + def get(self, key, default=None): + """Get first value matching the key. + + If the key is not found, returns the default (or None if no default is provided) + """ + return self.getone(key, default) + + def __iter__(self): + return iter(self.keys()) + + def __len__(self): + return len(self._impl._items) + + def keys(self): + """Return a new view of the dictionary's keys.""" + return _KeysView(self._impl) + + def items(self): + """Return a new view of the dictionary's items *(key, value) pairs).""" + return _ItemsView(self._impl) + + def values(self): + """Return a new view of the dictionary's values.""" + return _ValuesView(self._impl) + + def __eq__(self, other): + if not isinstance(other, abc.Mapping): + return NotImplemented + if isinstance(other, _Base): + lft = self._impl._items + rht = other._impl._items + if len(lft) != len(rht): + return False + for (i1, k2, v1), (i2, k2, v2) in zip(lft, rht): + if i1 != i2 or v1 != v2: + return False + return True + if len(self._impl._items) != len(other): + return False + for k, v in self.items(): + nv = other.get(k, _marker) + if v != nv: + return False + return True + + def __contains__(self, key): + identity = self._title(key) + for i, k, v in self._impl._items: + if i == identity: + return True + return False + + def __repr__(self): + body = ", ".join("'{}': {!r}".format(k, v) for k, v in self.items()) + return "<{}({})>".format(self.__class__.__name__, body) + + __class_getitem__ = classmethod(GenericAlias) + + +class MultiDictProxy(_Base, MultiMapping): + """Read-only proxy for MultiDict instance.""" + + def __init__(self, arg): + if not isinstance(arg, (MultiDict, MultiDictProxy)): + raise TypeError( + "ctor requires MultiDict or MultiDictProxy instance" + ", not {}".format(type(arg)) + ) + + self._impl = arg._impl + + def __reduce__(self): + raise TypeError("can't pickle {} objects".format(self.__class__.__name__)) + + def copy(self): + """Return a copy of itself.""" + return MultiDict(self.items()) + + +class CIMultiDictProxy(MultiDictProxy): + """Read-only proxy for CIMultiDict instance.""" + + def __init__(self, arg): + if not isinstance(arg, (CIMultiDict, CIMultiDictProxy)): + raise TypeError( + "ctor requires CIMultiDict or CIMultiDictProxy instance" + ", not {}".format(type(arg)) + ) + + self._impl = arg._impl + + def _title(self, key): + return key.title() + + def copy(self): + """Return a copy of itself.""" + return CIMultiDict(self.items()) + + +class MultiDict(_Base, MutableMultiMapping): + """Dictionary with the support for duplicate keys.""" + + def __init__(self, *args, **kwargs): + self._impl = _Impl() + + self._extend(args, kwargs, self.__class__.__name__, self._extend_items) + + if sys.implementation.name != "pypy": + + def __sizeof__(self): + return object.__sizeof__(self) + sys.getsizeof(self._impl) + + def __reduce__(self): + return (self.__class__, (list(self.items()),)) + + def _title(self, key): + return key + + def _key(self, key): + if isinstance(key, str): + return key + else: + raise TypeError( + "MultiDict keys should be either str " "or subclasses of str" + ) + + def add(self, key, value): + identity = self._title(key) + self._impl._items.append((identity, self._key(key), value)) + self._impl.incr_version() + + def copy(self): + """Return a copy of itself.""" + cls = self.__class__ + return cls(self.items()) + + __copy__ = copy + + def extend(self, *args, **kwargs): + """Extend current MultiDict with more values. + + This method must be used instead of update. + """ + self._extend(args, kwargs, "extend", self._extend_items) + + def _extend(self, args, kwargs, name, method): + if len(args) > 1: + raise TypeError( + "{} takes at most 1 positional argument" + " ({} given)".format(name, len(args)) + ) + if args: + arg = args[0] + if isinstance(args[0], (MultiDict, MultiDictProxy)) and not kwargs: + items = arg._impl._items + else: + if hasattr(arg, "items"): + arg = arg.items() + if kwargs: + arg = list(arg) + arg.extend(list(kwargs.items())) + items = [] + for item in arg: + if not len(item) == 2: + raise TypeError( + "{} takes either dict or list of (key, value) " + "tuples".format(name) + ) + items.append((self._title(item[0]), self._key(item[0]), item[1])) + + method(items) + else: + method( + [ + (self._title(key), self._key(key), value) + for key, value in kwargs.items() + ] + ) + + def _extend_items(self, items): + for identity, key, value in items: + self.add(key, value) + + def clear(self): + """Remove all items from MultiDict.""" + self._impl._items.clear() + self._impl.incr_version() + + # Mapping interface # + + def __setitem__(self, key, value): + self._replace(key, value) + + def __delitem__(self, key): + identity = self._title(key) + items = self._impl._items + found = False + for i in range(len(items) - 1, -1, -1): + if items[i][0] == identity: + del items[i] + found = True + if not found: + raise KeyError(key) + else: + self._impl.incr_version() + + def setdefault(self, key, default=None): + """Return value for key, set value to default if key is not present.""" + identity = self._title(key) + for i, k, v in self._impl._items: + if i == identity: + return v + self.add(key, default) + return default + + def popone(self, key, default=_marker): + """Remove specified key and return the corresponding value. + + If key is not found, d is returned if given, otherwise + KeyError is raised. + + """ + identity = self._title(key) + for i in range(len(self._impl._items)): + if self._impl._items[i][0] == identity: + value = self._impl._items[i][2] + del self._impl._items[i] + self._impl.incr_version() + return value + if default is _marker: + raise KeyError(key) + else: + return default + + pop = popone # type: ignore + + def popall(self, key, default=_marker): + """Remove all occurrences of key and return the list of corresponding + values. + + If key is not found, default is returned if given, otherwise + KeyError is raised. + + """ + found = False + identity = self._title(key) + ret = [] + for i in range(len(self._impl._items) - 1, -1, -1): + item = self._impl._items[i] + if item[0] == identity: + ret.append(item[2]) + del self._impl._items[i] + self._impl.incr_version() + found = True + if not found: + if default is _marker: + raise KeyError(key) + else: + return default + else: + ret.reverse() + return ret + + def popitem(self): + """Remove and return an arbitrary (key, value) pair.""" + if self._impl._items: + i = self._impl._items.pop(0) + self._impl.incr_version() + return i[1], i[2] + else: + raise KeyError("empty multidict") + + def update(self, *args, **kwargs): + """Update the dictionary from *other*, overwriting existing keys.""" + self._extend(args, kwargs, "update", self._update_items) + + def _update_items(self, items): + if not items: + return + used_keys = {} + for identity, key, value in items: + start = used_keys.get(identity, 0) + for i in range(start, len(self._impl._items)): + item = self._impl._items[i] + if item[0] == identity: + used_keys[identity] = i + 1 + self._impl._items[i] = (identity, key, value) + break + else: + self._impl._items.append((identity, key, value)) + used_keys[identity] = len(self._impl._items) + + # drop tails + i = 0 + while i < len(self._impl._items): + item = self._impl._items[i] + identity = item[0] + pos = used_keys.get(identity) + if pos is None: + i += 1 + continue + if i >= pos: + del self._impl._items[i] + else: + i += 1 + + self._impl.incr_version() + + def _replace(self, key, value): + key = self._key(key) + identity = self._title(key) + items = self._impl._items + + for i in range(len(items)): + item = items[i] + if item[0] == identity: + items[i] = (identity, key, value) + # i points to last found item + rgt = i + self._impl.incr_version() + break + else: + self._impl._items.append((identity, key, value)) + self._impl.incr_version() + return + + # remove all tail items + i = rgt + 1 + while i < len(items): + item = items[i] + if item[0] == identity: + del items[i] + else: + i += 1 + + +class CIMultiDict(MultiDict): + """Dictionary with the support for duplicate case-insensitive keys.""" + + def _title(self, key): + return key.title() + + +class _Iter: + __slots__ = ("_size", "_iter") + + def __init__(self, size, iterator): + self._size = size + self._iter = iterator + + def __iter__(self): + return self + + def __next__(self): + return next(self._iter) + + def __length_hint__(self): + return self._size + + +class _ViewBase: + def __init__(self, impl): + self._impl = impl + + def __len__(self): + return len(self._impl._items) + + +class _ItemsView(_ViewBase, abc.ItemsView): + def __contains__(self, item): + assert isinstance(item, tuple) or isinstance(item, list) + assert len(item) == 2 + for i, k, v in self._impl._items: + if item[0] == k and item[1] == v: + return True + return False + + def __iter__(self): + return _Iter(len(self), self._iter(self._impl._version)) + + def _iter(self, version): + for i, k, v in self._impl._items: + if version != self._impl._version: + raise RuntimeError("Dictionary changed during iteration") + yield k, v + + def __repr__(self): + lst = [] + for item in self._impl._items: + lst.append("{!r}: {!r}".format(item[1], item[2])) + body = ", ".join(lst) + return "{}({})".format(self.__class__.__name__, body) + + +class _ValuesView(_ViewBase, abc.ValuesView): + def __contains__(self, value): + for item in self._impl._items: + if item[2] == value: + return True + return False + + def __iter__(self): + return _Iter(len(self), self._iter(self._impl._version)) + + def _iter(self, version): + for item in self._impl._items: + if version != self._impl._version: + raise RuntimeError("Dictionary changed during iteration") + yield item[2] + + def __repr__(self): + lst = [] + for item in self._impl._items: + lst.append("{!r}".format(item[2])) + body = ", ".join(lst) + return "{}({})".format(self.__class__.__name__, body) + + +class _KeysView(_ViewBase, abc.KeysView): + def __contains__(self, key): + for item in self._impl._items: + if item[1] == key: + return True + return False + + def __iter__(self): + return _Iter(len(self), self._iter(self._impl._version)) + + def _iter(self, version): + for item in self._impl._items: + if version != self._impl._version: + raise RuntimeError("Dictionary changed during iteration") + yield item[1] + + def __repr__(self): + lst = [] + for item in self._impl._items: + lst.append("{!r}".format(item[1])) + body = ", ".join(lst) + return "{}({})".format(self.__class__.__name__, body) diff --git a/contrib/python/multidict/multidict/_multilib/defs.h b/contrib/python/multidict/multidict/_multilib/defs.h new file mode 100644 index 0000000000..c7027c817e --- /dev/null +++ b/contrib/python/multidict/multidict/_multilib/defs.h @@ -0,0 +1,22 @@ +#ifndef _MULTIDICT_DEFS_H +#define _MULTIDICT_DEFS_H + +#ifdef __cplusplus +extern "C" { +#endif + +_Py_IDENTIFIER(lower); + +/* We link this module statically for convenience. If compiled as a shared + library instead, some compilers don't allow addresses of Python objects + defined in other libraries to be used in static initializers here. The + DEFERRED_ADDRESS macro is used to tag the slots where such addresses + appear; the module init function must fill in the tagged slots at runtime. + The argument is for documentation -- the macro ignores it. +*/ +#define DEFERRED_ADDRESS(ADDR) 0 + +#ifdef __cplusplus +} +#endif +#endif diff --git a/contrib/python/multidict/multidict/_multilib/dict.h b/contrib/python/multidict/multidict/_multilib/dict.h new file mode 100644 index 0000000000..3caf83e5b4 --- /dev/null +++ b/contrib/python/multidict/multidict/_multilib/dict.h @@ -0,0 +1,24 @@ +#ifndef _MULTIDICT_C_H +#define _MULTIDICT_C_H + +#ifdef __cplusplus +extern "C" { +#endif + +typedef struct { // 16 or 24 for GC prefix + PyObject_HEAD // 16 + PyObject *weaklist; + pair_list_t pairs; +} MultiDictObject; + +typedef struct { + PyObject_HEAD + PyObject *weaklist; + MultiDictObject *md; +} MultiDictProxyObject; + +#ifdef __cplusplus +} +#endif + +#endif diff --git a/contrib/python/multidict/multidict/_multilib/istr.h b/contrib/python/multidict/multidict/_multilib/istr.h new file mode 100644 index 0000000000..2688f48914 --- /dev/null +++ b/contrib/python/multidict/multidict/_multilib/istr.h @@ -0,0 +1,85 @@ +#ifndef _MULTIDICT_ISTR_H +#define _MULTIDICT_ISTR_H + +#ifdef __cplusplus +extern "C" { +#endif + +typedef struct { + PyUnicodeObject str; + PyObject * canonical; +} istrobject; + +PyDoc_STRVAR(istr__doc__, "istr class implementation"); + +static PyTypeObject istr_type; + +static inline void +istr_dealloc(istrobject *self) +{ + Py_XDECREF(self->canonical); + PyUnicode_Type.tp_dealloc((PyObject*)self); +} + +static inline PyObject * +istr_new(PyTypeObject *type, PyObject *args, PyObject *kwds) +{ + PyObject *x = NULL; + static char *kwlist[] = {"object", "encoding", "errors", 0}; + PyObject *encoding = NULL; + PyObject *errors = NULL; + PyObject *s = NULL; + PyObject * ret = NULL; + + if (!PyArg_ParseTupleAndKeywords(args, kwds, "|OOO:str", + kwlist, &x, &encoding, &errors)) { + return NULL; + } + if (x != NULL && Py_TYPE(x) == &istr_type) { + Py_INCREF(x); + return x; + } + ret = PyUnicode_Type.tp_new(type, args, kwds); + if (!ret) { + goto fail; + } + s =_PyObject_CallMethodId(ret, &PyId_lower, NULL); + if (!s) { + goto fail; + } + ((istrobject*)ret)->canonical = s; + s = NULL; /* the reference is stollen by .canonical */ + return ret; +fail: + Py_XDECREF(ret); + return NULL; +} + +static PyTypeObject istr_type = { + PyVarObject_HEAD_INIT(DEFERRED_ADDRESS(&PyType_Type), 0) + "multidict._multidict.istr", + sizeof(istrobject), + .tp_dealloc = (destructor)istr_dealloc, + .tp_flags = Py_TPFLAGS_DEFAULT + | Py_TPFLAGS_BASETYPE + | Py_TPFLAGS_UNICODE_SUBCLASS, + .tp_doc = istr__doc__, + .tp_base = DEFERRED_ADDRESS(&PyUnicode_Type), + .tp_new = (newfunc)istr_new, +}; + + +static inline int +istr_init(void) +{ + istr_type.tp_base = &PyUnicode_Type; + if (PyType_Ready(&istr_type) < 0) { + return -1; + } + return 0; +} + +#ifdef __cplusplus +} +#endif +#endif diff --git a/contrib/python/multidict/multidict/_multilib/iter.h b/contrib/python/multidict/multidict/_multilib/iter.h new file mode 100644 index 0000000000..4e2e32b387 --- /dev/null +++ b/contrib/python/multidict/multidict/_multilib/iter.h @@ -0,0 +1,238 @@ +#ifndef _MULTIDICT_ITER_H +#define _MULTIDICT_ITER_H + +#ifdef __cplusplus +extern "C" { +#endif + +static PyTypeObject multidict_items_iter_type; +static PyTypeObject multidict_values_iter_type; +static PyTypeObject multidict_keys_iter_type; + +typedef struct multidict_iter { + PyObject_HEAD + MultiDictObject *md; // MultiDict or CIMultiDict + Py_ssize_t current; + uint64_t version; +} MultidictIter; + +static inline void +_init_iter(MultidictIter *it, MultiDictObject *md) +{ + Py_INCREF(md); + + it->md = md; + it->current = 0; + it->version = pair_list_version(&md->pairs); +} + +static inline PyObject * +multidict_items_iter_new(MultiDictObject *md) +{ + MultidictIter *it = PyObject_GC_New( + MultidictIter, &multidict_items_iter_type); + if (it == NULL) { + return NULL; + } + + _init_iter(it, md); + + PyObject_GC_Track(it); + return (PyObject *)it; +} + +static inline PyObject * +multidict_keys_iter_new(MultiDictObject *md) +{ + MultidictIter *it = PyObject_GC_New( + MultidictIter, &multidict_keys_iter_type); + if (it == NULL) { + return NULL; + } + + _init_iter(it, md); + + PyObject_GC_Track(it); + return (PyObject *)it; +} + +static inline PyObject * +multidict_values_iter_new(MultiDictObject *md) +{ + MultidictIter *it = PyObject_GC_New( + MultidictIter, &multidict_values_iter_type); + if (it == NULL) { + return NULL; + } + + _init_iter(it, md); + + PyObject_GC_Track(it); + return (PyObject *)it; +} + +static inline PyObject * +multidict_items_iter_iternext(MultidictIter *self) +{ + PyObject *key = NULL; + PyObject *value = NULL; + PyObject *ret = NULL; + + if (self->version != pair_list_version(&self->md->pairs)) { + PyErr_SetString(PyExc_RuntimeError, "Dictionary changed during iteration"); + return NULL; + } + + if (!_pair_list_next(&self->md->pairs, &self->current, NULL, &key, &value, NULL)) { + PyErr_SetNone(PyExc_StopIteration); + return NULL; + } + + ret = PyTuple_Pack(2, key, value); + if (ret == NULL) { + return NULL; + } + + return ret; +} + +static inline PyObject * +multidict_values_iter_iternext(MultidictIter *self) +{ + PyObject *value = NULL; + + if (self->version != pair_list_version(&self->md->pairs)) { + PyErr_SetString(PyExc_RuntimeError, "Dictionary changed during iteration"); + return NULL; + } + + if (!pair_list_next(&self->md->pairs, &self->current, NULL, NULL, &value)) { + PyErr_SetNone(PyExc_StopIteration); + return NULL; + } + + Py_INCREF(value); + + return value; +} + +static inline PyObject * +multidict_keys_iter_iternext(MultidictIter *self) +{ + PyObject *key = NULL; + + if (self->version != pair_list_version(&self->md->pairs)) { + PyErr_SetString(PyExc_RuntimeError, "Dictionary changed during iteration"); + return NULL; + } + + if (!pair_list_next(&self->md->pairs, &self->current, NULL, &key, NULL)) { + PyErr_SetNone(PyExc_StopIteration); + return NULL; + } + + Py_INCREF(key); + + return key; +} + +static inline void +multidict_iter_dealloc(MultidictIter *self) +{ + PyObject_GC_UnTrack(self); + Py_XDECREF(self->md); + PyObject_GC_Del(self); +} + +static inline int +multidict_iter_traverse(MultidictIter *self, visitproc visit, void *arg) +{ + Py_VISIT(self->md); + return 0; +} + +static inline int +multidict_iter_clear(MultidictIter *self) +{ + Py_CLEAR(self->md); + return 0; +} + +static inline PyObject * +multidict_iter_len(MultidictIter *self) +{ + return PyLong_FromLong(pair_list_len(&self->md->pairs)); +} + +PyDoc_STRVAR(length_hint_doc, + "Private method returning an estimate of len(list(it))."); + +static PyMethodDef multidict_iter_methods[] = { + { + "__length_hint__", + (PyCFunction)(void(*)(void))multidict_iter_len, + METH_NOARGS, + length_hint_doc + }, + { + NULL, + NULL + } /* sentinel */ +}; + +/***********************************************************************/ + +static PyTypeObject multidict_items_iter_type = { + PyVarObject_HEAD_INIT(DEFERRED_ADDRESS(&PyType_Type), 0) + "multidict._multidict._itemsiter", /* tp_name */ + sizeof(MultidictIter), /* tp_basicsize */ + .tp_dealloc = (destructor)multidict_iter_dealloc, + .tp_flags = Py_TPFLAGS_DEFAULT | Py_TPFLAGS_HAVE_GC, + .tp_traverse = (traverseproc)multidict_iter_traverse, + .tp_clear = (inquiry)multidict_iter_clear, + .tp_iter = PyObject_SelfIter, + .tp_iternext = (iternextfunc)multidict_items_iter_iternext, + .tp_methods = multidict_iter_methods, +}; + +static PyTypeObject multidict_values_iter_type = { + PyVarObject_HEAD_INIT(DEFERRED_ADDRESS(&PyType_Type), 0) + "multidict._multidict._valuesiter", /* tp_name */ + sizeof(MultidictIter), /* tp_basicsize */ + .tp_dealloc = (destructor)multidict_iter_dealloc, + .tp_flags = Py_TPFLAGS_DEFAULT | Py_TPFLAGS_HAVE_GC, + .tp_traverse = (traverseproc)multidict_iter_traverse, + .tp_clear = (inquiry)multidict_iter_clear, + .tp_iter = PyObject_SelfIter, + .tp_iternext = (iternextfunc)multidict_values_iter_iternext, + .tp_methods = multidict_iter_methods, +}; + +static PyTypeObject multidict_keys_iter_type = { + PyVarObject_HEAD_INIT(DEFERRED_ADDRESS(&PyType_Type), 0) + "multidict._multidict._keysiter", /* tp_name */ + sizeof(MultidictIter), /* tp_basicsize */ + .tp_dealloc = (destructor)multidict_iter_dealloc, + .tp_flags = Py_TPFLAGS_DEFAULT | Py_TPFLAGS_HAVE_GC, + .tp_traverse = (traverseproc)multidict_iter_traverse, + .tp_clear = (inquiry)multidict_iter_clear, + .tp_iter = PyObject_SelfIter, + .tp_iternext = (iternextfunc)multidict_keys_iter_iternext, + .tp_methods = multidict_iter_methods, +}; + +static inline int +multidict_iter_init() +{ + if (PyType_Ready(&multidict_items_iter_type) < 0 || + PyType_Ready(&multidict_values_iter_type) < 0 || + PyType_Ready(&multidict_keys_iter_type) < 0) { + return -1; + } + return 0; +} + +#ifdef __cplusplus +} +#endif +#endif diff --git a/contrib/python/multidict/multidict/_multilib/pair_list.h b/contrib/python/multidict/multidict/_multilib/pair_list.h new file mode 100644 index 0000000000..7eafd215b5 --- /dev/null +++ b/contrib/python/multidict/multidict/_multilib/pair_list.h @@ -0,0 +1,1244 @@ +#ifndef _MULTIDICT_PAIR_LIST_H +#define _MULTIDICT_PAIR_LIST_H + +#ifdef __cplusplus +extern "C" { +#endif + +#include <string.h> +#include <stddef.h> +#include <stdint.h> + +typedef PyObject * (*calc_identity_func)(PyObject *key); + +typedef struct pair { + PyObject *identity; // 8 + PyObject *key; // 8 + PyObject *value; // 8 + Py_hash_t hash; // 8 +} pair_t; + +/* Note about the structure size +With 29 pairs the MultiDict object size is slightly less than 1KiB +(1000-1008 bytes depending on Python version, +plus extra 12 bytes for memory allocator internal structures). +As the result the max reserved size is 1020 bytes at most. + +To fit into 512 bytes, the structure can contain only 13 pairs +which is too small, e.g. https://www.python.org returns 16 headers +(9 of them are caching proxy information though). + +The embedded buffer intention is to fit the vast majority of possible +HTTP headers into the buffer without allocating an extra memory block. +*/ + +#if (PY_VERSION_HEX < 0x03080000) +#define EMBEDDED_CAPACITY 28 +#else +#define EMBEDDED_CAPACITY 29 +#endif + +typedef struct pair_list { // 40 + Py_ssize_t capacity; // 8 + Py_ssize_t size; // 8 + uint64_t version; // 8 + calc_identity_func calc_identity; // 8 + pair_t *pairs; // 8 + pair_t buffer[EMBEDDED_CAPACITY]; +} pair_list_t; + +#define MIN_CAPACITY 63 +#define CAPACITY_STEP 64 + +/* Global counter used to set ma_version_tag field of dictionary. + * It is incremented each time that a dictionary is created and each + * time that a dictionary is modified. */ +static uint64_t pair_list_global_version = 0; + +#define NEXT_VERSION() (++pair_list_global_version) + + +static inline int +str_cmp(PyObject *s1, PyObject *s2) +{ + PyObject *ret = PyUnicode_RichCompare(s1, s2, Py_EQ); + if (ret == Py_True) { + Py_DECREF(ret); + return 1; + } + else if (ret == NULL) { + return -1; + } + else { + Py_DECREF(ret); + return 0; + } +} + + +static inline PyObject * +key_to_str(PyObject *key) +{ + PyObject *ret; + PyTypeObject *type = Py_TYPE(key); + if (type == &istr_type) { + ret = ((istrobject*)key)->canonical; + Py_INCREF(ret); + return ret; + } + if (PyUnicode_CheckExact(key)) { + Py_INCREF(key); + return key; + } + if (PyUnicode_Check(key)) { + return PyObject_Str(key); + } + PyErr_SetString(PyExc_TypeError, + "MultiDict keys should be either str " + "or subclasses of str"); + return NULL; +} + + +static inline PyObject * +ci_key_to_str(PyObject *key) +{ + PyObject *ret; + PyTypeObject *type = Py_TYPE(key); + if (type == &istr_type) { + ret = ((istrobject*)key)->canonical; + Py_INCREF(ret); + return ret; + } + if (PyUnicode_Check(key)) { + return _PyObject_CallMethodId(key, &PyId_lower, NULL); + } + PyErr_SetString(PyExc_TypeError, + "CIMultiDict keys should be either str " + "or subclasses of str"); + return NULL; +} + +static inline pair_t * +pair_list_get(pair_list_t *list, Py_ssize_t i) +{ + pair_t *item = list->pairs + i; + return item; +} + + +static inline int +pair_list_grow(pair_list_t *list) +{ + // Grow by one element if needed + Py_ssize_t new_capacity; + pair_t *new_pairs; + + if (list->size < list->capacity) { + return 0; + } + + if (list->pairs == list->buffer) { + new_pairs = PyMem_New(pair_t, MIN_CAPACITY); + memcpy(new_pairs, list->buffer, (size_t)list->capacity * sizeof(pair_t)); + + list->pairs = new_pairs; + list->capacity = MIN_CAPACITY; + return 0; + } else { + new_capacity = list->capacity + CAPACITY_STEP; + new_pairs = PyMem_Resize(list->pairs, pair_t, (size_t)new_capacity); + + if (NULL == new_pairs) { + // Resizing error + return -1; + } + + list->pairs = new_pairs; + list->capacity = new_capacity; + return 0; + } +} + + +static inline int +pair_list_shrink(pair_list_t *list) +{ + // Shrink by one element if needed. + // Optimization is applied to prevent jitter + // (grow-shrink-grow-shrink on adding-removing the single element + // when the buffer is full). + // To prevent this, the buffer is resized if the size is less than the capacity + // by 2*CAPACITY_STEP factor. + // The switch back to embedded buffer is never performed for both reasons: + // the code simplicity and the jitter prevention. + + pair_t *new_pairs; + Py_ssize_t new_capacity; + + if (list->capacity - list->size < 2 * CAPACITY_STEP) { + return 0; + } + new_capacity = list->capacity - CAPACITY_STEP; + if (new_capacity < MIN_CAPACITY) { + return 0; + } + + new_pairs = PyMem_Resize(list->pairs, pair_t, (size_t)new_capacity); + + if (NULL == new_pairs) { + // Resizing error + return -1; + } + + list->pairs = new_pairs; + list->capacity = new_capacity; + + return 0; +} + + +static inline int +_pair_list_init(pair_list_t *list, calc_identity_func calc_identity) +{ + list->pairs = list->buffer; + list->capacity = EMBEDDED_CAPACITY; + list->size = 0; + list->version = NEXT_VERSION(); + list->calc_identity = calc_identity; + return 0; +} + +static inline int +pair_list_init(pair_list_t *list) +{ + return _pair_list_init(list, key_to_str); +} + + +static inline int +ci_pair_list_init(pair_list_t *list) +{ + return _pair_list_init(list, ci_key_to_str); +} + + +static inline void +pair_list_dealloc(pair_list_t *list) +{ + pair_t *pair; + Py_ssize_t pos; + + for (pos = 0; pos < list->size; pos++) { + pair = pair_list_get(list, pos); + + Py_XDECREF(pair->identity); + Py_XDECREF(pair->key); + Py_XDECREF(pair->value); + } + + /* + Strictly speaking, resetting size and capacity and + assigning pairs to buffer is not necessary. + Do it to consistency and idemotency. + The cleanup doesn't hurt performance. + !!! + !!! The buffer deletion is crucial though. + !!! + */ + list->size = 0; + if (list->pairs != list->buffer) { + PyMem_Del(list->pairs); + list->pairs = list->buffer; + list->capacity = EMBEDDED_CAPACITY; + } +} + + +static inline Py_ssize_t +pair_list_len(pair_list_t *list) +{ + return list->size; +} + + +static inline int +_pair_list_add_with_hash(pair_list_t *list, + PyObject *identity, + PyObject *key, + PyObject *value, + Py_hash_t hash) +{ + pair_t *pair; + + if (pair_list_grow(list) < 0) { + return -1; + } + + pair = pair_list_get(list, list->size); + + Py_INCREF(identity); + pair->identity = identity; + + Py_INCREF(key); + pair->key = key; + + Py_INCREF(value); + pair->value = value; + + pair->hash = hash; + + list->version = NEXT_VERSION(); + list->size += 1; + + return 0; +} + + +static inline int +pair_list_add(pair_list_t *list, + PyObject *key, + PyObject *value) +{ + Py_hash_t hash; + PyObject *identity = NULL; + int ret; + + identity = list->calc_identity(key); + if (identity == NULL) { + goto fail; + } + hash = PyObject_Hash(identity); + if (hash == -1) { + goto fail; + } + ret = _pair_list_add_with_hash(list, identity, key, value, hash); + Py_DECREF(identity); + return ret; +fail: + Py_XDECREF(identity); + return -1; +} + + +static inline int +pair_list_del_at(pair_list_t *list, Py_ssize_t pos) +{ + // return 1 on success, -1 on failure + Py_ssize_t tail; + pair_t *pair; + + pair = pair_list_get(list, pos); + Py_DECREF(pair->identity); + Py_DECREF(pair->key); + Py_DECREF(pair->value); + + list->size -= 1; + list->version = NEXT_VERSION(); + + if (list->size == pos) { + // remove from tail, no need to shift body + return 0; + } + + tail = list->size - pos; + // TODO: raise an error if tail < 0 + memmove((void *)pair_list_get(list, pos), + (void *)pair_list_get(list, pos + 1), + sizeof(pair_t) * (size_t)tail); + + return pair_list_shrink(list); +} + + +static inline int +_pair_list_drop_tail(pair_list_t *list, PyObject *identity, Py_hash_t hash, + Py_ssize_t pos) +{ + // return 1 if deleted, 0 if not found + pair_t *pair; + int ret; + int found = 0; + + if (pos >= list->size) { + return 0; + } + + for (; pos < list->size; pos++) { + pair = pair_list_get(list, pos); + if (pair->hash != hash) { + continue; + } + ret = str_cmp(pair->identity, identity); + if (ret > 0) { + if (pair_list_del_at(list, pos) < 0) { + return -1; + } + found = 1; + pos--; + } + else if (ret == -1) { + return -1; + } + } + + return found; +} + +static inline int +_pair_list_del_hash(pair_list_t *list, PyObject *identity, + PyObject *key, Py_hash_t hash) +{ + int ret = _pair_list_drop_tail(list, identity, hash, 0); + + if (ret < 0) { + return -1; + } + else if (ret == 0) { + PyErr_SetObject(PyExc_KeyError, key); + return -1; + } + else { + list->version = NEXT_VERSION(); + return 0; + } +} + + +static inline int +pair_list_del(pair_list_t *list, PyObject *key) +{ + PyObject *identity = NULL; + Py_hash_t hash; + int ret; + + identity = list->calc_identity(key); + if (identity == NULL) { + goto fail; + } + + hash = PyObject_Hash(identity); + if (hash == -1) { + goto fail; + } + + ret = _pair_list_del_hash(list, identity, key, hash); + Py_DECREF(identity); + return ret; +fail: + Py_XDECREF(identity); + return -1; +} + + +static inline uint64_t +pair_list_version(pair_list_t *list) +{ + return list->version; +} + + +static inline int +_pair_list_next(pair_list_t *list, Py_ssize_t *ppos, PyObject **pidentity, + PyObject **pkey, PyObject **pvalue, Py_hash_t *phash) +{ + pair_t *pair; + + if (*ppos >= list->size) { + return 0; + } + + pair = pair_list_get(list, *ppos); + + if (pidentity) { + *pidentity = pair->identity; + } + if (pkey) { + *pkey = pair->key; + } + if (pvalue) { + *pvalue = pair->value; + } + if (phash) { + *phash = pair->hash; + } + + *ppos += 1; + return 1; +} + + +static inline int +pair_list_next(pair_list_t *list, Py_ssize_t *ppos, PyObject **pidentity, + PyObject **pkey, PyObject **pvalue) +{ + Py_hash_t hash; + return _pair_list_next(list, ppos, pidentity, pkey, pvalue, &hash); +} + + +static inline int +pair_list_contains(pair_list_t *list, PyObject *key) +{ + Py_hash_t hash1, hash2; + Py_ssize_t pos = 0; + PyObject *ident = NULL; + PyObject *identity = NULL; + int tmp; + + ident = list->calc_identity(key); + if (ident == NULL) { + goto fail; + } + + hash1 = PyObject_Hash(ident); + if (hash1 == -1) { + goto fail; + } + + while (_pair_list_next(list, &pos, &identity, NULL, NULL, &hash2)) { + if (hash1 != hash2) { + continue; + } + tmp = str_cmp(ident, identity); + if (tmp > 0) { + Py_DECREF(ident); + return 1; + } + else if (tmp < 0) { + goto fail; + } + } + + Py_DECREF(ident); + return 0; +fail: + Py_XDECREF(ident); + return -1; +} + + +static inline PyObject * +pair_list_get_one(pair_list_t *list, PyObject *key) +{ + Py_hash_t hash1, hash2; + Py_ssize_t pos = 0; + PyObject *ident = NULL; + PyObject *identity = NULL; + PyObject *value = NULL; + int tmp; + + ident = list->calc_identity(key); + if (ident == NULL) { + goto fail; + } + + hash1 = PyObject_Hash(ident); + if (hash1 == -1) { + goto fail; + } + + while (_pair_list_next(list, &pos, &identity, NULL, &value, &hash2)) { + if (hash1 != hash2) { + continue; + } + tmp = str_cmp(ident, identity); + if (tmp > 0) { + Py_INCREF(value); + Py_DECREF(ident); + return value; + } + else if (tmp < 0) { + goto fail; + } + } + + Py_DECREF(ident); + PyErr_SetObject(PyExc_KeyError, key); + return NULL; +fail: + Py_XDECREF(ident); + return NULL; +} + + +static inline PyObject * +pair_list_get_all(pair_list_t *list, PyObject *key) +{ + Py_hash_t hash1, hash2; + Py_ssize_t pos = 0; + PyObject *ident = NULL; + PyObject *identity = NULL; + PyObject *value = NULL; + PyObject *res = NULL; + int tmp; + + ident = list->calc_identity(key); + if (ident == NULL) { + goto fail; + } + + hash1 = PyObject_Hash(ident); + if (hash1 == -1) { + goto fail; + } + + while (_pair_list_next(list, &pos, &identity, NULL, &value, &hash2)) { + if (hash1 != hash2) { + continue; + } + tmp = str_cmp(ident, identity); + if (tmp > 0) { + if (res == NULL) { + res = PyList_New(1); + if (res == NULL) { + goto fail; + } + if (PyList_SetItem(res, 0, value) < 0) { + goto fail; + } + Py_INCREF(value); + } + else if (PyList_Append(res, value) < 0) { + goto fail; + } + } + else if (tmp < 0) { + goto fail; + } + } + + if (res == NULL) { + PyErr_SetObject(PyExc_KeyError, key); + } + Py_DECREF(ident); + return res; + +fail: + Py_XDECREF(ident); + Py_XDECREF(res); + return NULL; +} + + +static inline PyObject * +pair_list_set_default(pair_list_t *list, PyObject *key, PyObject *value) +{ + Py_hash_t hash1, hash2; + Py_ssize_t pos = 0; + PyObject *ident = NULL; + PyObject *identity = NULL; + PyObject *value2 = NULL; + int tmp; + + ident = list->calc_identity(key); + if (ident == NULL) { + goto fail; + } + + hash1 = PyObject_Hash(ident); + if (hash1 == -1) { + goto fail; + } + + while (_pair_list_next(list, &pos, &identity, NULL, &value2, &hash2)) { + if (hash1 != hash2) { + continue; + } + tmp = str_cmp(ident, identity); + if (tmp > 0) { + Py_INCREF(value2); + Py_DECREF(ident); + return value2; + } + else if (tmp < 0) { + goto fail; + } + } + + if (_pair_list_add_with_hash(list, ident, key, value, hash1) < 0) { + goto fail; + } + + Py_INCREF(value); + Py_DECREF(ident); + return value; +fail: + Py_XDECREF(ident); + return NULL; +} + + +static inline PyObject * +pair_list_pop_one(pair_list_t *list, PyObject *key) +{ + pair_t *pair; + + Py_hash_t hash; + Py_ssize_t pos; + PyObject *value = NULL; + int tmp; + PyObject *ident = NULL; + + ident = list->calc_identity(key); + if (ident == NULL) { + goto fail; + } + + hash = PyObject_Hash(ident); + if (hash == -1) { + goto fail; + } + + for (pos=0; pos < list->size; pos++) { + pair = pair_list_get(list, pos); + if (pair->hash != hash) { + continue; + } + tmp = str_cmp(ident, pair->identity); + if (tmp > 0) { + value = pair->value; + Py_INCREF(value); + if (pair_list_del_at(list, pos) < 0) { + goto fail; + } + Py_DECREF(ident); + return value; + } + else if (tmp < 0) { + goto fail; + } + } + + PyErr_SetObject(PyExc_KeyError, key); + goto fail; + +fail: + Py_XDECREF(value); + Py_XDECREF(ident); + return NULL; +} + + +static inline PyObject * +pair_list_pop_all(pair_list_t *list, PyObject *key) +{ + Py_hash_t hash; + Py_ssize_t pos; + pair_t *pair; + int tmp; + PyObject *res = NULL; + PyObject *ident = NULL; + + ident = list->calc_identity(key); + if (ident == NULL) { + goto fail; + } + + hash = PyObject_Hash(ident); + if (hash == -1) { + goto fail; + } + + if (list->size == 0) { + PyErr_SetObject(PyExc_KeyError, ident); + goto fail; + } + + for (pos = list->size - 1; pos >= 0; pos--) { + pair = pair_list_get(list, pos); + if (hash != pair->hash) { + continue; + } + tmp = str_cmp(ident, pair->identity); + if (tmp > 0) { + if (res == NULL) { + res = PyList_New(1); + if (res == NULL) { + goto fail; + } + if (PyList_SetItem(res, 0, pair->value) < 0) { + goto fail; + } + Py_INCREF(pair->value); + } else if (PyList_Append(res, pair->value) < 0) { + goto fail; + } + if (pair_list_del_at(list, pos) < 0) { + goto fail; + } + } + else if (tmp < 0) { + goto fail; + } + } + + if (res == NULL) { + PyErr_SetObject(PyExc_KeyError, key); + } else if (PyList_Reverse(res) < 0) { + goto fail; + } + Py_DECREF(ident); + return res; + +fail: + Py_XDECREF(ident); + Py_XDECREF(res); + return NULL; +} + + +static inline PyObject * +pair_list_pop_item(pair_list_t *list) +{ + PyObject *ret; + pair_t *pair; + + if (list->size == 0) { + PyErr_SetString(PyExc_KeyError, "empty multidict"); + return NULL; + } + + pair = pair_list_get(list, 0); + ret = PyTuple_Pack(2, pair->key, pair->value); + if (ret == NULL) { + return NULL; + } + + if (pair_list_del_at(list, 0) < 0) { + Py_DECREF(ret); + return NULL; + } + + return ret; +} + + +static inline int +pair_list_replace(pair_list_t *list, PyObject * key, PyObject *value) +{ + pair_t *pair; + + Py_ssize_t pos; + int tmp; + int found = 0; + + PyObject *identity = NULL; + Py_hash_t hash; + + identity = list->calc_identity(key); + if (identity == NULL) { + goto fail; + } + + hash = PyObject_Hash(identity); + if (hash == -1) { + goto fail; + } + + + for (pos = 0; pos < list->size; pos++) { + pair = pair_list_get(list, pos); + if (hash != pair->hash) { + continue; + } + tmp = str_cmp(identity, pair->identity); + if (tmp > 0) { + found = 1; + Py_INCREF(key); + Py_DECREF(pair->key); + pair->key = key; + Py_INCREF(value); + Py_DECREF(pair->value); + pair->value = value; + break; + } + else if (tmp < 0) { + goto fail; + } + } + + if (!found) { + if (_pair_list_add_with_hash(list, identity, key, value, hash) < 0) { + goto fail; + } + Py_DECREF(identity); + return 0; + } + else { + list->version = NEXT_VERSION(); + if (_pair_list_drop_tail(list, identity, hash, pos+1) < 0) { + goto fail; + } + Py_DECREF(identity); + return 0; + } +fail: + Py_XDECREF(identity); + return -1; +} + + +static inline int +_dict_set_number(PyObject *dict, PyObject *key, Py_ssize_t num) +{ + PyObject *tmp = PyLong_FromSsize_t(num); + if (tmp == NULL) { + return -1; + } + + if (PyDict_SetItem(dict, key, tmp) < 0) { + Py_DECREF(tmp); + return -1; + } + + return 0; +} + + +static inline int +_pair_list_post_update(pair_list_t *list, PyObject* used_keys, Py_ssize_t pos) +{ + pair_t *pair; + PyObject *tmp; + Py_ssize_t num; + + for (; pos < list->size; pos++) { + pair = pair_list_get(list, pos); + tmp = PyDict_GetItem(used_keys, pair->identity); + if (tmp == NULL) { + // not found + continue; + } + + num = PyLong_AsSsize_t(tmp); + if (num == -1) { + if (!PyErr_Occurred()) { + PyErr_SetString(PyExc_RuntimeError, "invalid internal state"); + } + return -1; + } + + if (pos >= num) { + // del self[pos] + if (pair_list_del_at(list, pos) < 0) { + return -1; + } + pos--; + } + } + + list->version = NEXT_VERSION(); + return 0; +} + +// TODO: need refactoring function name +static inline int +_pair_list_update(pair_list_t *list, PyObject *key, + PyObject *value, PyObject *used_keys, + PyObject *identity, Py_hash_t hash) +{ + PyObject *item = NULL; + pair_t *pair = NULL; + Py_ssize_t pos; + int found; + int ident_cmp_res; + + item = PyDict_GetItem(used_keys, identity); + if (item == NULL) { + pos = 0; + } + else { + pos = PyLong_AsSsize_t(item); + if (pos == -1) { + if (!PyErr_Occurred()) { + PyErr_SetString(PyExc_RuntimeError, "invalid internal state"); + } + return -1; + } + } + + found = 0; + for (; pos < list->size; pos++) { + pair = pair_list_get(list, pos); + if (pair->hash != hash) { + continue; + } + + ident_cmp_res = str_cmp(pair->identity, identity); + if (ident_cmp_res > 0) { + Py_INCREF(key); + Py_DECREF(pair->key); + pair->key = key; + + Py_INCREF(value); + Py_DECREF(pair->value); + pair->value = value; + + if (_dict_set_number(used_keys, pair->identity, pos + 1) < 0) { + return -1; + } + + found = 1; + break; + } + else if (ident_cmp_res < 0) { + return -1; + } + } + + if (!found) { + if (_pair_list_add_with_hash(list, identity, key, value, hash) < 0) { + return -1; + } + if (_dict_set_number(used_keys, identity, list->size) < 0) { + return -1; + } + } + + return 0; +} + + +static inline int +pair_list_update(pair_list_t *list, pair_list_t *other) +{ + PyObject *used_keys = NULL; + pair_t *pair = NULL; + + Py_ssize_t pos; + + if (other->size == 0) { + return 0; + } + + used_keys = PyDict_New(); + if (used_keys == NULL) { + return -1; + } + + for (pos = 0; pos < other->size; pos++) { + pair = pair_list_get(other, pos); + if (_pair_list_update(list, pair->key, pair->value, used_keys, + pair->identity, pair->hash) < 0) { + goto fail; + } + } + + if (_pair_list_post_update(list, used_keys, 0) < 0) { + goto fail; + } + + Py_DECREF(used_keys); + return 0; + +fail: + Py_XDECREF(used_keys); + return -1; +} + + +static inline int +pair_list_update_from_seq(pair_list_t *list, PyObject *seq) +{ + PyObject *it = NULL; // iter(seq) + PyObject *fast = NULL; // item as a 2-tuple or 2-list + PyObject *item = NULL; // seq[i] + PyObject *used_keys = NULL; // dict(<Identitty: Pos>) + + PyObject *key = NULL; + PyObject *value = NULL; + PyObject *identity = NULL; + + Py_hash_t hash; + + Py_ssize_t i; + Py_ssize_t n; + + it = PyObject_GetIter(seq); + if (it == NULL) { + return -1; + } + + used_keys = PyDict_New(); + if (used_keys == NULL) { + goto fail_1; + } + + for (i = 0; ; ++i) { // i - index into seq of current element + fast = NULL; + item = PyIter_Next(it); + if (item == NULL) { + if (PyErr_Occurred()) { + goto fail_1; + } + break; + } + + // Convert item to sequence, and verify length 2. + fast = PySequence_Fast(item, ""); + if (fast == NULL) { + if (PyErr_ExceptionMatches(PyExc_TypeError)) { + PyErr_Format(PyExc_TypeError, + "multidict cannot convert sequence element #%zd" + " to a sequence", + i); + } + goto fail_1; + } + + n = PySequence_Fast_GET_SIZE(fast); + if (n != 2) { + PyErr_Format(PyExc_ValueError, + "multidict update sequence element #%zd " + "has length %zd; 2 is required", + i, n); + goto fail_1; + } + + key = PySequence_Fast_GET_ITEM(fast, 0); + value = PySequence_Fast_GET_ITEM(fast, 1); + Py_INCREF(key); + Py_INCREF(value); + + identity = list->calc_identity(key); + if (identity == NULL) { + goto fail_1; + } + + hash = PyObject_Hash(identity); + if (hash == -1) { + goto fail_1; + } + + if (_pair_list_update(list, key, value, used_keys, identity, hash) < 0) { + goto fail_1; + } + + Py_DECREF(key); + Py_DECREF(value); + Py_DECREF(fast); + Py_DECREF(item); + Py_DECREF(identity); + } + + if (_pair_list_post_update(list, used_keys, 0) < 0) { + goto fail_2; + } + + Py_DECREF(it); + Py_DECREF(used_keys); + return 0; + +fail_1: + Py_XDECREF(key); + Py_XDECREF(value); + Py_XDECREF(fast); + Py_XDECREF(item); + Py_XDECREF(identity); + +fail_2: + Py_XDECREF(it); + Py_XDECREF(used_keys); + return -1; +} + +static inline int +pair_list_eq_to_mapping(pair_list_t *list, PyObject *other) +{ + PyObject *key = NULL; + PyObject *avalue = NULL; + PyObject *bvalue; + + Py_ssize_t pos, other_len; + + int eq; + + if (!PyMapping_Check(other)) { + PyErr_Format(PyExc_TypeError, + "other argument must be a mapping, not %s", + Py_TYPE(other)->tp_name); + return -1; + } + + other_len = PyMapping_Size(other); + if (other_len < 0) { + return -1; + } + if (pair_list_len(list) != other_len) { + return 0; + } + + pos = 0; + while (pair_list_next(list, &pos, NULL, &key, &avalue)) { + bvalue = PyObject_GetItem(other, key); + if (bvalue == NULL) { + if (PyErr_ExceptionMatches(PyExc_KeyError)) { + PyErr_Clear(); + return 0; + } + return -1; + } + + eq = PyObject_RichCompareBool(avalue, bvalue, Py_EQ); + Py_DECREF(bvalue); + + if (eq <= 0) { + return eq; + } + } + + return 1; +} + + +/***********************************************************************/ + +static inline int +pair_list_traverse(pair_list_t *list, visitproc visit, void *arg) +{ + pair_t *pair = NULL; + Py_ssize_t pos; + + for (pos = 0; pos < list->size; pos++) { + pair = pair_list_get(list, pos); + // Don't need traverse the identity: it is a terminal + Py_VISIT(pair->key); + Py_VISIT(pair->value); + } + + return 0; +} + + +static inline int +pair_list_clear(pair_list_t *list) +{ + pair_t *pair = NULL; + Py_ssize_t pos; + + if (list->size == 0) { + return 0; + } + + list->version = NEXT_VERSION(); + for (pos = 0; pos < list->size; pos++) { + pair = pair_list_get(list, pos); + Py_CLEAR(pair->key); + Py_CLEAR(pair->identity); + Py_CLEAR(pair->value); + } + list->size = 0; + if (list->pairs != list->buffer) { + PyMem_Del(list->pairs); + list->pairs = list->buffer; + } + + return 0; +} + + +#ifdef __cplusplus +} +#endif +#endif diff --git a/contrib/python/multidict/multidict/_multilib/views.h b/contrib/python/multidict/multidict/_multilib/views.h new file mode 100644 index 0000000000..5b1ebfe77c --- /dev/null +++ b/contrib/python/multidict/multidict/_multilib/views.h @@ -0,0 +1,464 @@ +#ifndef _MULTIDICT_VIEWS_H +#define _MULTIDICT_VIEWS_H + +#ifdef __cplusplus +extern "C" { +#endif + +static PyTypeObject multidict_itemsview_type; +static PyTypeObject multidict_valuesview_type; +static PyTypeObject multidict_keysview_type; + +static PyObject *viewbaseset_richcmp_func; +static PyObject *viewbaseset_and_func; +static PyObject *viewbaseset_or_func; +static PyObject *viewbaseset_sub_func; +static PyObject *viewbaseset_xor_func; + +static PyObject *abc_itemsview_register_func; +static PyObject *abc_keysview_register_func; +static PyObject *abc_valuesview_register_func; + +static PyObject *itemsview_isdisjoint_func; +static PyObject *itemsview_repr_func; + +static PyObject *keysview_repr_func; +static PyObject *keysview_isdisjoint_func; + +static PyObject *valuesview_repr_func; + +typedef struct { + PyObject_HEAD + PyObject *md; +} _Multidict_ViewObject; + + +/********** Base **********/ + +static inline void +_init_view(_Multidict_ViewObject *self, PyObject *md) +{ + Py_INCREF(md); + self->md = md; +} + +static inline void +multidict_view_dealloc(_Multidict_ViewObject *self) +{ + PyObject_GC_UnTrack(self); + Py_XDECREF(self->md); + PyObject_GC_Del(self); +} + +static inline int +multidict_view_traverse(_Multidict_ViewObject *self, visitproc visit, void *arg) +{ + Py_VISIT(self->md); + return 0; +} + +static inline int +multidict_view_clear(_Multidict_ViewObject *self) +{ + Py_CLEAR(self->md); + return 0; +} + +static inline Py_ssize_t +multidict_view_len(_Multidict_ViewObject *self) +{ + return pair_list_len(&((MultiDictObject*)self->md)->pairs); +} + +static inline PyObject * +multidict_view_richcompare(PyObject *self, PyObject *other, int op) +{ + PyObject *ret; + PyObject *op_obj = PyLong_FromLong(op); + if (op_obj == NULL) { + return NULL; + } + ret = PyObject_CallFunctionObjArgs( + viewbaseset_richcmp_func, self, other, op_obj, NULL); + Py_DECREF(op_obj); + return ret; +} + +static inline PyObject * +multidict_view_and(PyObject *self, PyObject *other) +{ + return PyObject_CallFunctionObjArgs( + viewbaseset_and_func, self, other, NULL); +} + +static inline PyObject * +multidict_view_or(PyObject *self, PyObject *other) +{ + return PyObject_CallFunctionObjArgs( + viewbaseset_or_func, self, other, NULL); +} + +static inline PyObject * +multidict_view_sub(PyObject *self, PyObject *other) +{ + return PyObject_CallFunctionObjArgs( + viewbaseset_sub_func, self, other, NULL); +} + +static inline PyObject * +multidict_view_xor(PyObject *self, PyObject *other) +{ + return PyObject_CallFunctionObjArgs( + viewbaseset_xor_func, self, other, NULL); +} + +static PyNumberMethods multidict_view_as_number = { + .nb_subtract = (binaryfunc)multidict_view_sub, + .nb_and = (binaryfunc)multidict_view_and, + .nb_xor = (binaryfunc)multidict_view_xor, + .nb_or = (binaryfunc)multidict_view_or, +}; + +/********** Items **********/ + +static inline PyObject * +multidict_itemsview_new(PyObject *md) +{ + _Multidict_ViewObject *mv = PyObject_GC_New( + _Multidict_ViewObject, &multidict_itemsview_type); + if (mv == NULL) { + return NULL; + } + + _init_view(mv, md); + + PyObject_GC_Track(mv); + return (PyObject *)mv; +} + +static inline PyObject * +multidict_itemsview_iter(_Multidict_ViewObject *self) +{ + return multidict_items_iter_new((MultiDictObject*)self->md); +} + +static inline PyObject * +multidict_itemsview_repr(_Multidict_ViewObject *self) +{ + return PyObject_CallFunctionObjArgs( + itemsview_repr_func, self, NULL); +} + +static inline PyObject * +multidict_itemsview_isdisjoint(_Multidict_ViewObject *self, PyObject *other) +{ + return PyObject_CallFunctionObjArgs( + itemsview_isdisjoint_func, self, other, NULL); +} + +PyDoc_STRVAR(itemsview_isdisjoint_doc, + "Return True if two sets have a null intersection."); + +static PyMethodDef multidict_itemsview_methods[] = { + { + "isdisjoint", + (PyCFunction)multidict_itemsview_isdisjoint, + METH_O, + itemsview_isdisjoint_doc + }, + { + NULL, + NULL + } /* sentinel */ +}; + +static inline int +multidict_itemsview_contains(_Multidict_ViewObject *self, PyObject *obj) +{ + PyObject *akey = NULL, + *aval = NULL, + *bkey = NULL, + *bval = NULL, + *iter = NULL, + *item = NULL; + int ret1, ret2; + + if (!PyTuple_Check(obj) || PyTuple_GET_SIZE(obj) != 2) { + return 0; + } + + bkey = PyTuple_GET_ITEM(obj, 0); + bval = PyTuple_GET_ITEM(obj, 1); + + iter = multidict_itemsview_iter(self); + if (iter == NULL) { + return 0; + } + + while ((item = PyIter_Next(iter)) != NULL) { + akey = PyTuple_GET_ITEM(item, 0); + aval = PyTuple_GET_ITEM(item, 1); + + ret1 = PyObject_RichCompareBool(akey, bkey, Py_EQ); + if (ret1 < 0) { + Py_DECREF(iter); + Py_DECREF(item); + return -1; + } + ret2 = PyObject_RichCompareBool(aval, bval, Py_EQ); + if (ret2 < 0) { + Py_DECREF(iter); + Py_DECREF(item); + return -1; + } + if (ret1 > 0 && ret2 > 0) + { + Py_DECREF(iter); + Py_DECREF(item); + return 1; + } + + Py_DECREF(item); + } + + Py_DECREF(iter); + + if (PyErr_Occurred()) { + return -1; + } + + return 0; +} + +static PySequenceMethods multidict_itemsview_as_sequence = { + .sq_length = (lenfunc)multidict_view_len, + .sq_contains = (objobjproc)multidict_itemsview_contains, +}; + +static PyTypeObject multidict_itemsview_type = { + PyVarObject_HEAD_INIT(DEFERRED_ADDRESS(&PyType_Type), 0) + "multidict._multidict._ItemsView", /* tp_name */ + sizeof(_Multidict_ViewObject), /* tp_basicsize */ + .tp_dealloc = (destructor)multidict_view_dealloc, + .tp_repr = (reprfunc)multidict_itemsview_repr, + .tp_as_number = &multidict_view_as_number, + .tp_as_sequence = &multidict_itemsview_as_sequence, + .tp_getattro = PyObject_GenericGetAttr, + .tp_flags = Py_TPFLAGS_DEFAULT | Py_TPFLAGS_HAVE_GC, + .tp_traverse = (traverseproc)multidict_view_traverse, + .tp_clear = (inquiry)multidict_view_clear, + .tp_richcompare = multidict_view_richcompare, + .tp_iter = (getiterfunc)multidict_itemsview_iter, + .tp_methods = multidict_itemsview_methods, +}; + + +/********** Keys **********/ + +static inline PyObject * +multidict_keysview_new(PyObject *md) +{ + _Multidict_ViewObject *mv = PyObject_GC_New( + _Multidict_ViewObject, &multidict_keysview_type); + if (mv == NULL) { + return NULL; + } + + _init_view(mv, md); + + PyObject_GC_Track(mv); + return (PyObject *)mv; +} + +static inline PyObject * +multidict_keysview_iter(_Multidict_ViewObject *self) +{ + return multidict_keys_iter_new(((MultiDictObject*)self->md)); +} + +static inline PyObject * +multidict_keysview_repr(_Multidict_ViewObject *self) +{ + return PyObject_CallFunctionObjArgs( + keysview_repr_func, self, NULL); +} + +static inline PyObject * +multidict_keysview_isdisjoint(_Multidict_ViewObject *self, PyObject *other) +{ + return PyObject_CallFunctionObjArgs( + keysview_isdisjoint_func, self, other, NULL); +} + +PyDoc_STRVAR(keysview_isdisjoint_doc, + "Return True if two sets have a null intersection."); + +static PyMethodDef multidict_keysview_methods[] = { + { + "isdisjoint", + (PyCFunction)multidict_keysview_isdisjoint, + METH_O, + keysview_isdisjoint_doc + }, + { + NULL, + NULL + } /* sentinel */ +}; + +static inline int +multidict_keysview_contains(_Multidict_ViewObject *self, PyObject *key) +{ + return pair_list_contains(&((MultiDictObject*)self->md)->pairs, key); +} + +static PySequenceMethods multidict_keysview_as_sequence = { + .sq_length = (lenfunc)multidict_view_len, + .sq_contains = (objobjproc)multidict_keysview_contains, +}; + +static PyTypeObject multidict_keysview_type = { + PyVarObject_HEAD_INIT(DEFERRED_ADDRESS(&PyType_Type), 0) + "multidict._multidict._KeysView", /* tp_name */ + sizeof(_Multidict_ViewObject), /* tp_basicsize */ + .tp_dealloc = (destructor)multidict_view_dealloc, + .tp_repr = (reprfunc)multidict_keysview_repr, + .tp_as_number = &multidict_view_as_number, + .tp_as_sequence = &multidict_keysview_as_sequence, + .tp_getattro = PyObject_GenericGetAttr, + .tp_flags = Py_TPFLAGS_DEFAULT | Py_TPFLAGS_HAVE_GC, + .tp_traverse = (traverseproc)multidict_view_traverse, + .tp_clear = (inquiry)multidict_view_clear, + .tp_richcompare = multidict_view_richcompare, + .tp_iter = (getiterfunc)multidict_keysview_iter, + .tp_methods = multidict_keysview_methods, +}; + + +/********** Values **********/ + +static inline PyObject * +multidict_valuesview_new(PyObject *md) +{ + _Multidict_ViewObject *mv = PyObject_GC_New( + _Multidict_ViewObject, &multidict_valuesview_type); + if (mv == NULL) { + return NULL; + } + + _init_view(mv, md); + + PyObject_GC_Track(mv); + return (PyObject *)mv; +} + +static inline PyObject * +multidict_valuesview_iter(_Multidict_ViewObject *self) +{ + return multidict_values_iter_new(((MultiDictObject*)self->md)); +} + +static inline PyObject * +multidict_valuesview_repr(_Multidict_ViewObject *self) +{ + return PyObject_CallFunctionObjArgs( + valuesview_repr_func, self, NULL); +} + +static PySequenceMethods multidict_valuesview_as_sequence = { + .sq_length = (lenfunc)multidict_view_len, +}; + +static PyTypeObject multidict_valuesview_type = { + PyVarObject_HEAD_INIT(DEFERRED_ADDRESS(&PyType_Type), 0) + "multidict._multidict._ValuesView", /* tp_name */ + sizeof(_Multidict_ViewObject), /* tp_basicsize */ + .tp_dealloc = (destructor)multidict_view_dealloc, + .tp_repr = (reprfunc)multidict_valuesview_repr, + .tp_as_sequence = &multidict_valuesview_as_sequence, + .tp_getattro = PyObject_GenericGetAttr, + .tp_flags = Py_TPFLAGS_DEFAULT | Py_TPFLAGS_HAVE_GC, + .tp_traverse = (traverseproc)multidict_view_traverse, + .tp_clear = (inquiry)multidict_view_clear, + .tp_iter = (getiterfunc)multidict_valuesview_iter, +}; + + +static inline int +multidict_views_init() +{ + PyObject *reg_func_call_result = NULL; + PyObject *module = PyImport_ImportModule("multidict._multidict_base"); + if (module == NULL) { + goto fail; + } + +#define GET_MOD_ATTR(VAR, NAME) \ + VAR = PyObject_GetAttrString(module, NAME); \ + if (VAR == NULL) { \ + goto fail; \ + } + + GET_MOD_ATTR(viewbaseset_richcmp_func, "_viewbaseset_richcmp"); + GET_MOD_ATTR(viewbaseset_and_func, "_viewbaseset_and"); + GET_MOD_ATTR(viewbaseset_or_func, "_viewbaseset_or"); + GET_MOD_ATTR(viewbaseset_sub_func, "_viewbaseset_sub"); + GET_MOD_ATTR(viewbaseset_xor_func, "_viewbaseset_xor"); + + GET_MOD_ATTR(abc_itemsview_register_func, "_abc_itemsview_register"); + GET_MOD_ATTR(abc_keysview_register_func, "_abc_keysview_register"); + GET_MOD_ATTR(abc_valuesview_register_func, "_abc_valuesview_register"); + + GET_MOD_ATTR(itemsview_repr_func, "_itemsview_isdisjoint"); + GET_MOD_ATTR(itemsview_repr_func, "_itemsview_repr"); + + GET_MOD_ATTR(keysview_repr_func, "_keysview_repr"); + GET_MOD_ATTR(keysview_isdisjoint_func, "_keysview_isdisjoint"); + + GET_MOD_ATTR(valuesview_repr_func, "_valuesview_repr"); + + if (PyType_Ready(&multidict_itemsview_type) < 0 || + PyType_Ready(&multidict_valuesview_type) < 0 || + PyType_Ready(&multidict_keysview_type) < 0) + { + goto fail; + } + + // abc.ItemsView.register(_ItemsView) + reg_func_call_result = PyObject_CallFunctionObjArgs( + abc_itemsview_register_func, (PyObject*)&multidict_itemsview_type, NULL); + if (reg_func_call_result == NULL) { + goto fail; + } + Py_DECREF(reg_func_call_result); + + // abc.KeysView.register(_KeysView) + reg_func_call_result = PyObject_CallFunctionObjArgs( + abc_keysview_register_func, (PyObject*)&multidict_keysview_type, NULL); + if (reg_func_call_result == NULL) { + goto fail; + } + Py_DECREF(reg_func_call_result); + + // abc.ValuesView.register(_KeysView) + reg_func_call_result = PyObject_CallFunctionObjArgs( + abc_valuesview_register_func, (PyObject*)&multidict_valuesview_type, NULL); + if (reg_func_call_result == NULL) { + goto fail; + } + Py_DECREF(reg_func_call_result); + + Py_DECREF(module); + return 0; + +fail: + Py_CLEAR(module); + return -1; + +#undef GET_MOD_ATTR +} + +#ifdef __cplusplus +} +#endif +#endif diff --git a/contrib/python/multidict/multidict/py.typed b/contrib/python/multidict/multidict/py.typed new file mode 100644 index 0000000000..dfe8cc048e --- /dev/null +++ b/contrib/python/multidict/multidict/py.typed @@ -0,0 +1 @@ +PEP-561 marker.
\ No newline at end of file diff --git a/ydb/public/sdk/python3/ydb/_topic_reader/__init__.py b/contrib/python/multidict/tests/__init__.py index e69de29bb2..e69de29bb2 100644 --- a/ydb/public/sdk/python3/ydb/_topic_reader/__init__.py +++ b/contrib/python/multidict/tests/__init__.py diff --git a/contrib/python/multidict/tests/cimultidict.pickle.0 b/contrib/python/multidict/tests/cimultidict.pickle.0 new file mode 100644 index 0000000000..7b2ed00845 --- /dev/null +++ b/contrib/python/multidict/tests/cimultidict.pickle.0 @@ -0,0 +1,14 @@ +cmultidict._multidict +CIMultiDict +p0 +((lp1 +(Va +p2 +L1L +tp3 +a(g2 +L2L +tp4 +atp5 +Rp6 +.
\ No newline at end of file diff --git a/contrib/python/multidict/tests/cimultidict.pickle.1 b/contrib/python/multidict/tests/cimultidict.pickle.1 Binary files differnew file mode 100644 index 0000000000..225458ba29 --- /dev/null +++ b/contrib/python/multidict/tests/cimultidict.pickle.1 diff --git a/contrib/python/multidict/tests/cimultidict.pickle.2 b/contrib/python/multidict/tests/cimultidict.pickle.2 Binary files differnew file mode 100644 index 0000000000..d33600e615 --- /dev/null +++ b/contrib/python/multidict/tests/cimultidict.pickle.2 diff --git a/contrib/python/multidict/tests/cimultidict.pickle.3 b/contrib/python/multidict/tests/cimultidict.pickle.3 Binary files differnew file mode 100644 index 0000000000..cbb8624db0 --- /dev/null +++ b/contrib/python/multidict/tests/cimultidict.pickle.3 diff --git a/contrib/python/multidict/tests/cimultidict.pickle.4 b/contrib/python/multidict/tests/cimultidict.pickle.4 Binary files differnew file mode 100644 index 0000000000..1f5164ca37 --- /dev/null +++ b/contrib/python/multidict/tests/cimultidict.pickle.4 diff --git a/contrib/python/multidict/tests/cimultidict.pickle.5 b/contrib/python/multidict/tests/cimultidict.pickle.5 Binary files differnew file mode 100644 index 0000000000..11bf552c43 --- /dev/null +++ b/contrib/python/multidict/tests/cimultidict.pickle.5 diff --git a/contrib/python/multidict/tests/conftest.py b/contrib/python/multidict/tests/conftest.py new file mode 100644 index 0000000000..bd3b4de027 --- /dev/null +++ b/contrib/python/multidict/tests/conftest.py @@ -0,0 +1,29 @@ +import pickle + +import pytest + +from multidict._compat import USE_EXTENSIONS + +OPTIONAL_CYTHON = ( + () + if USE_EXTENSIONS + else pytest.mark.skip(reason="No extensions available") +) + + +@pytest.fixture( # type: ignore[call-overload] + scope="session", + params=[ + pytest.param("multidict._multidict", marks=OPTIONAL_CYTHON), # type: ignore + "multidict._multidict_py", + ], +) +def _multidict(request): + return pytest.importorskip(request.param) + + +def pytest_generate_tests(metafunc): + if "pickle_protocol" in metafunc.fixturenames: + metafunc.parametrize( + "pickle_protocol", list(range(pickle.HIGHEST_PROTOCOL + 1)), scope="session" + ) diff --git a/contrib/python/multidict/tests/gen_pickles.py b/contrib/python/multidict/tests/gen_pickles.py new file mode 100644 index 0000000000..028a01f8cc --- /dev/null +++ b/contrib/python/multidict/tests/gen_pickles.py @@ -0,0 +1,32 @@ +import pickle + +from multidict._compat import USE_EXTENSIONS +from multidict._multidict_py import CIMultiDict as PyCIMultiDict # noqa +from multidict._multidict_py import MultiDict as PyMultiDict # noqa + +try: + from multidict._multidict import ( # type: ignore # noqa + CIMultiDict, + MultiDict, + ) +except ImportError: + pass + + +def write(name, proto): + cls = globals()[name] + d = cls([("a", 1), ("a", 2)]) + with open("{}.pickle.{}".format(name.lower(), proto), "wb") as f: + pickle.dump(d, f, proto) + + +def generate(): + if not USE_EXTENSIONS: + raise RuntimeError("C Extension is required") + for proto in range(pickle.HIGHEST_PROTOCOL + 1): + for name in ("MultiDict", "CIMultiDict", "PyMultiDict", "PyCIMultiDict"): + write(name, proto) + + +if __name__ == "__main__": + generate() diff --git a/contrib/python/multidict/tests/multidict.pickle.0 b/contrib/python/multidict/tests/multidict.pickle.0 new file mode 100644 index 0000000000..eb979fcf72 --- /dev/null +++ b/contrib/python/multidict/tests/multidict.pickle.0 @@ -0,0 +1,14 @@ +cmultidict._multidict +MultiDict +p0 +((lp1 +(Va +p2 +L1L +tp3 +a(g2 +L2L +tp4 +atp5 +Rp6 +.
\ No newline at end of file diff --git a/contrib/python/multidict/tests/multidict.pickle.1 b/contrib/python/multidict/tests/multidict.pickle.1 Binary files differnew file mode 100644 index 0000000000..a4f211d7b1 --- /dev/null +++ b/contrib/python/multidict/tests/multidict.pickle.1 diff --git a/contrib/python/multidict/tests/multidict.pickle.2 b/contrib/python/multidict/tests/multidict.pickle.2 Binary files differnew file mode 100644 index 0000000000..b4563f879d --- /dev/null +++ b/contrib/python/multidict/tests/multidict.pickle.2 diff --git a/contrib/python/multidict/tests/multidict.pickle.3 b/contrib/python/multidict/tests/multidict.pickle.3 Binary files differnew file mode 100644 index 0000000000..415960a3ee --- /dev/null +++ b/contrib/python/multidict/tests/multidict.pickle.3 diff --git a/contrib/python/multidict/tests/multidict.pickle.4 b/contrib/python/multidict/tests/multidict.pickle.4 Binary files differnew file mode 100644 index 0000000000..00ef17c3f7 --- /dev/null +++ b/contrib/python/multidict/tests/multidict.pickle.4 diff --git a/contrib/python/multidict/tests/multidict.pickle.5 b/contrib/python/multidict/tests/multidict.pickle.5 Binary files differnew file mode 100644 index 0000000000..2c4ae0a0d3 --- /dev/null +++ b/contrib/python/multidict/tests/multidict.pickle.5 diff --git a/contrib/python/multidict/tests/pycimultidict.pickle.0 b/contrib/python/multidict/tests/pycimultidict.pickle.0 new file mode 100644 index 0000000000..bd39b6db20 --- /dev/null +++ b/contrib/python/multidict/tests/pycimultidict.pickle.0 @@ -0,0 +1,14 @@ +cmultidict._multidict_py +CIMultiDict +p0 +((lp1 +(Va +p2 +L1L +tp3 +a(g2 +L2L +tp4 +atp5 +Rp6 +.
\ No newline at end of file diff --git a/contrib/python/multidict/tests/pycimultidict.pickle.1 b/contrib/python/multidict/tests/pycimultidict.pickle.1 Binary files differnew file mode 100644 index 0000000000..866003d26c --- /dev/null +++ b/contrib/python/multidict/tests/pycimultidict.pickle.1 diff --git a/contrib/python/multidict/tests/pycimultidict.pickle.2 b/contrib/python/multidict/tests/pycimultidict.pickle.2 Binary files differnew file mode 100644 index 0000000000..c9e43fef9c --- /dev/null +++ b/contrib/python/multidict/tests/pycimultidict.pickle.2 diff --git a/contrib/python/multidict/tests/pycimultidict.pickle.3 b/contrib/python/multidict/tests/pycimultidict.pickle.3 Binary files differnew file mode 100644 index 0000000000..821659fe0c --- /dev/null +++ b/contrib/python/multidict/tests/pycimultidict.pickle.3 diff --git a/contrib/python/multidict/tests/pycimultidict.pickle.4 b/contrib/python/multidict/tests/pycimultidict.pickle.4 Binary files differnew file mode 100644 index 0000000000..a17c6e9b73 --- /dev/null +++ b/contrib/python/multidict/tests/pycimultidict.pickle.4 diff --git a/contrib/python/multidict/tests/pycimultidict.pickle.5 b/contrib/python/multidict/tests/pycimultidict.pickle.5 Binary files differnew file mode 100644 index 0000000000..479bfe3a5d --- /dev/null +++ b/contrib/python/multidict/tests/pycimultidict.pickle.5 diff --git a/contrib/python/multidict/tests/pymultidict.pickle.0 b/contrib/python/multidict/tests/pymultidict.pickle.0 new file mode 100644 index 0000000000..e91023ecf9 --- /dev/null +++ b/contrib/python/multidict/tests/pymultidict.pickle.0 @@ -0,0 +1,14 @@ +cmultidict._multidict_py +MultiDict +p0 +((lp1 +(Va +p2 +L1L +tp3 +a(g2 +L2L +tp4 +atp5 +Rp6 +.
\ No newline at end of file diff --git a/contrib/python/multidict/tests/pymultidict.pickle.1 b/contrib/python/multidict/tests/pymultidict.pickle.1 Binary files differnew file mode 100644 index 0000000000..acce9bf793 --- /dev/null +++ b/contrib/python/multidict/tests/pymultidict.pickle.1 diff --git a/contrib/python/multidict/tests/pymultidict.pickle.2 b/contrib/python/multidict/tests/pymultidict.pickle.2 Binary files differnew file mode 100644 index 0000000000..900446ad8d --- /dev/null +++ b/contrib/python/multidict/tests/pymultidict.pickle.2 diff --git a/contrib/python/multidict/tests/pymultidict.pickle.3 b/contrib/python/multidict/tests/pymultidict.pickle.3 Binary files differnew file mode 100644 index 0000000000..9b9073515a --- /dev/null +++ b/contrib/python/multidict/tests/pymultidict.pickle.3 diff --git a/contrib/python/multidict/tests/pymultidict.pickle.4 b/contrib/python/multidict/tests/pymultidict.pickle.4 Binary files differnew file mode 100644 index 0000000000..db363f8d3c --- /dev/null +++ b/contrib/python/multidict/tests/pymultidict.pickle.4 diff --git a/contrib/python/multidict/tests/pymultidict.pickle.5 b/contrib/python/multidict/tests/pymultidict.pickle.5 Binary files differnew file mode 100644 index 0000000000..7dc772d58f --- /dev/null +++ b/contrib/python/multidict/tests/pymultidict.pickle.5 diff --git a/contrib/python/multidict/tests/test_abc.py b/contrib/python/multidict/tests/test_abc.py new file mode 100644 index 0000000000..4636b3bc1a --- /dev/null +++ b/contrib/python/multidict/tests/test_abc.py @@ -0,0 +1,132 @@ +from collections.abc import Mapping, MutableMapping + +import pytest + +from multidict import MultiMapping, MutableMultiMapping +from multidict._compat import USE_EXTENSIONS +from multidict._multidict_py import CIMultiDict as PyCIMultiDict +from multidict._multidict_py import CIMultiDictProxy as PyCIMultiDictProxy +from multidict._multidict_py import MultiDict as PyMultiDict # noqa: E402 +from multidict._multidict_py import MultiDictProxy as PyMultiDictProxy + +if USE_EXTENSIONS: + from multidict._multidict import ( # type: ignore + CIMultiDict, + CIMultiDictProxy, + MultiDict, + MultiDictProxy, + ) + + +@pytest.fixture( + params=([MultiDict, CIMultiDict] if USE_EXTENSIONS else []) + + [PyMultiDict, PyCIMultiDict], + ids=(["MultiDict", "CIMultiDict"] if USE_EXTENSIONS else []) + + ["PyMultiDict", "PyCIMultiDict"], +) +def cls(request): + return request.param + + +@pytest.fixture( + params=( + [(MultiDictProxy, MultiDict), (CIMultiDictProxy, CIMultiDict)] + if USE_EXTENSIONS + else [] + ) + + [(PyMultiDictProxy, PyMultiDict), (PyCIMultiDictProxy, PyCIMultiDict)], + ids=(["MultiDictProxy", "CIMultiDictProxy"] if USE_EXTENSIONS else []) + + ["PyMultiDictProxy", "PyCIMultiDictProxy"], +) +def proxy_classes(request): + return request.param + + +def test_abc_inheritance(): + assert issubclass(MultiMapping, Mapping) + assert not issubclass(MultiMapping, MutableMapping) + assert issubclass(MutableMultiMapping, Mapping) + assert issubclass(MutableMultiMapping, MutableMapping) + + +class A(MultiMapping): + def __getitem__(self, key): + pass + + def __iter__(self): + pass + + def __len__(self): + pass + + def getall(self, key, default=None): + super().getall(key, default) + + def getone(self, key, default=None): + super().getone(key, default) + + +def test_abc_getall(): + with pytest.raises(KeyError): + A().getall("key") + + +def test_abc_getone(): + with pytest.raises(KeyError): + A().getone("key") + + +class B(A, MutableMultiMapping): + def __setitem__(self, key, value): + pass + + def __delitem__(self, key): + pass + + def add(self, key, value): + super().add(key, value) + + def extend(self, *args, **kwargs): + super().extend(*args, **kwargs) + + def popall(self, key, default=None): + super().popall(key, default) + + def popone(self, key, default=None): + super().popone(key, default) + + +def test_abc_add(): + with pytest.raises(NotImplementedError): + B().add("key", "val") + + +def test_abc_extend(): + with pytest.raises(NotImplementedError): + B().extend() + + +def test_abc_popone(): + with pytest.raises(KeyError): + B().popone("key") + + +def test_abc_popall(): + with pytest.raises(KeyError): + B().popall("key") + + +def test_multidict_inheritance(cls): + assert issubclass(cls, MultiMapping) + assert issubclass(cls, MutableMultiMapping) + + +def test_proxy_inheritance(proxy_classes): + proxy, _ = proxy_classes + assert issubclass(proxy, MultiMapping) + assert not issubclass(proxy, MutableMultiMapping) + + +def test_generic_type_in_runtime(): + MultiMapping[str] + MutableMultiMapping[str] diff --git a/contrib/python/multidict/tests/test_copy.py b/contrib/python/multidict/tests/test_copy.py new file mode 100644 index 0000000000..564cdde597 --- /dev/null +++ b/contrib/python/multidict/tests/test_copy.py @@ -0,0 +1,79 @@ +import copy + +import pytest + +from multidict._compat import USE_EXTENSIONS +from multidict._multidict_py import CIMultiDict as PyCIMultiDict +from multidict._multidict_py import CIMultiDictProxy as PyCIMultiDictProxy +from multidict._multidict_py import MultiDict as PyMultiDict # noqa: E402 +from multidict._multidict_py import MultiDictProxy as PyMultiDictProxy + +if USE_EXTENSIONS: + from multidict._multidict import ( # type: ignore + CIMultiDict, + CIMultiDictProxy, + MultiDict, + MultiDictProxy, + ) + + +@pytest.fixture( + params=([MultiDict, CIMultiDict] if USE_EXTENSIONS else []) + + [PyMultiDict, PyCIMultiDict], + ids=(["MultiDict", "CIMultiDict"] if USE_EXTENSIONS else []) + + ["PyMultiDict", "PyCIMultiDict"], +) +def cls(request): + return request.param + + +@pytest.fixture( + params=( + [(MultiDictProxy, MultiDict), (CIMultiDictProxy, CIMultiDict)] + if USE_EXTENSIONS + else [] + ) + + [(PyMultiDictProxy, PyMultiDict), (PyCIMultiDictProxy, PyCIMultiDict)], + ids=(["MultiDictProxy", "CIMultiDictProxy"] if USE_EXTENSIONS else []) + + ["PyMultiDictProxy", "PyCIMultiDictProxy"], +) +def proxy_classes(request): + return request.param + + +def test_copy(cls): + d = cls() + d["foo"] = 6 + d2 = d.copy() + d2["foo"] = 7 + assert d["foo"] == 6 + assert d2["foo"] == 7 + + +def test_copy_proxy(proxy_classes): + proxy_cls, dict_cls = proxy_classes + d = dict_cls() + d["foo"] = 6 + p = proxy_cls(d) + d2 = p.copy() + d2["foo"] = 7 + assert d["foo"] == 6 + assert p["foo"] == 6 + assert d2["foo"] == 7 + + +def test_copy_std_copy(cls): + d = cls() + d["foo"] = 6 + d2 = copy.copy(d) + d2["foo"] = 7 + assert d["foo"] == 6 + assert d2["foo"] == 7 + + +def test_ci_multidict_clone(cls): + d = cls(foo=6) + d2 = cls(d) + d2["foo"] = 7 + assert d["foo"] == 6 + assert d2["foo"] == 7 diff --git a/contrib/python/multidict/tests/test_guard.py b/contrib/python/multidict/tests/test_guard.py new file mode 100644 index 0000000000..823cc1afb8 --- /dev/null +++ b/contrib/python/multidict/tests/test_guard.py @@ -0,0 +1,39 @@ +import pytest + +from multidict._compat import USE_EXTENSIONS +from multidict._multidict_py import MultiDict as PyMultiDict # noqa: E402 + +if USE_EXTENSIONS: + from multidict._multidict import MultiDict # type: ignore + + +@pytest.fixture( + params=([MultiDict] if USE_EXTENSIONS else []) + [PyMultiDict], + ids=(["MultiDict"] if USE_EXTENSIONS else []) + ["PyMultiDict"], +) +def cls(request): + return request.param + + +def test_guard_items(cls): + md = cls({"a": "b"}) + it = iter(md.items()) + md["a"] = "c" + with pytest.raises(RuntimeError): + next(it) + + +def test_guard_keys(cls): + md = cls({"a": "b"}) + it = iter(md.keys()) + md["a"] = "c" + with pytest.raises(RuntimeError): + next(it) + + +def test_guard_values(cls): + md = cls({"a": "b"}) + it = iter(md.values()) + md["a"] = "c" + with pytest.raises(RuntimeError): + next(it) diff --git a/contrib/python/multidict/tests/test_istr.py b/contrib/python/multidict/tests/test_istr.py new file mode 100644 index 0000000000..caae397f2d --- /dev/null +++ b/contrib/python/multidict/tests/test_istr.py @@ -0,0 +1,83 @@ +import gc +import sys +from typing import Type + +import pytest + +from multidict._compat import USE_EXTENSIONS +from multidict._multidict_py import istr as _istr # noqa: E402 + +if USE_EXTENSIONS: + from multidict._multidict import istr # type: ignore +else: + from multidict import istr + + +IMPLEMENTATION = getattr(sys, "implementation") # to suppress mypy error + + +class IStrMixin: + + cls = Type[istr] + + def test_ctor(self): + s = self.cls() + assert "" == s + + def test_ctor_str(self): + s = self.cls("aBcD") + assert "aBcD" == s + + def test_ctor_istr(self): + s = self.cls("A") + s2 = self.cls(s) + assert "A" == s + assert s == s2 + + def test_ctor_buffer(self): + s = self.cls(b"aBc") + assert "b'aBc'" == s + + def test_ctor_repr(self): + s = self.cls(None) + assert "None" == s + + def test_str(self): + s = self.cls("aBcD") + s1 = str(s) + assert s1 == "aBcD" + assert type(s1) is str + + def test_eq(self): + s1 = "Abc" + s2 = self.cls(s1) + assert s1 == s2 + + +class TestPyIStr(IStrMixin): + cls = _istr + + @staticmethod + def _create_strs(): + _istr("foobarbaz") + istr2 = _istr() + _istr(istr2) + + @pytest.mark.skipif( + IMPLEMENTATION.name != "cpython", reason="PyPy has different GC implementation" + ) + def test_leak(self): + gc.collect() + cnt = len(gc.get_objects()) + for _ in range(10000): + self._create_strs() + + gc.collect() + cnt2 = len(gc.get_objects()) + assert abs(cnt - cnt2) < 10 # on PyPy these numbers are not equal + + +if USE_EXTENSIONS: + + class TestIStr(IStrMixin): + cls = istr diff --git a/contrib/python/multidict/tests/test_multidict.py b/contrib/python/multidict/tests/test_multidict.py new file mode 100644 index 0000000000..706fc93e75 --- /dev/null +++ b/contrib/python/multidict/tests/test_multidict.py @@ -0,0 +1,602 @@ +import gc +import operator +import sys +import weakref +from collections import deque +from collections.abc import Mapping +from functools import reduce +from typing import ( + Any, + Callable, + Dict, + Iterable, + Iterator, + List, + Mapping, + Set, + Tuple, + Type, + TypeVar, + Union, +) + +import pytest + +import multidict +from multidict import CIMultiDict, CIMultiDictProxy, MultiDict, MultiDictProxy + +_MultiDictClasses = Union[Type[MultiDict[str]], Type[CIMultiDict[str]]] + + +def chained_callable( + module: object, callables: Union[str, Iterable[str]] +) -> Callable[..., Any]: + """ + Returns callable that will get and call all given objects in module in + exact order. If `names` is a single object's name function will return + object itself. + + Will treat `names` of type `str` as a list of single element. + """ + callables = (callables,) if isinstance(callables, str) else callables + _callable, *rest = (getattr(module, name) for name in callables) + + def chained_call(*args: object, **kwargs: object) -> Any: + return reduce(lambda res, c: c(res), rest, _callable(*args, **kwargs)) + + return chained_call if len(rest) > 0 else _callable # type: ignore[no-any-return] + + +@pytest.fixture(scope="function") +def cls(request: Any, _multidict: Any) -> Any: + return chained_callable(_multidict, request.param) + + +@pytest.fixture(scope="function") +def classes(request: Any, _multidict: Any) -> Any: + return tuple(chained_callable(_multidict, n) for n in request.param) + + +@pytest.mark.parametrize("cls", ["MultiDict", "CIMultiDict"], indirect=True) +def test_exposed_names( + cls: Union[Type[MultiDict[object]], Type[CIMultiDict[object]]] +) -> None: + name = cls.__name__ + + while name.startswith("_"): + name = name[1:] + + assert name in multidict.__all__ # type: ignore[attr-defined] + + +@pytest.mark.parametrize( + "cls, key_cls", + [("MultiDict", str), (("MultiDict", "MultiDictProxy"), str)], + indirect=["cls"], +) +def test__iter__types( + cls: Type[MultiDict[Union[str, int]]], key_cls: Type[object] +) -> None: + d = cls([("key", "one"), ("key2", "two"), ("key", 3)]) + for i in d: + assert type(i) is key_cls, (type(i), key_cls) + + +_ClsPair = TypeVar( + "_ClsPair", + Tuple[Type[MultiDict[str]], Type[MultiDictProxy[str]]], + Tuple[Type[CIMultiDict[str]], Type[CIMultiDictProxy[str]]], +) + + +@pytest.mark.parametrize( + "classes", + [("MultiDict", "MultiDictProxy"), ("CIMultiDict", "CIMultiDictProxy")], + indirect=True, +) +def test_proxy_copy(classes: _ClsPair) -> None: + dict_cls, proxy_cls = classes + d1 = dict_cls(key="value", a="b") + p1 = proxy_cls(d1) + + d2 = p1.copy() + assert d1 == d2 + assert d1 is not d2 + + +@pytest.mark.parametrize( + "cls", + ["MultiDict", "CIMultiDict", "MultiDictProxy", "CIMultiDictProxy"], + indirect=True, +) +def test_subclassing(cls: Any) -> None: + class MyClass(cls): # type: ignore[valid-type,misc] + pass + + +class BaseMultiDictTest: + def test_instantiate__empty(self, cls: _MultiDictClasses) -> None: + d = cls() + empty: Mapping[str, str] = {} + assert d == empty + assert len(d) == 0 + assert list(d.keys()) == [] + assert list(d.values()) == [] + assert list(d.items()) == [] + + assert cls() != list() # type: ignore[comparison-overlap] + with pytest.raises(TypeError, match=r"(2 given)"): + cls(("key1", "value1"), ("key2", "value2")) # type: ignore[arg-type,call-arg] # noqa: E501 + + @pytest.mark.parametrize("arg0", [[("key", "value1")], {"key": "value1"}]) + def test_instantiate__from_arg0( + self, + cls: _MultiDictClasses, + arg0: Union[List[Tuple[str, str]], Dict[str, str]], + ) -> None: + d = cls(arg0) + + assert d == {"key": "value1"} + assert len(d) == 1 + assert list(d.keys()) == ["key"] + assert list(d.values()) == ["value1"] + assert list(d.items()) == [("key", "value1")] + + def test_instantiate__with_kwargs(self, cls: _MultiDictClasses) -> None: + d = cls([("key", "value1")], key2="value2") + + assert d == {"key": "value1", "key2": "value2"} + assert len(d) == 2 + assert sorted(d.keys()) == ["key", "key2"] + assert sorted(d.values()) == ["value1", "value2"] + assert sorted(d.items()) == [("key", "value1"), ("key2", "value2")] + + def test_instantiate__from_generator( + self, cls: Union[Type[MultiDict[int]], Type[CIMultiDict[int]]] + ) -> None: + d = cls((str(i), i) for i in range(2)) + + assert d == {"0": 0, "1": 1} + assert len(d) == 2 + assert sorted(d.keys()) == ["0", "1"] + assert sorted(d.values()) == [0, 1] + assert sorted(d.items()) == [("0", 0), ("1", 1)] + + def test_instantiate__from_list_of_lists(self, cls: _MultiDictClasses) -> None: + # Should work at runtime, but won't type check. + d = cls([["key", "value1"]]) # type: ignore[list-item] + assert d == {"key": "value1"} + + def test_instantiate__from_list_of_custom_pairs( + self, cls: _MultiDictClasses + ) -> None: + class Pair: + def __len__(self) -> int: + return 2 + + def __getitem__(self, pos: int) -> str: + if pos == 0: + return "key" + elif pos == 1: + return "value1" + else: + raise IndexError + + # Works at runtime, but won't type check. + d = cls([Pair()]) # type: ignore[list-item] + assert d == {"key": "value1"} + + def test_getone(self, cls: _MultiDictClasses) -> None: + d = cls([("key", "value1")], key="value2") + + assert d.getone("key") == "value1" + assert d.get("key") == "value1" + assert d["key"] == "value1" + + with pytest.raises(KeyError, match="key2"): + d["key2"] + with pytest.raises(KeyError, match="key2"): + d.getone("key2") + + assert d.getone("key2", "default") == "default" + + def test__iter__( + self, + cls: Union[Type[MultiDict[Union[str, int]]], Type[CIMultiDict[Union[str, int]]]] + ) -> None: + d = cls([("key", "one"), ("key2", "two"), ("key", 3)]) + assert list(d) == ["key", "key2", "key"] + + def test_keys__contains( + self, + cls: Union[Type[MultiDict[Union[str, int]]], Type[CIMultiDict[Union[str, int]]]] + ) -> None: + d = cls([("key", "one"), ("key2", "two"), ("key", 3)]) + + assert list(d.keys()) == ["key", "key2", "key"] + + assert "key" in d.keys() + assert "key2" in d.keys() + + assert "foo" not in d.keys() + + def test_values__contains( + self, + cls: Union[Type[MultiDict[Union[str, int]]], Type[CIMultiDict[Union[str, int]]]] + ) -> None: + d = cls([("key", "one"), ("key", "two"), ("key", 3)]) + + assert list(d.values()) == ["one", "two", 3] + + assert "one" in d.values() + assert "two" in d.values() + assert 3 in d.values() + + assert "foo" not in d.values() + + def test_items__contains( + self, + cls: Union[Type[MultiDict[Union[str, int]]], Type[CIMultiDict[Union[str, int]]]] + ) -> None: + d = cls([("key", "one"), ("key", "two"), ("key", 3)]) + + assert list(d.items()) == [("key", "one"), ("key", "two"), ("key", 3)] + + assert ("key", "one") in d.items() + assert ("key", "two") in d.items() + assert ("key", 3) in d.items() + + assert ("foo", "bar") not in d.items() + + def test_cannot_create_from_unaccepted(self, cls: _MultiDictClasses) -> None: + with pytest.raises(TypeError): + cls([(1, 2, 3)]) # type: ignore[list-item] + + def test_keys_is_set_less(self, cls: _MultiDictClasses) -> None: + d = cls([("key", "value1")]) + + assert d.keys() < {"key", "key2"} + + def test_keys_is_set_less_equal(self, cls: _MultiDictClasses) -> None: + d = cls([("key", "value1")]) + + assert d.keys() <= {"key"} + + def test_keys_is_set_equal(self, cls: _MultiDictClasses) -> None: + d = cls([("key", "value1")]) + + assert d.keys() == {"key"} + + def test_keys_is_set_greater(self, cls: _MultiDictClasses) -> None: + d = cls([("key", "value1")]) + + assert {"key", "key2"} > d.keys() + + def test_keys_is_set_greater_equal(self, cls: _MultiDictClasses) -> None: + d = cls([("key", "value1")]) + + assert {"key"} >= d.keys() + + def test_keys_is_set_not_equal(self, cls: _MultiDictClasses) -> None: + d = cls([("key", "value1")]) + + assert d.keys() != {"key2"} + + def test_eq(self, cls: _MultiDictClasses) -> None: + d = cls([("key", "value1")]) + + assert {"key": "value1"} == d + + def test_eq2(self, cls: _MultiDictClasses) -> None: + d1 = cls([("key", "value1")]) + d2 = cls([("key2", "value1")]) + + assert d1 != d2 + + def test_eq3(self, cls: _MultiDictClasses) -> None: + d1 = cls([("key", "value1")]) + d2 = cls() + + assert d1 != d2 + + def test_eq_other_mapping_contains_more_keys(self, cls: _MultiDictClasses) -> None: + d1 = cls(foo="bar") + d2 = dict(foo="bar", bar="baz") + + assert d1 != d2 + + def test_eq_bad_mapping_len( + self, cls: Union[Type[MultiDict[int]], Type[CIMultiDict[int]]] + ) -> None: + class BadMapping(Mapping[str, int]): + def __getitem__(self, key: str) -> int: + return 1 + + def __iter__(self) -> Iterator[str]: + yield "a" + + def __len__(self) -> int: # type: ignore[return] + 1 / 0 + + d1 = cls(a=1) + d2 = BadMapping() + with pytest.raises(ZeroDivisionError): + d1 == d2 + + def test_eq_bad_mapping_getitem( + self, + cls: Union[Type[MultiDict[int]], Type[CIMultiDict[int]]] + ) -> None: + class BadMapping(Mapping[str, int]): + def __getitem__(self, key: str) -> int: # type: ignore[return] + 1 / 0 + + def __iter__(self) -> Iterator[str]: + yield "a" + + def __len__(self) -> int: + return 1 + + d1 = cls(a=1) + d2 = BadMapping() + with pytest.raises(ZeroDivisionError): + d1 == d2 + + def test_ne(self, cls: _MultiDictClasses) -> None: + d = cls([("key", "value1")]) + + assert d != {"key": "another_value"} + + def test_and(self, cls: _MultiDictClasses) -> None: + d = cls([("key", "value1")]) + + assert {"key"} == d.keys() & {"key", "key2"} + + def test_and2(self, cls: _MultiDictClasses) -> None: + d = cls([("key", "value1")]) + + assert {"key"} == {"key", "key2"} & d.keys() + + def test_or(self, cls: _MultiDictClasses) -> None: + d = cls([("key", "value1")]) + + assert {"key", "key2"} == d.keys() | {"key2"} + + def test_or2(self, cls: _MultiDictClasses) -> None: + d = cls([("key", "value1")]) + + assert {"key", "key2"} == {"key2"} | d.keys() + + def test_sub(self, cls: _MultiDictClasses) -> None: + d = cls([("key", "value1"), ("key2", "value2")]) + + assert {"key"} == d.keys() - {"key2"} + + def test_sub2(self, cls: _MultiDictClasses) -> None: + d = cls([("key", "value1"), ("key2", "value2")]) + + assert {"key3"} == {"key", "key2", "key3"} - d.keys() + + def test_xor(self, cls: _MultiDictClasses) -> None: + d = cls([("key", "value1"), ("key2", "value2")]) + + assert {"key", "key3"} == d.keys() ^ {"key2", "key3"} + + def test_xor2(self, cls: _MultiDictClasses) -> None: + d = cls([("key", "value1"), ("key2", "value2")]) + + assert {"key", "key3"} == {"key2", "key3"} ^ d.keys() + + @pytest.mark.parametrize("_set, expected", [({"key2"}, True), ({"key"}, False)]) + def test_isdisjoint( + self, cls: _MultiDictClasses, _set: Set[str], expected: bool + ) -> None: + d = cls([("key", "value1")]) + + assert d.keys().isdisjoint(_set) == expected + + def test_repr_issue_410(self, cls: _MultiDictClasses) -> None: + d = cls() + + try: + raise Exception + pytest.fail("Should never happen") # pragma: no cover + except Exception as e: + repr(d) + + assert sys.exc_info()[1] == e + + @pytest.mark.parametrize( + "op", [operator.or_, operator.and_, operator.sub, operator.xor] + ) + @pytest.mark.parametrize("other", [{"other"}]) + def test_op_issue_410( + self, + cls: _MultiDictClasses, + op: Callable[[object, object], object], + other: Set[str], + ) -> None: + d = cls([("key", "value")]) + + try: + raise Exception + pytest.fail("Should never happen") # pragma: no cover + except Exception as e: + op(d.keys(), other) + + assert sys.exc_info()[1] == e + + def test_weakref(self, cls: _MultiDictClasses) -> None: + called = False + + def cb(wr: object) -> None: + nonlocal called + called = True + + d = cls() + wr = weakref.ref(d, cb) + del d + gc.collect() + assert called + del wr + + def test_iter_length_hint_keys( + self, + cls: Union[Type[MultiDict[int]], Type[CIMultiDict[int]]] + ) -> None: + md = cls(a=1, b=2) + it = iter(md.keys()) + assert it.__length_hint__() == 2 # type: ignore[attr-defined] + + def test_iter_length_hint_items( + self, + cls: Union[Type[MultiDict[int]], Type[CIMultiDict[int]]] + ) -> None: + md = cls(a=1, b=2) + it = iter(md.items()) + assert it.__length_hint__() == 2 # type: ignore[attr-defined] + + def test_iter_length_hint_values( + self, + cls: Union[Type[MultiDict[int]], Type[CIMultiDict[int]]] + ) -> None: + md = cls(a=1, b=2) + it = iter(md.values()) + assert it.__length_hint__() == 2 # type: ignore[attr-defined] + + def test_ctor_list_arg_and_kwds( + self, + cls: Union[Type[MultiDict[int]], Type[CIMultiDict[int]]] + ) -> None: + arg = [("a", 1)] + obj = cls(arg, b=2) + assert list(obj.items()) == [("a", 1), ("b", 2)] + assert arg == [("a", 1)] + + def test_ctor_tuple_arg_and_kwds( + self, + cls: Union[Type[MultiDict[int]], Type[CIMultiDict[int]]] + ) -> None: + arg = (("a", 1),) + obj = cls(arg, b=2) + assert list(obj.items()) == [("a", 1), ("b", 2)] + assert arg == (("a", 1),) + + def test_ctor_deque_arg_and_kwds( + self, + cls: Union[Type[MultiDict[int]], Type[CIMultiDict[int]]] + ) -> None: + arg = deque([("a", 1)]) + obj = cls(arg, b=2) + assert list(obj.items()) == [("a", 1), ("b", 2)] + assert arg == deque([("a", 1)]) + + +class TestMultiDict(BaseMultiDictTest): + @pytest.fixture(params=["MultiDict", ("MultiDict", "MultiDictProxy")]) + def cls(self, request: Any, _multidict: Any) -> Any: + return chained_callable(_multidict, request.param) + + def test__repr__(self, cls: Type[MultiDict[str]]) -> None: + d = cls() + _cls = type(d) + + assert str(d) == "<%s()>" % _cls.__name__ + + d = cls([("key", "one"), ("key", "two")]) + + assert str(d) == "<%s('key': 'one', 'key': 'two')>" % _cls.__name__ + + def test_getall(self, cls: Type[MultiDict[str]]) -> None: + d = cls([("key", "value1")], key="value2") + + assert d != {"key": "value1"} + assert len(d) == 2 + + assert d.getall("key") == ["value1", "value2"] + + with pytest.raises(KeyError, match="some_key"): + d.getall("some_key") + + default = object() + assert d.getall("some_key", default) is default + + def test_preserve_stable_ordering( + self, cls: Type[MultiDict[Union[str, int]]] + ) -> None: + d = cls([("a", 1), ("b", "2"), ("a", 3)]) + s = "&".join("{}={}".format(k, v) for k, v in d.items()) + + assert s == "a=1&b=2&a=3" + + def test_get(self, cls: Type[MultiDict[int]]) -> None: + d = cls([("a", 1), ("a", 2)]) + assert d["a"] == 1 + + def test_items__repr__(self, cls: Type[MultiDict[str]]) -> None: + d = cls([("key", "value1")], key="value2") + expected = "_ItemsView('key': 'value1', 'key': 'value2')" + assert repr(d.items()) == expected + + def test_keys__repr__(self, cls: Type[MultiDict[str]]) -> None: + d = cls([("key", "value1")], key="value2") + assert repr(d.keys()) == "_KeysView('key', 'key')" + + def test_values__repr__(self, cls: Type[MultiDict[str]]) -> None: + d = cls([("key", "value1")], key="value2") + assert repr(d.values()) == "_ValuesView('value1', 'value2')" + + +class TestCIMultiDict(BaseMultiDictTest): + @pytest.fixture(params=["CIMultiDict", ("CIMultiDict", "CIMultiDictProxy")]) + def cls(self, request: Any, _multidict: Any) -> Any: + return chained_callable(_multidict, request.param) + + def test_basics(self, cls: Type[CIMultiDict[str]]) -> None: + d = cls([("KEY", "value1")], KEY="value2") + + assert d.getone("key") == "value1" + assert d.get("key") == "value1" + assert d.get("key2", "val") == "val" + assert d["key"] == "value1" + assert "key" in d + + with pytest.raises(KeyError, match="key2"): + d["key2"] + with pytest.raises(KeyError, match="key2"): + d.getone("key2") + + def test_getall(self, cls: Type[CIMultiDict[str]]) -> None: + d = cls([("KEY", "value1")], KEY="value2") + + assert not d == {"KEY": "value1"} + assert len(d) == 2 + + assert d.getall("key") == ["value1", "value2"] + + with pytest.raises(KeyError, match="some_key"): + d.getall("some_key") + + def test_get(self, cls: Type[CIMultiDict[int]]) -> None: + d = cls([("A", 1), ("a", 2)]) + assert 1 == d["a"] + + def test__repr__(self, cls: Type[CIMultiDict[str]]) -> None: + d = cls([("KEY", "value1")], key="value2") + _cls = type(d) + + expected = "<%s('KEY': 'value1', 'key': 'value2')>" % _cls.__name__ + assert str(d) == expected + + def test_items__repr__(self, cls: Type[CIMultiDict[str]]) -> None: + d = cls([("KEY", "value1")], key="value2") + expected = "_ItemsView('KEY': 'value1', 'key': 'value2')" + assert repr(d.items()) == expected + + def test_keys__repr__(self, cls: Type[CIMultiDict[str]]) -> None: + d = cls([("KEY", "value1")], key="value2") + assert repr(d.keys()) == "_KeysView('KEY', 'key')" + + def test_values__repr__(self, cls: Type[CIMultiDict[str]]) -> None: + d = cls([("KEY", "value1")], key="value2") + assert repr(d.values()) == "_ValuesView('value1', 'value2')" diff --git a/contrib/python/multidict/tests/test_mutable_multidict.py b/contrib/python/multidict/tests/test_mutable_multidict.py new file mode 100644 index 0000000000..3d4d16ac03 --- /dev/null +++ b/contrib/python/multidict/tests/test_mutable_multidict.py @@ -0,0 +1,510 @@ +import string +import sys + +import pytest + + +class TestMutableMultiDict: + @pytest.fixture + def cls(self, _multidict): + return _multidict.MultiDict + + @pytest.fixture + def proxy_cls(self, _multidict): + return _multidict.MultiDictProxy + + @pytest.fixture + def istr(self, _multidict): + return _multidict.istr + + def test_copy(self, cls): + d1 = cls(key="value", a="b") + + d2 = d1.copy() + assert d1 == d2 + assert d1 is not d2 + + def test__repr__(self, cls): + d = cls() + assert str(d) == "<%s()>" % cls.__name__ + + d = cls([("key", "one"), ("key", "two")]) + + expected = "<%s('key': 'one', 'key': 'two')>" % cls.__name__ + assert str(d) == expected + + def test_getall(self, cls): + d = cls([("key", "value1")], key="value2") + assert len(d) == 2 + + assert d.getall("key") == ["value1", "value2"] + + with pytest.raises(KeyError, match="some_key"): + d.getall("some_key") + + default = object() + assert d.getall("some_key", default) is default + + def test_add(self, cls): + d = cls() + + assert d == {} + d["key"] = "one" + assert d == {"key": "one"} + assert d.getall("key") == ["one"] + + d["key"] = "two" + assert d == {"key": "two"} + assert d.getall("key") == ["two"] + + d.add("key", "one") + assert 2 == len(d) + assert d.getall("key") == ["two", "one"] + + d.add("foo", "bar") + assert 3 == len(d) + assert d.getall("foo") == ["bar"] + + def test_extend(self, cls): + d = cls() + assert d == {} + + d.extend([("key", "one"), ("key", "two")], key=3, foo="bar") + assert d != {"key": "one", "foo": "bar"} + assert 4 == len(d) + itms = d.items() + # we can't guarantee order of kwargs + assert ("key", "one") in itms + assert ("key", "two") in itms + assert ("key", 3) in itms + assert ("foo", "bar") in itms + + other = cls(bar="baz") + assert other == {"bar": "baz"} + + d.extend(other) + assert ("bar", "baz") in d.items() + + d.extend({"foo": "moo"}) + assert ("foo", "moo") in d.items() + + d.extend() + assert 6 == len(d) + + with pytest.raises(TypeError): + d.extend("foo", "bar") + + def test_extend_from_proxy(self, cls, proxy_cls): + d = cls([("a", "a"), ("b", "b")]) + proxy = proxy_cls(d) + + d2 = cls() + d2.extend(proxy) + + assert [("a", "a"), ("b", "b")] == list(d2.items()) + + def test_clear(self, cls): + d = cls([("key", "one")], key="two", foo="bar") + + d.clear() + assert d == {} + assert list(d.items()) == [] + + def test_del(self, cls): + d = cls([("key", "one"), ("key", "two")], foo="bar") + assert list(d.keys()) == ["key", "key", "foo"] + + del d["key"] + assert d == {"foo": "bar"} + assert list(d.items()) == [("foo", "bar")] + + with pytest.raises(KeyError, match="key"): + del d["key"] + + def test_set_default(self, cls): + d = cls([("key", "one"), ("key", "two")], foo="bar") + assert "one" == d.setdefault("key", "three") + assert "three" == d.setdefault("otherkey", "three") + assert "otherkey" in d + assert "three" == d["otherkey"] + + def test_popitem(self, cls): + d = cls() + d.add("key", "val1") + d.add("key", "val2") + + assert ("key", "val1") == d.popitem() + assert [("key", "val2")] == list(d.items()) + + def test_popitem_empty_multidict(self, cls): + d = cls() + + with pytest.raises(KeyError): + d.popitem() + + def test_pop(self, cls): + d = cls() + d.add("key", "val1") + d.add("key", "val2") + + assert "val1" == d.pop("key") + assert {"key": "val2"} == d + + def test_pop2(self, cls): + d = cls() + d.add("key", "val1") + d.add("key2", "val2") + d.add("key", "val3") + + assert "val1" == d.pop("key") + assert [("key2", "val2"), ("key", "val3")] == list(d.items()) + + def test_pop_default(self, cls): + d = cls(other="val") + + assert "default" == d.pop("key", "default") + assert "other" in d + + def test_pop_raises(self, cls): + d = cls(other="val") + + with pytest.raises(KeyError, match="key"): + d.pop("key") + + assert "other" in d + + def test_replacement_order(self, cls): + d = cls() + d.add("key1", "val1") + d.add("key2", "val2") + d.add("key1", "val3") + d.add("key2", "val4") + + d["key1"] = "val" + + expected = [("key1", "val"), ("key2", "val2"), ("key2", "val4")] + + assert expected == list(d.items()) + + def test_nonstr_key(self, cls): + d = cls() + with pytest.raises(TypeError): + d[1] = "val" + + def test_istr_key(self, cls, istr): + d = cls() + d[istr("1")] = "val" + assert type(list(d.keys())[0]) is istr + + def test_str_derived_key(self, cls): + class A(str): + pass + + d = cls() + d[A("1")] = "val" + assert type(list(d.keys())[0]) is A + + def test_istr_key_add(self, cls, istr): + d = cls() + d.add(istr("1"), "val") + assert type(list(d.keys())[0]) is istr + + def test_str_derived_key_add(self, cls): + class A(str): + pass + + d = cls() + d.add(A("1"), "val") + assert type(list(d.keys())[0]) is A + + def test_popall(self, cls): + d = cls() + d.add("key1", "val1") + d.add("key2", "val2") + d.add("key1", "val3") + ret = d.popall("key1") + assert ["val1", "val3"] == ret + assert {"key2": "val2"} == d + + def test_popall_default(self, cls): + d = cls() + assert "val" == d.popall("key", "val") + + def test_popall_key_error(self, cls): + d = cls() + with pytest.raises(KeyError, match="key"): + d.popall("key") + + def test_large_multidict_resizing(self, cls): + SIZE = 1024 + d = cls() + for i in range(SIZE): + d["key" + str(i)] = i + + for i in range(SIZE - 1): + del d["key" + str(i)] + + assert {"key" + str(SIZE - 1): SIZE - 1} == d + + +class TestCIMutableMultiDict: + @pytest.fixture + def cls(self, _multidict): + return _multidict.CIMultiDict + + @pytest.fixture + def proxy_cls(self, _multidict): + return _multidict.CIMultiDictProxy + + @pytest.fixture + def istr(self, _multidict): + return _multidict.istr + + def test_getall(self, cls): + d = cls([("KEY", "value1")], KEY="value2") + + assert d != {"KEY": "value1"} + assert len(d) == 2 + + assert d.getall("key") == ["value1", "value2"] + + with pytest.raises(KeyError, match="some_key"): + d.getall("some_key") + + def test_ctor(self, cls): + d = cls(k1="v1") + assert "v1" == d["K1"] + assert ("k1", "v1") in d.items() + + def test_setitem(self, cls): + d = cls() + d["k1"] = "v1" + assert "v1" == d["K1"] + assert ("k1", "v1") in d.items() + + def test_delitem(self, cls): + d = cls() + d["k1"] = "v1" + assert "K1" in d + del d["k1"] + assert "K1" not in d + + def test_copy(self, cls): + d1 = cls(key="KEY", a="b") + + d2 = d1.copy() + assert d1 == d2 + assert d1.items() == d2.items() + assert d1 is not d2 + + def test__repr__(self, cls): + d = cls() + assert str(d) == "<%s()>" % cls.__name__ + + d = cls([("KEY", "one"), ("KEY", "two")]) + + expected = "<%s('KEY': 'one', 'KEY': 'two')>" % cls.__name__ + assert str(d) == expected + + def test_add(self, cls): + d = cls() + + assert d == {} + d["KEY"] = "one" + assert ("KEY", "one") in d.items() + assert d == cls({"Key": "one"}) + assert d.getall("key") == ["one"] + + d["KEY"] = "two" + assert ("KEY", "two") in d.items() + assert d == cls({"Key": "two"}) + assert d.getall("key") == ["two"] + + d.add("KEY", "one") + assert ("KEY", "one") in d.items() + assert 2 == len(d) + assert d.getall("key") == ["two", "one"] + + d.add("FOO", "bar") + assert ("FOO", "bar") in d.items() + assert 3 == len(d) + assert d.getall("foo") == ["bar"] + + d.add(key="test", value="test") + assert ("test", "test") in d.items() + assert 4 == len(d) + assert d.getall("test") == ["test"] + + def test_extend(self, cls): + d = cls() + assert d == {} + + d.extend([("KEY", "one"), ("key", "two")], key=3, foo="bar") + assert 4 == len(d) + itms = d.items() + # we can't guarantee order of kwargs + assert ("KEY", "one") in itms + assert ("key", "two") in itms + assert ("key", 3) in itms + assert ("foo", "bar") in itms + + other = cls(Bar="baz") + assert other == {"Bar": "baz"} + + d.extend(other) + assert ("Bar", "baz") in d.items() + assert "bar" in d + + d.extend({"Foo": "moo"}) + assert ("Foo", "moo") in d.items() + assert "foo" in d + + d.extend() + assert 6 == len(d) + + with pytest.raises(TypeError): + d.extend("foo", "bar") + + def test_extend_from_proxy(self, cls, proxy_cls): + d = cls([("a", "a"), ("b", "b")]) + proxy = proxy_cls(d) + + d2 = cls() + d2.extend(proxy) + + assert [("a", "a"), ("b", "b")] == list(d2.items()) + + def test_clear(self, cls): + d = cls([("KEY", "one")], key="two", foo="bar") + + d.clear() + assert d == {} + assert list(d.items()) == [] + + def test_del(self, cls): + d = cls([("KEY", "one"), ("key", "two")], foo="bar") + + del d["key"] + assert d == {"foo": "bar"} + assert list(d.items()) == [("foo", "bar")] + + with pytest.raises(KeyError, match="key"): + del d["key"] + + def test_set_default(self, cls): + d = cls([("KEY", "one"), ("key", "two")], foo="bar") + assert "one" == d.setdefault("key", "three") + assert "three" == d.setdefault("otherkey", "three") + assert "otherkey" in d + assert ("otherkey", "three") in d.items() + assert "three" == d["OTHERKEY"] + + def test_popitem(self, cls): + d = cls() + d.add("KEY", "val1") + d.add("key", "val2") + + pair = d.popitem() + assert ("KEY", "val1") == pair + assert isinstance(pair[0], str) + assert [("key", "val2")] == list(d.items()) + + def test_popitem_empty_multidict(self, cls): + d = cls() + + with pytest.raises(KeyError): + d.popitem() + + def test_pop(self, cls): + d = cls() + d.add("KEY", "val1") + d.add("key", "val2") + + assert "val1" == d.pop("KEY") + assert {"key": "val2"} == d + + def test_pop_lowercase(self, cls): + d = cls() + d.add("KEY", "val1") + d.add("key", "val2") + + assert "val1" == d.pop("key") + assert {"key": "val2"} == d + + def test_pop_default(self, cls): + d = cls(OTHER="val") + + assert "default" == d.pop("key", "default") + assert "other" in d + + def test_pop_raises(self, cls): + d = cls(OTHER="val") + + with pytest.raises(KeyError, match="KEY"): + d.pop("KEY") + + assert "other" in d + + def test_extend_with_istr(self, cls, istr): + us = istr("aBc") + d = cls() + + d.extend([(us, "val")]) + assert [("aBc", "val")] == list(d.items()) + + def test_copy_istr(self, cls, istr): + d = cls({istr("Foo"): "bar"}) + d2 = d.copy() + assert d == d2 + + def test_eq(self, cls): + d1 = cls(Key="val") + d2 = cls(KEY="val") + + assert d1 == d2 + + @pytest.mark.skipif( + sys.implementation.name == "pypy", + reason="getsizeof() is not implemented on PyPy", + ) + def test_sizeof(self, cls): + md = cls() + s1 = sys.getsizeof(md) + for i in string.ascii_lowercase: + for j in string.ascii_uppercase: + md[i + j] = i + j + # multidict should be resized + s2 = sys.getsizeof(md) + assert s2 > s1 + + @pytest.mark.skipif( + sys.implementation.name == "pypy", + reason="getsizeof() is not implemented on PyPy", + ) + def test_min_sizeof(self, cls): + md = cls() + assert sys.getsizeof(md) < 1024 + + def test_issue_620_items(self, cls): + # https://github.com/aio-libs/multidict/issues/620 + d = cls({"a": "123, 456", "b": "789"}) + before_mutation_items = d.items() + d["c"] = "000" + # This causes an error on pypy. + list(before_mutation_items) + + def test_issue_620_keys(self, cls): + # https://github.com/aio-libs/multidict/issues/620 + d = cls({"a": "123, 456", "b": "789"}) + before_mutation_keys = d.keys() + d["c"] = "000" + # This causes an error on pypy. + list(before_mutation_keys) + + def test_issue_620_values(self, cls): + # https://github.com/aio-libs/multidict/issues/620 + d = cls({"a": "123, 456", "b": "789"}) + before_mutation_values = d.values() + d["c"] = "000" + # This causes an error on pypy. + list(before_mutation_values) diff --git a/contrib/python/multidict/tests/test_mypy.py b/contrib/python/multidict/tests/test_mypy.py new file mode 100644 index 0000000000..62bb62e52d --- /dev/null +++ b/contrib/python/multidict/tests/test_mypy.py @@ -0,0 +1,278 @@ +# This file doesn't use test parametrization because mypy doesn't nothing about it. +# Concrete types are required + +import multidict + + +def test_classes_not_abstract() -> None: + d1: multidict.MultiDict[str] = multidict.MultiDict({"a": "b"}) + d2: multidict.CIMultiDict[str] = multidict.CIMultiDict({"a": "b"}) + + multidict.MultiDictProxy(d1) + multidict.CIMultiDictProxy(d2) + + +def test_getitem() -> None: + d1: multidict.MultiDict[str] = multidict.MultiDict({"a": "b"}) + d2: multidict.CIMultiDict[str] = multidict.CIMultiDict({"a": "b"}) + + d3 = multidict.MultiDictProxy(d1) + d4 = multidict.CIMultiDictProxy(d2) + + key = multidict.istr("a") + + assert d1["a"] == "b" + assert d2["a"] == "b" + assert d3["a"] == "b" + assert d4["a"] == "b" + + assert d1[key] == "b" + assert d2[key] == "b" + assert d3[key] == "b" + assert d4[key] == "b" + + +def test_get() -> None: + d1: multidict.MultiDict[str] = multidict.MultiDict({"a": "b"}) + d2: multidict.CIMultiDict[str] = multidict.CIMultiDict({"a": "b"}) + + d3 = multidict.MultiDictProxy(d1) + d4 = multidict.CIMultiDictProxy(d2) + + key = multidict.istr("a") + + assert d1.get("a") == "b" + assert d2.get("a") == "b" + assert d3.get("a") == "b" + assert d4.get("a") == "b" + + assert d1.get(key) == "b" + assert d2.get(key) == "b" + assert d3.get(key) == "b" + assert d4.get(key) == "b" + + +def test_get_default() -> None: + d1: multidict.MultiDict[str] = multidict.MultiDict({"a": "b"}) + d2: multidict.CIMultiDict[str] = multidict.CIMultiDict({"a": "b"}) + + d3 = multidict.MultiDictProxy(d1) + d4 = multidict.CIMultiDictProxy(d2) + + key = multidict.istr("b") + + assert d1.get("b", "d") == "d" + assert d2.get("b", "d") == "d" + assert d3.get("b", "d") == "d" + assert d4.get("b", "d") == "d" + + assert d1.get(key, "d") == "d" + assert d2.get(key, "d") == "d" + assert d3.get(key, "d") == "d" + assert d4.get(key, "d") == "d" + + +def test_getone() -> None: + d1: multidict.MultiDict[str] = multidict.MultiDict({"a": "b"}) + d2: multidict.CIMultiDict[str] = multidict.CIMultiDict({"a": "b"}) + + d3 = multidict.MultiDictProxy(d1) + d4 = multidict.CIMultiDictProxy(d2) + + key = multidict.istr("a") + + assert d1.getone("a") == "b" + assert d2.getone("a") == "b" + assert d3.getone("a") == "b" + assert d4.getone("a") == "b" + + assert d1.getone(key) == "b" + assert d2.getone(key) == "b" + assert d3.getone(key) == "b" + assert d4.getone(key) == "b" + + +def test_getone_default() -> None: + d1: multidict.MultiDict[str] = multidict.MultiDict({"a": "b"}) + d2: multidict.CIMultiDict[str] = multidict.CIMultiDict({"a": "b"}) + + d3 = multidict.MultiDictProxy(d1) + d4 = multidict.CIMultiDictProxy(d2) + + key = multidict.istr("b") + + assert d1.getone("b", 1) == 1 + assert d2.getone("b", 1) == 1 + assert d3.getone("b", 1) == 1 + assert d4.getone("b", 1) == 1 + + assert d1.getone(key, 1) == 1 + assert d2.getone(key, 1) == 1 + assert d3.getone(key, 1) == 1 + assert d4.getone(key, 1) == 1 + + +def test_getall() -> None: + d1: multidict.MultiDict[str] = multidict.MultiDict({"a": "b"}) + d2: multidict.CIMultiDict[str] = multidict.CIMultiDict({"a": "b"}) + + d3 = multidict.MultiDictProxy(d1) + d4 = multidict.CIMultiDictProxy(d2) + + key = multidict.istr("a") + + assert d1.getall("a") == ["b"] + assert d2.getall("a") == ["b"] + assert d3.getall("a") == ["b"] + assert d4.getall("a") == ["b"] + + assert d1.getall(key) == ["b"] + assert d2.getall(key) == ["b"] + assert d3.getall(key) == ["b"] + assert d4.getall(key) == ["b"] + + +def test_getall_default() -> None: + d1: multidict.MultiDict[str] = multidict.MultiDict({"a": "b"}) + d2: multidict.CIMultiDict[str] = multidict.CIMultiDict({"a": "b"}) + + d3 = multidict.MultiDictProxy(d1) + d4 = multidict.CIMultiDictProxy(d2) + + key = multidict.istr("b") + + assert d1.getall("b", 1) == 1 + assert d2.getall("b", 1) == 1 + assert d3.getall("b", 1) == 1 + assert d4.getall("b", 1) == 1 + + assert d1.getall(key, 1) == 1 + assert d2.getall(key, 1) == 1 + assert d3.getall(key, 1) == 1 + assert d4.getall(key, 1) == 1 + + +def test_copy() -> None: + d1: multidict.MultiDict[str] = multidict.MultiDict({"a": "b"}) + d2: multidict.CIMultiDict[str] = multidict.CIMultiDict({"a": "b"}) + + d3 = multidict.MultiDictProxy(d1) + d4 = multidict.CIMultiDictProxy(d2) + + assert d1.copy() == d1 + assert d2.copy() == d2 + assert d3.copy() == d1 + assert d4.copy() == d2 + + +def test_iter() -> None: + d1: multidict.MultiDict[str] = multidict.MultiDict({"a": "b"}) + d2: multidict.CIMultiDict[str] = multidict.CIMultiDict({"a": "b"}) + + d3 = multidict.MultiDictProxy(d1) + d4 = multidict.CIMultiDictProxy(d2) + + for i in d1: + i.lower() # str-specific class + for i in d2: + i.lower() # str-specific class + for i in d3: + i.lower() # str-specific class + for i in d4: + i.lower() # str-specific class + + +def test_setitem() -> None: + d1: multidict.MultiDict[str] = multidict.MultiDict({"a": "b"}) + d2: multidict.CIMultiDict[str] = multidict.CIMultiDict({"a": "b"}) + + key = multidict.istr("a") + + d1["a"] = "b" + d2["a"] = "b" + + d1[key] = "b" + d2[key] = "b" + +def test_delitem() -> None: + d1: multidict.MultiDict[str] = multidict.MultiDict({"a": "b"}) + d2: multidict.CIMultiDict[str] = multidict.CIMultiDict({"a": "b"}) + + del d1["a"] + del d2["a"] + + key = multidict.istr("a") + + d3: multidict.MultiDict[str] = multidict.MultiDict({"a": "b"}) + d4: multidict.CIMultiDict[str] = multidict.CIMultiDict({"a": "b"}) + + del d3[key] + del d4[key] + + +def test_additem() -> None: + d1: multidict.MultiDict[str] = multidict.MultiDict({"a": "b"}) + d2: multidict.CIMultiDict[str] = multidict.CIMultiDict({"a": "b"}) + + key = multidict.istr("a") + + d1.add("a", "b") + d2.add("a", "b") + + d1.add(key, "b") + d2.add(key, "b") + + +def test_extend_mapping() -> None: + d1: multidict.MultiDict[str] = multidict.MultiDict({"a": "b"}) + d2: multidict.CIMultiDict[str] = multidict.CIMultiDict({"a": "b"}) + + key = multidict.istr("a") + + d1.extend({"a": "b"}) + d2.extend({"a": "b"}) + + d1.extend({key: "b"}) + d2.extend({key: "b"}) + + +def test_update_mapping() -> None: + d1: multidict.MultiDict[str] = multidict.MultiDict({"a": "b"}) + d2: multidict.CIMultiDict[str] = multidict.CIMultiDict({"a": "b"}) + + key = multidict.istr("a") + + d1.update({"a": "b"}) + d2.update({"a": "b"}) + + d1.update({key: "b"}) + d2.update({key: "b"}) + +def test_popone() -> None: + d1: multidict.MultiDict[str] = multidict.MultiDict({"a": "b"}) + d2: multidict.CIMultiDict[str] = multidict.CIMultiDict({"a": "b"}) + + assert d1.popone("a") == "b" + assert d2.popone("a") == "b" + + key = multidict.istr("a") + d1 = multidict.MultiDict({"a": "b"}) + d2 = multidict.CIMultiDict({"a": "b"}) + + assert d1.popone(key) == "b" + assert d2.popone(key) == "b" + + +def test_popall() -> None: + d1: multidict.MultiDict[str] = multidict.MultiDict({"a": "b"}) + d2: multidict.CIMultiDict[str] = multidict.CIMultiDict({"a": "b"}) + + assert d1.popall("a") == ["b"] + assert d2.popall("a") == ["b"] + + key = multidict.istr("a") + d1 = multidict.MultiDict({"a": "b"}) + d2 = multidict.CIMultiDict({"a": "b"}) + + assert d1.popall(key) == ["b"] + assert d2.popall(key) == ["b"] diff --git a/contrib/python/multidict/tests/test_pickle.py b/contrib/python/multidict/tests/test_pickle.py new file mode 100644 index 0000000000..ce7383684f --- /dev/null +++ b/contrib/python/multidict/tests/test_pickle.py @@ -0,0 +1,82 @@ +import pickle +from pathlib import Path + +import pytest + +from multidict._compat import USE_EXTENSIONS +from multidict._multidict_py import CIMultiDict as PyCIMultiDict +from multidict._multidict_py import CIMultiDictProxy as PyCIMultiDictProxy +from multidict._multidict_py import MultiDict as PyMultiDict # noqa: E402 +from multidict._multidict_py import MultiDictProxy as PyMultiDictProxy + +import yatest.common + +if USE_EXTENSIONS: + from multidict._multidict import ( # type: ignore + CIMultiDict, + CIMultiDictProxy, + MultiDict, + MultiDictProxy, + ) + + +here = Path(yatest.common.test_source_path()).resolve() + + +@pytest.fixture( + params=(["MultiDict", "CIMultiDict"] if USE_EXTENSIONS else []) + + ["PyMultiDict", "PyCIMultiDict"] +) +def cls_name(request): + return request.param + + +@pytest.fixture( + params=([MultiDict, CIMultiDict] if USE_EXTENSIONS else []) + + [PyMultiDict, PyCIMultiDict], + ids=(["MultiDict", "CIMultiDict"] if USE_EXTENSIONS else []) + + ["PyMultiDict", "PyCIMultiDict"], +) +def cls(request): + return request.param + + +@pytest.fixture( + params=( + [(MultiDictProxy, MultiDict), (CIMultiDictProxy, CIMultiDict)] + if USE_EXTENSIONS + else [] + ) + + [(PyMultiDictProxy, PyMultiDict), (PyCIMultiDictProxy, PyCIMultiDict)], + ids=(["MultiDictProxy", "CIMultiDictProxy"] if USE_EXTENSIONS else []) + + ["PyMultiDictProxy", "PyCIMultiDictProxy"], +) +def proxy_classes(request): + return request.param + + +def test_pickle(cls, pickle_protocol): + d = cls([("a", 1), ("a", 2)]) + pbytes = pickle.dumps(d, pickle_protocol) + obj = pickle.loads(pbytes) + assert d == obj + assert isinstance(obj, cls) + + +def test_pickle_proxy(proxy_classes): + proxy_cls, dict_cls = proxy_classes + d = dict_cls([("a", 1), ("a", 2)]) + proxy = proxy_cls(d) + with pytest.raises(TypeError): + pickle.dumps(proxy) + + +def test_load_from_file(pickle_protocol, cls_name): + cls = globals()[cls_name] + d = cls([("a", 1), ("a", 2)]) + fname = "{}.pickle.{}".format(cls_name.lower(), pickle_protocol) + p = here / fname + with p.open("rb") as f: + obj = pickle.load(f) + assert d == obj + assert isinstance(obj, cls) diff --git a/contrib/python/multidict/tests/test_types.py b/contrib/python/multidict/tests/test_types.py new file mode 100644 index 0000000000..3ae2cbb844 --- /dev/null +++ b/contrib/python/multidict/tests/test_types.py @@ -0,0 +1,109 @@ +import sys +import types + +import pytest + + +def test_proxies(_multidict): + assert issubclass(_multidict.CIMultiDictProxy, _multidict.MultiDictProxy) + + +def test_dicts(_multidict): + assert issubclass(_multidict.CIMultiDict, _multidict.MultiDict) + + +def test_proxy_not_inherited_from_dict(_multidict): + assert not issubclass(_multidict.MultiDictProxy, _multidict.MultiDict) + + +def test_dict_not_inherited_from_proxy(_multidict): + assert not issubclass(_multidict.MultiDict, _multidict.MultiDictProxy) + + +def test_multidict_proxy_copy_type(_multidict): + d = _multidict.MultiDict(key="val") + p = _multidict.MultiDictProxy(d) + assert isinstance(p.copy(), _multidict.MultiDict) + + +def test_cimultidict_proxy_copy_type(_multidict): + d = _multidict.CIMultiDict(key="val") + p = _multidict.CIMultiDictProxy(d) + assert isinstance(p.copy(), _multidict.CIMultiDict) + + +def test_create_multidict_proxy_from_nonmultidict(_multidict): + with pytest.raises(TypeError): + _multidict.MultiDictProxy({}) + + +def test_create_multidict_proxy_from_cimultidict(_multidict): + d = _multidict.CIMultiDict(key="val") + p = _multidict.MultiDictProxy(d) + assert p == d + + +def test_create_multidict_proxy_from_multidict_proxy_from_mdict(_multidict): + d = _multidict.MultiDict(key="val") + p = _multidict.MultiDictProxy(d) + assert p == d + p2 = _multidict.MultiDictProxy(p) + assert p2 == p + + +def test_create_cimultidict_proxy_from_cimultidict_proxy_from_ci(_multidict): + d = _multidict.CIMultiDict(key="val") + p = _multidict.CIMultiDictProxy(d) + assert p == d + p2 = _multidict.CIMultiDictProxy(p) + assert p2 == p + + +def test_create_cimultidict_proxy_from_nonmultidict(_multidict): + with pytest.raises( + TypeError, + match=( + "ctor requires CIMultiDict or CIMultiDictProxy instance, " + "not <class 'dict'>" + ), + ): + _multidict.CIMultiDictProxy({}) + + +def test_create_ci_multidict_proxy_from_multidict(_multidict): + d = _multidict.MultiDict(key="val") + with pytest.raises( + TypeError, + match=( + "ctor requires CIMultiDict or CIMultiDictProxy instance, " + "not <class 'multidict._multidict.*.MultiDict'>" + ), + ): + _multidict.CIMultiDictProxy(d) + + +@pytest.mark.skipif( + sys.version_info >= (3, 9), reason="Python 3.9 uses GenericAlias which is different" +) +def test_generic_exists(_multidict) -> None: + assert _multidict.MultiDict[int] is _multidict.MultiDict + assert _multidict.MultiDictProxy[int] is _multidict.MultiDictProxy + assert _multidict.CIMultiDict[int] is _multidict.CIMultiDict + assert _multidict.CIMultiDictProxy[int] is _multidict.CIMultiDictProxy + + +@pytest.mark.skipif( + sys.version_info < (3, 9), reason="Python 3.9 is required for GenericAlias" +) +def test_generic_alias(_multidict) -> None: + + assert _multidict.MultiDict[int] == types.GenericAlias(_multidict.MultiDict, (int,)) + assert _multidict.MultiDictProxy[int] == types.GenericAlias( + _multidict.MultiDictProxy, (int,) + ) + assert _multidict.CIMultiDict[int] == types.GenericAlias( + _multidict.CIMultiDict, (int,) + ) + assert _multidict.CIMultiDictProxy[int] == types.GenericAlias( + _multidict.CIMultiDictProxy, (int,) + ) diff --git a/contrib/python/multidict/tests/test_update.py b/contrib/python/multidict/tests/test_update.py new file mode 100644 index 0000000000..4bacdbce77 --- /dev/null +++ b/contrib/python/multidict/tests/test_update.py @@ -0,0 +1,147 @@ +from collections import deque + +import pytest + +from multidict._compat import USE_EXTENSIONS +from multidict._multidict_py import CIMultiDict as PyCIMultiDict +from multidict._multidict_py import MultiDict as PyMultiDict # noqa: E402 + +if USE_EXTENSIONS: + from multidict._multidict import CIMultiDict, MultiDict # type: ignore + + +@pytest.fixture( + params=([MultiDict, CIMultiDict] if USE_EXTENSIONS else []) + + [PyMultiDict, PyCIMultiDict], + ids=(["MultiDict", "CIMultiDict"] if USE_EXTENSIONS else []) + + ["PyMultiDict", "PyCIMultiDict"], +) +def cls(request): + return request.param + + +@pytest.fixture +def md_cls(_multidict): + return _multidict.MultiDict + + +@pytest.fixture +def ci_md_cls(_multidict): + return _multidict.CIMultiDict + + +@pytest.fixture +def istr(_multidict): + return _multidict.istr + + +def test_update_replace(cls): + obj1 = cls([("a", 1), ("b", 2), ("a", 3), ("c", 10)]) + obj2 = cls([("a", 4), ("b", 5), ("a", 6)]) + obj1.update(obj2) + expected = [("a", 4), ("b", 5), ("a", 6), ("c", 10)] + assert list(obj1.items()) == expected + + +def test_update_append(cls): + obj1 = cls([("a", 1), ("b", 2), ("a", 3), ("c", 10)]) + obj2 = cls([("a", 4), ("a", 5), ("a", 6)]) + obj1.update(obj2) + expected = [("a", 4), ("b", 2), ("a", 5), ("c", 10), ("a", 6)] + assert list(obj1.items()) == expected + + +def test_update_remove(cls): + obj1 = cls([("a", 1), ("b", 2), ("a", 3), ("c", 10)]) + obj2 = cls([("a", 4)]) + obj1.update(obj2) + expected = [("a", 4), ("b", 2), ("c", 10)] + assert list(obj1.items()) == expected + + +def test_update_replace_seq(cls): + obj1 = cls([("a", 1), ("b", 2), ("a", 3), ("c", 10)]) + obj2 = [("a", 4), ("b", 5), ("a", 6)] + obj1.update(obj2) + expected = [("a", 4), ("b", 5), ("a", 6), ("c", 10)] + assert list(obj1.items()) == expected + + +def test_update_replace_seq2(cls): + obj1 = cls([("a", 1), ("b", 2), ("a", 3), ("c", 10)]) + obj1.update([("a", 4)], b=5, a=6) + expected = [("a", 4), ("b", 5), ("a", 6), ("c", 10)] + assert list(obj1.items()) == expected + + +def test_update_append_seq(cls): + obj1 = cls([("a", 1), ("b", 2), ("a", 3), ("c", 10)]) + obj2 = [("a", 4), ("a", 5), ("a", 6)] + obj1.update(obj2) + expected = [("a", 4), ("b", 2), ("a", 5), ("c", 10), ("a", 6)] + assert list(obj1.items()) == expected + + +def test_update_remove_seq(cls): + obj1 = cls([("a", 1), ("b", 2), ("a", 3), ("c", 10)]) + obj2 = [("a", 4)] + obj1.update(obj2) + expected = [("a", 4), ("b", 2), ("c", 10)] + assert list(obj1.items()) == expected + + +def test_update_md(md_cls): + d = md_cls() + d.add("key", "val1") + d.add("key", "val2") + d.add("key2", "val3") + + d.update(key="val") + + assert [("key", "val"), ("key2", "val3")] == list(d.items()) + + +def test_update_istr_ci_md(ci_md_cls, istr): + d = ci_md_cls() + d.add(istr("KEY"), "val1") + d.add("key", "val2") + d.add("key2", "val3") + + d.update({istr("key"): "val"}) + + assert [("key", "val"), ("key2", "val3")] == list(d.items()) + + +def test_update_ci_md(ci_md_cls): + d = ci_md_cls() + d.add("KEY", "val1") + d.add("key", "val2") + d.add("key2", "val3") + + d.update(Key="val") + + assert [("Key", "val"), ("key2", "val3")] == list(d.items()) + + +def test_update_list_arg_and_kwds(cls): + obj = cls() + arg = [("a", 1)] + obj.update(arg, b=2) + assert list(obj.items()) == [("a", 1), ("b", 2)] + assert arg == [("a", 1)] + + +def test_update_tuple_arg_and_kwds(cls): + obj = cls() + arg = (("a", 1),) + obj.update(arg, b=2) + assert list(obj.items()) == [("a", 1), ("b", 2)] + assert arg == (("a", 1),) + + +def test_update_deque_arg_and_kwds(cls): + obj = cls() + arg = deque([("a", 1)]) + obj.update(arg, b=2) + assert list(obj.items()) == [("a", 1), ("b", 2)] + assert arg == deque([("a", 1)]) diff --git a/contrib/python/multidict/tests/test_version.py b/contrib/python/multidict/tests/test_version.py new file mode 100644 index 0000000000..067d6210ce --- /dev/null +++ b/contrib/python/multidict/tests/test_version.py @@ -0,0 +1,199 @@ +from typing import Type + +import pytest + +from multidict import MultiMapping +from multidict._compat import USE_EXTENSIONS +from multidict._multidict_py import CIMultiDict as _CIMultiDict +from multidict._multidict_py import MultiDict as _MultiDict # noqa: E402 +from multidict._multidict_py import getversion as _getversion + +if USE_EXTENSIONS: + from multidict._multidict import ( # type: ignore + CIMultiDict, + MultiDict, + getversion, + ) + + +class VersionMixin: + cls: Type[MultiMapping[str]] + + def getver(self, md): + raise NotImplementedError + + def test_getversion_bad_param(self): + with pytest.raises(TypeError): + self.getver(1) + + def test_ctor(self): + m1 = self.cls() + v1 = self.getver(m1) + m2 = self.cls() + v2 = self.getver(m2) + assert v1 != v2 + + def test_add(self): + m = self.cls() + v = self.getver(m) + m.add("key", "val") + assert self.getver(m) > v + + def test_delitem(self): + m = self.cls() + m.add("key", "val") + v = self.getver(m) + del m["key"] + assert self.getver(m) > v + + def test_delitem_not_found(self): + m = self.cls() + m.add("key", "val") + v = self.getver(m) + with pytest.raises(KeyError): + del m["notfound"] + assert self.getver(m) == v + + def test_setitem(self): + m = self.cls() + m.add("key", "val") + v = self.getver(m) + m["key"] = "val2" + assert self.getver(m) > v + + def test_setitem_not_found(self): + m = self.cls() + m.add("key", "val") + v = self.getver(m) + m["notfound"] = "val2" + assert self.getver(m) > v + + def test_clear(self): + m = self.cls() + m.add("key", "val") + v = self.getver(m) + m.clear() + assert self.getver(m) > v + + def test_setdefault(self): + m = self.cls() + m.add("key", "val") + v = self.getver(m) + m.setdefault("key2", "val2") + assert self.getver(m) > v + + def test_popone(self): + m = self.cls() + m.add("key", "val") + v = self.getver(m) + m.popone("key") + assert self.getver(m) > v + + def test_popone_default(self): + m = self.cls() + m.add("key", "val") + v = self.getver(m) + m.popone("key2", "default") + assert self.getver(m) == v + + def test_popone_key_error(self): + m = self.cls() + m.add("key", "val") + v = self.getver(m) + with pytest.raises(KeyError): + m.popone("key2") + assert self.getver(m) == v + + def test_pop(self): + m = self.cls() + m.add("key", "val") + v = self.getver(m) + m.pop("key") + assert self.getver(m) > v + + def test_pop_default(self): + m = self.cls() + m.add("key", "val") + v = self.getver(m) + m.pop("key2", "default") + assert self.getver(m) == v + + def test_pop_key_error(self): + m = self.cls() + m.add("key", "val") + v = self.getver(m) + with pytest.raises(KeyError): + m.pop("key2") + assert self.getver(m) == v + + def test_popall(self): + m = self.cls() + m.add("key", "val") + v = self.getver(m) + m.popall("key") + assert self.getver(m) > v + + def test_popall_default(self): + m = self.cls() + m.add("key", "val") + v = self.getver(m) + m.popall("key2", "default") + assert self.getver(m) == v + + def test_popall_key_error(self): + m = self.cls() + m.add("key", "val") + v = self.getver(m) + with pytest.raises(KeyError): + m.popall("key2") + assert self.getver(m) == v + + def test_popitem(self): + m = self.cls() + m.add("key", "val") + v = self.getver(m) + m.popitem() + assert self.getver(m) > v + + def test_popitem_key_error(self): + m = self.cls() + v = self.getver(m) + with pytest.raises(KeyError): + m.popitem() + assert self.getver(m) == v + + +if USE_EXTENSIONS: + + class TestMultiDict(VersionMixin): + + cls = MultiDict + + def getver(self, md): + return getversion(md) + + +if USE_EXTENSIONS: + + class TestCIMultiDict(VersionMixin): + + cls = CIMultiDict + + def getver(self, md): + return getversion(md) + + +class TestPyMultiDict(VersionMixin): + + cls = _MultiDict # type: ignore[assignment] + + def getver(self, md): + return _getversion(md) + + +class TestPyCIMultiDict(VersionMixin): + + cls = _CIMultiDict # type: ignore[assignment] + + def getver(self, md): + return _getversion(md) diff --git a/contrib/python/multidict/tests/ya.make b/contrib/python/multidict/tests/ya.make new file mode 100644 index 0000000000..f9d823fb92 --- /dev/null +++ b/contrib/python/multidict/tests/ya.make @@ -0,0 +1,29 @@ +PY3TEST() + +PEERDIR( + contrib/python/multidict +) + +TEST_SRCS( + conftest.py + test_abc.py + test_copy.py + test_guard.py + test_istr.py + test_multidict.py + test_mutable_multidict.py + test_mypy.py + test_pickle.py + test_types.py + test_update.py + test_version.py +) + +DATA( + arcadia/contrib/python/multidict/tests +) + + +NO_LINT() + +END() diff --git a/contrib/python/multidict/ya.make b/contrib/python/multidict/ya.make new file mode 100644 index 0000000000..bdba18c601 --- /dev/null +++ b/contrib/python/multidict/ya.make @@ -0,0 +1,46 @@ +# Generated by devtools/yamaker (pypi). + +PY3_LIBRARY() + +VERSION(6.0.4) + +LICENSE(Apache-2.0) + +ADDINCL( + contrib/python/multidict/multidict +) + +NO_COMPILER_WARNINGS() + +NO_LINT() + +SRCS( + multidict/_multidict.c +) + +PY_REGISTER( + multidict._multidict +) + +PY_SRCS( + TOP_LEVEL + multidict/__init__.py + multidict/__init__.pyi + multidict/_abc.py + multidict/_compat.py + multidict/_multidict_base.py + multidict/_multidict_py.py +) + +RESOURCE_FILES( + PREFIX contrib/python/multidict/ + .dist-info/METADATA + .dist-info/top_level.txt + multidict/py.typed +) + +END() + +RECURSE_FOR_TESTS( + tests +) diff --git a/contrib/python/yarl/.dist-info/METADATA b/contrib/python/yarl/.dist-info/METADATA new file mode 100644 index 0000000000..8585ab738d --- /dev/null +++ b/contrib/python/yarl/.dist-info/METADATA @@ -0,0 +1,1010 @@ +Metadata-Version: 2.1 +Name: yarl +Version: 1.9.3 +Summary: Yet another URL library +Home-page: https://github.com/aio-libs/yarl +Author: Andrew Svetlov +Author-email: andrew.svetlov@gmail.com +Maintainer: aiohttp team <team@aiohttp.org> +Maintainer-email: team@aiohttp.org +License: Apache-2.0 +Project-URL: Chat: Matrix, https://matrix.to/#/#aio-libs:matrix.org +Project-URL: Chat: Matrix Space, https://matrix.to/#/#aio-libs-space:matrix.org +Project-URL: CI: GitHub Workflows, https://github.com/aio-libs/yarl/actions?query=branch:master +Project-URL: Code of Conduct, https://github.com/aio-libs/.github/blob/master/CODE_OF_CONDUCT.md +Project-URL: Coverage: codecov, https://codecov.io/github/aio-libs/yarl +Project-URL: Docs: Changelog, https://yarl.aio-libs.org/en/latest/changes/ +Project-URL: Docs: RTD, https://yarl.aio-libs.org +Project-URL: GitHub: issues, https://github.com/aio-libs/yarl/issues +Project-URL: GitHub: repo, https://github.com/aio-libs/yarl +Keywords: cython,cext,yarl +Classifier: Development Status :: 5 - Production/Stable +Classifier: Intended Audience :: Developers +Classifier: License :: OSI Approved :: Apache Software License +Classifier: Programming Language :: Cython +Classifier: Programming Language :: Python +Classifier: Programming Language :: Python :: 3 +Classifier: Programming Language :: Python :: 3.7 +Classifier: Programming Language :: Python :: 3.8 +Classifier: Programming Language :: Python :: 3.9 +Classifier: Programming Language :: Python :: 3.10 +Classifier: Programming Language :: Python :: 3.11 +Classifier: Programming Language :: Python :: 3.12 +Classifier: Topic :: Internet :: WWW/HTTP +Classifier: Topic :: Software Development :: Libraries :: Python Modules +Requires-Python: >=3.7 +Description-Content-Type: text/x-rst +License-File: LICENSE +License-File: NOTICE +Requires-Dist: idna >=2.0 +Requires-Dist: multidict >=4.0 +Requires-Dist: typing-extensions >=3.7.4 ; python_version < "3.8" + +yarl +==== + +The module provides handy URL class for URL parsing and changing. + +.. image:: https://github.com/aio-libs/yarl/workflows/CI/badge.svg + :target: https://github.com/aio-libs/yarl/actions?query=workflow%3ACI + :align: right + +.. image:: https://codecov.io/gh/aio-libs/yarl/branch/master/graph/badge.svg + :target: https://codecov.io/gh/aio-libs/yarl + +.. image:: https://badge.fury.io/py/yarl.svg + :target: https://badge.fury.io/py/yarl + + +.. image:: https://readthedocs.org/projects/yarl/badge/?version=latest + :target: https://yarl.aio-libs.org + + +.. image:: https://img.shields.io/pypi/pyversions/yarl.svg + :target: https://pypi.python.org/pypi/yarl + +.. image:: https://img.shields.io/matrix/aio-libs:matrix.org?label=Discuss%20on%20Matrix%20at%20%23aio-libs%3Amatrix.org&logo=matrix&server_fqdn=matrix.org&style=flat + :target: https://matrix.to/#/%23aio-libs:matrix.org + :alt: Matrix Room — #aio-libs:matrix.org + +.. image:: https://img.shields.io/matrix/aio-libs-space:matrix.org?label=Discuss%20on%20Matrix%20at%20%23aio-libs-space%3Amatrix.org&logo=matrix&server_fqdn=matrix.org&style=flat + :target: https://matrix.to/#/%23aio-libs-space:matrix.org + :alt: Matrix Space — #aio-libs-space:matrix.org + +Introduction +------------ + +Url is constructed from ``str``: + +.. code-block:: pycon + + >>> from yarl import URL + >>> url = URL('https://www.python.org/~guido?arg=1#frag') + >>> url + URL('https://www.python.org/~guido?arg=1#frag') + +All url parts: *scheme*, *user*, *password*, *host*, *port*, *path*, +*query* and *fragment* are accessible by properties: + +.. code-block:: pycon + + >>> url.scheme + 'https' + >>> url.host + 'www.python.org' + >>> url.path + '/~guido' + >>> url.query_string + 'arg=1' + >>> url.query + <MultiDictProxy('arg': '1')> + >>> url.fragment + 'frag' + +All url manipulations produce a new url object: + +.. code-block:: pycon + + >>> url = URL('https://www.python.org') + >>> url / 'foo' / 'bar' + URL('https://www.python.org/foo/bar') + >>> url / 'foo' % {'bar': 'baz'} + URL('https://www.python.org/foo?bar=baz') + +Strings passed to constructor and modification methods are +automatically encoded giving canonical representation as result: + +.. code-block:: pycon + + >>> url = URL('https://www.python.org/путь') + >>> url + URL('https://www.python.org/%D0%BF%D1%83%D1%82%D1%8C') + +Regular properties are *percent-decoded*, use ``raw_`` versions for +getting *encoded* strings: + +.. code-block:: pycon + + >>> url.path + '/путь' + + >>> url.raw_path + '/%D0%BF%D1%83%D1%82%D1%8C' + +Human readable representation of URL is available as ``.human_repr()``: + +.. code-block:: pycon + + >>> url.human_repr() + 'https://www.python.org/путь' + +For full documentation please read https://yarl.aio-libs.org. + + +Installation +------------ + +:: + + $ pip install yarl + +The library is Python 3 only! + +PyPI contains binary wheels for Linux, Windows and MacOS. If you want to install +``yarl`` on another operating system (like *Alpine Linux*, which is not +manylinux-compliant because of the missing glibc and therefore, cannot be +used with our wheels) the the tarball will be used to compile the library from +the source code. It requires a C compiler and and Python headers installed. + +To skip the compilation you must explicitly opt-in by using a PEP 517 +configuration setting ``--pure-python``, or setting the ``YARL_NO_EXTENSIONS`` +environment variable to a non-empty value, e.g.: + +.. code-block:: console + + $ pip install yarl --config-settings=--pure-python= + +Please note that the pure-Python (uncompiled) version is much slower. However, +PyPy always uses a pure-Python implementation, and, as such, it is unaffected +by this variable. + +Dependencies +------------ + +YARL requires multidict_ library. + + +API documentation +------------------ + +The documentation is located at https://yarl.aio-libs.org. + + +Why isn't boolean supported by the URL query API? +------------------------------------------------- + +There is no standard for boolean representation of boolean values. + +Some systems prefer ``true``/``false``, others like ``yes``/``no``, ``on``/``off``, +``Y``/``N``, ``1``/``0``, etc. + +``yarl`` cannot make an unambiguous decision on how to serialize ``bool`` values because +it is specific to how the end-user's application is built and would be different for +different apps. The library doesn't accept booleans in the API; a user should convert +bools into strings using own preferred translation protocol. + + +Comparison with other URL libraries +------------------------------------ + +* furl (https://pypi.python.org/pypi/furl) + + The library has rich functionality but the ``furl`` object is mutable. + + I'm afraid to pass this object into foreign code: who knows if the + code will modify my url in a terrible way while I just want to send URL + with handy helpers for accessing URL properties. + + ``furl`` has other non-obvious tricky things but the main objection + is mutability. + +* URLObject (https://pypi.python.org/pypi/URLObject) + + URLObject is immutable, that's pretty good. + + Every URL change generates a new URL object. + + But the library doesn't do any decode/encode transformations leaving the + end user to cope with these gory details. + + +Source code +----------- + +The project is hosted on GitHub_ + +Please file an issue on the `bug tracker +<https://github.com/aio-libs/yarl/issues>`_ if you have found a bug +or have some suggestion in order to improve the library. + +The library uses `Azure Pipelines <https://dev.azure.com/aio-libs/yarl>`_ for +Continuous Integration. + +Discussion list +--------------- + +*aio-libs* google group: https://groups.google.com/forum/#!forum/aio-libs + +Feel free to post your questions and ideas here. + + +Authors and License +------------------- + +The ``yarl`` package is written by Andrew Svetlov. + +It's *Apache 2* licensed and freely available. + + +.. _GitHub: https://github.com/aio-libs/yarl + +.. _multidict: https://github.com/aio-libs/multidict + +.. + You should *NOT* be adding new change log entries to this file, this + file is managed by towncrier. You *may* edit previous change logs to + fix problems like typo corrections or such. + To add a new change log entry, please see + https://pip.pypa.io/en/latest/development/#adding-a-news-entry + we named the news folder "changes". + + WARNING: Don't drop the next directive! + +.. towncrier release notes start + +1.9.3 (2023-11-20) +================== + +Bug fixes +--------- + +- Stopped dropping trailing slashes in ``yarl.URL.joinpath()`` -- by `@mjpieters <https://github.com/sponsors/mjpieters>`__. (`#862 <https://github.com/aio-libs/yarl/issues/862>`__, `#866 <https://github.com/aio-libs/yarl/issues/866>`__) +- Started accepting string subclasses in ``__truediv__()`` operations (``URL / segment``) -- by `@mjpieters <https://github.com/sponsors/mjpieters>`__. (`#871 <https://github.com/aio-libs/yarl/issues/871>`__, `#884 <https://github.com/aio-libs/yarl/issues/884>`__) +- Fixed the human representation of URLs with square brackets in usernames and passwords -- by `@mjpieters <https://github.com/sponsors/mjpieters>`__. (`#876 <https://github.com/aio-libs/yarl/issues/876>`__, `#882 <https://github.com/aio-libs/yarl/issues/882>`__) +- Updated type hints to include ``URL.missing_port()``, ``URL.__bytes__()`` + and the ``encoding`` argument to ``yarl.URL.joinpath()`` + -- by `@mjpieters <https://github.com/sponsors/mjpieters>`__. (`#891 <https://github.com/aio-libs/yarl/issues/891>`__) + + +Packaging updates and notes for downstreams +------------------------------------------- + +- Integrated Cython 3 to enable building *yarl* under Python 3.12 -- by `@mjpieters <https://github.com/sponsors/mjpieters>`__. (`#829 <https://github.com/aio-libs/yarl/issues/829>`__, `#881 <https://github.com/aio-libs/yarl/issues/881>`__) +- Declared modern ``setuptools.build_meta`` as the ``517`` build + backend in ``pyproject.toml`` explicitly -- by `@webknjaz <https://github.com/sponsors/webknjaz>`__. (`#886 <https://github.com/aio-libs/yarl/issues/886>`__) +- Converted most of the packaging setup into a declarative ``setup.cfg`` + config -- by `@webknjaz <https://github.com/sponsors/webknjaz>`__. (`#890 <https://github.com/aio-libs/yarl/issues/890>`__) +- Replaced the packaging is replaced from an old-fashioned ``setup.py`` to an + in-tree ``517`` build backend -- by `@webknjaz <https://github.com/sponsors/webknjaz>`__. + + Whenever the end-users or downstream packagers need to build ``yarl`` from + source (a Git checkout or an sdist), they may pass a ``config_settings`` + flag ``--pure-python``. If this flag is not set, a C-extension will be built + and included into the distribution. + + Here is how this can be done with ``pip``: + + .. code-block:: console + + $ python -m pip install . --config-settings=--pure-python= + + This will also work with ``-e | --editable``. + + The same can be achieved via ``pypa/build``: + + .. code-block:: console + + $ python -m build --config-setting=--pure-python= + + Adding ``-w | --wheel`` can force ``pypa/build`` produce a wheel from source + directly, as opposed to building an ``sdist`` and then building from it. (`#893 <https://github.com/aio-libs/yarl/issues/893>`__) +- Declared Python 3.12 supported officially in the distribution package metadata + -- by `@edgarrmondragon <https://github.com/sponsors/edgarrmondragon>`__. (`#942 <https://github.com/aio-libs/yarl/issues/942>`__) + + +Contributor-facing changes +-------------------------- + +- A regression test for no-host URLs was added per `#821 <https://github.com/aio-libs/yarl/issues/821>`__ + and ``3986`` -- by `@kenballus <https://github.com/sponsors/kenballus>`__. (`#821 <https://github.com/aio-libs/yarl/issues/821>`__, `#822 <https://github.com/aio-libs/yarl/issues/822>`__) +- Started testing *yarl* against Python 3.12 in CI -- by `@mjpieters <https://github.com/sponsors/mjpieters>`__. (`#881 <https://github.com/aio-libs/yarl/issues/881>`__) +- All Python 3.12 jobs are now marked as required to pass in CI + -- by `@edgarrmondragon <https://github.com/sponsors/edgarrmondragon>`__. (`#942 <https://github.com/aio-libs/yarl/issues/942>`__) +- MyST is now integrated in Sphinx -- by `@webknjaz <https://github.com/sponsors/webknjaz>`__. + + This allows the contributors to author new documents in Markdown + when they have difficulties with going straight RST. (`#953 <https://github.com/aio-libs/yarl/issues/953>`__) + + +1.9.2 (2023-04-25) +================== + +Bugfixes +-------- + +- Fix regression with ``__truediv__`` and absolute URLs with empty paths causing the raw path to lack the leading ``/``. + (`#854 <https://github.com/aio-libs/yarl/issues/854>`_) + + +1.9.1 (2023-04-21) +================== + +Bugfixes +-------- + +- Marked tests that fail on older Python patch releases (< 3.7.10, < 3.8.8 and < 3.9.2) as expected to fail due to missing a security fix for CVE-2021-23336. (`#850 <https://github.com/aio-libs/yarl/issues/850>`_) + + +1.9.0 (2023-04-19) +================== + +This release was never published to PyPI, due to issues with the build process. + +Features +-------- + +- Added ``URL.joinpath(*elements)``, to create a new URL appending multiple path elements. (`#704 <https://github.com/aio-libs/yarl/issues/704>`_) +- Made ``URL.__truediv__()`` return ``NotImplemented`` if called with an + unsupported type — by `@michaeljpeters <https://github.com/sponsors/michaeljpeters>`__. + (`#832 <https://github.com/aio-libs/yarl/issues/832>`_) + + +Bugfixes +-------- + +- Path normalization for absolute URLs no longer raises a ValueError exception + when ``..`` segments would otherwise go beyond the URL path root. + (`#536 <https://github.com/aio-libs/yarl/issues/536>`_) +- Fixed an issue with update_query() not getting rid of the query when argument is None. (`#792 <https://github.com/aio-libs/yarl/issues/792>`_) +- Added some input restrictions on with_port() function to prevent invalid boolean inputs or out of valid port inputs; handled incorrect 0 port representation. (`#793 <https://github.com/aio-libs/yarl/issues/793>`_) +- Made ``yarl.URL.build()`` raise a ``TypeError`` if the ``host`` argument is ``None`` — by `@paulpapacz <https://github.com/sponsors/paulpapacz>`__. (`#808 <https://github.com/aio-libs/yarl/issues/808>`_) +- Fixed an issue with ``update_query()`` getting rid of the query when the argument + is empty but not ``None``. (`#845 <https://github.com/aio-libs/yarl/issues/845>`_) + + +Misc +---- + +- `#220 <https://github.com/aio-libs/yarl/issues/220>`_ + + +1.8.2 (2022-12-03) +================== + +This is the first release that started shipping wheels for Python 3.11. + + +1.8.1 (2022-08-01) +================== + +Misc +---- + +- `#694 <https://github.com/aio-libs/yarl/issues/694>`_, `#699 <https://github.com/aio-libs/yarl/issues/699>`_, `#700 <https://github.com/aio-libs/yarl/issues/700>`_, `#701 <https://github.com/aio-libs/yarl/issues/701>`_, `#702 <https://github.com/aio-libs/yarl/issues/702>`_, `#703 <https://github.com/aio-libs/yarl/issues/703>`_, `#739 <https://github.com/aio-libs/yarl/issues/739>`_ + + +1.8.0 (2022-08-01) +================== + +Features +-------- + +- Added ``URL.raw_suffix``, ``URL.suffix``, ``URL.raw_suffixes``, ``URL.suffixes``, ``URL.with_suffix``. (`#613 <https://github.com/aio-libs/yarl/issues/613>`_) + + +Improved Documentation +---------------------- + +- Fixed broken internal references to ``yarl.URL.human_repr()``. + (`#665 <https://github.com/aio-libs/yarl/issues/665>`_) +- Fixed broken external references to ``multidict:index`` docs. (`#665 <https://github.com/aio-libs/yarl/issues/665>`_) + + +Deprecations and Removals +------------------------- + +- Dropped Python 3.6 support. (`#672 <https://github.com/aio-libs/yarl/issues/672>`_) + + +Misc +---- + +- `#646 <https://github.com/aio-libs/yarl/issues/646>`_, `#699 <https://github.com/aio-libs/yarl/issues/699>`_, `#701 <https://github.com/aio-libs/yarl/issues/701>`_ + + +1.7.2 (2021-11-01) +================== + +Bugfixes +-------- + +- Changed call in ``with_port()`` to stop reencoding parts of the URL that were already encoded. (`#623 <https://github.com/aio-libs/yarl/issues/623>`_) + + +1.7.1 (2021-10-07) +================== + +Bugfixes +-------- + +- Fix 1.7.0 build error + +1.7.0 (2021-10-06) +================== + +Features +-------- + +- Add ``__bytes__()`` magic method so that ``bytes(url)`` will work and use optimal ASCII encoding. + (`#582 <https://github.com/aio-libs/yarl/issues/582>`_) +- Started shipping platform-specific arm64 wheels for Apple Silicon. (`#622 <https://github.com/aio-libs/yarl/issues/622>`_) +- Started shipping platform-specific wheels with the ``musl`` tag targeting typical Alpine Linux runtimes. (`#622 <https://github.com/aio-libs/yarl/issues/622>`_) +- Added support for Python 3.10. (`#622 <https://github.com/aio-libs/yarl/issues/622>`_) + + +1.6.3 (2020-11-14) +================== + +Bugfixes +-------- + +- No longer loose characters when decoding incorrect percent-sequences (like ``%e2%82%f8``). All non-decodable percent-sequences are now preserved. + `#517 <https://github.com/aio-libs/yarl/issues/517>`_ +- Provide x86 Windows wheels. + `#535 <https://github.com/aio-libs/yarl/issues/535>`_ + + +---- + + +1.6.2 (2020-10-12) +================== + + +Bugfixes +-------- + +- Provide generated ``.c`` files in TarBall distribution. + `#530 <https://github.com/aio-libs/multidict/issues/530>`_ + +1.6.1 (2020-10-12) +================== + +Features +-------- + +- Provide wheels for ``aarch64``, ``i686``, ``ppc64le``, ``s390x`` architectures on + Linux as well as ``x86_64``. + `#507 <https://github.com/aio-libs/yarl/issues/507>`_ +- Provide wheels for Python 3.9. + `#526 <https://github.com/aio-libs/yarl/issues/526>`_ + +Bugfixes +-------- + +- ``human_repr()`` now always produces valid representation equivalent to the original URL (if the original URL is valid). + `#511 <https://github.com/aio-libs/yarl/issues/511>`_ +- Fixed requoting a single percent followed by a percent-encoded character in the Cython implementation. + `#514 <https://github.com/aio-libs/yarl/issues/514>`_ +- Fix ValueError when decoding ``%`` which is not followed by two hexadecimal digits. + `#516 <https://github.com/aio-libs/yarl/issues/516>`_ +- Fix decoding ``%`` followed by a space and hexadecimal digit. + `#520 <https://github.com/aio-libs/yarl/issues/520>`_ +- Fix annotation of ``with_query()``/``update_query()`` methods for ``key=[val1, val2]`` case. + `#528 <https://github.com/aio-libs/yarl/issues/528>`_ + +Removal +------- + +- Drop Python 3.5 support; Python 3.6 is the minimal supported Python version. + + +---- + + +1.6.0 (2020-09-23) +================== + +Features +-------- + +- Allow for int and float subclasses in query, while still denying bool. + `#492 <https://github.com/aio-libs/yarl/issues/492>`_ + + +Bugfixes +-------- + +- Do not requote arguments in ``URL.build()``, ``with_xxx()`` and in ``/`` operator. + `#502 <https://github.com/aio-libs/yarl/issues/502>`_ +- Keep IPv6 brackets in ``origin()``. + `#504 <https://github.com/aio-libs/yarl/issues/504>`_ + + +---- + + +1.5.1 (2020-08-01) +================== + +Bugfixes +-------- + +- Fix including relocated internal ``yarl._quoting_c`` C-extension into published PyPI dists. + `#485 <https://github.com/aio-libs/yarl/issues/485>`_ + + +Misc +---- + +- `#484 <https://github.com/aio-libs/yarl/issues/484>`_ + + +---- + + +1.5.0 (2020-07-26) +================== + +Features +-------- + +- Convert host to lowercase on URL building. + `#386 <https://github.com/aio-libs/yarl/issues/386>`_ +- Allow using ``mod`` operator (``%``) for updating query string (an alias for ``update_query()`` method). + `#435 <https://github.com/aio-libs/yarl/issues/435>`_ +- Allow use of sequences such as ``list`` and ``tuple`` in the values + of a mapping such as ``dict`` to represent that a key has many values:: + + url = URL("http://example.com") + assert url.with_query({"a": [1, 2]}) == URL("http://example.com/?a=1&a=2") + + `#443 <https://github.com/aio-libs/yarl/issues/443>`_ +- Support ``URL.build()`` with scheme and path (creates a relative URL). + `#464 <https://github.com/aio-libs/yarl/issues/464>`_ +- Cache slow IDNA encode/decode calls. + `#476 <https://github.com/aio-libs/yarl/issues/476>`_ +- Add ``@final`` / ``Final`` type hints + `#477 <https://github.com/aio-libs/yarl/issues/477>`_ +- Support URL authority/raw_authority properties and authority argument of ``URL.build()`` method. + `#478 <https://github.com/aio-libs/yarl/issues/478>`_ +- Hide the library implementation details, make the exposed public list very clean. + `#483 <https://github.com/aio-libs/yarl/issues/483>`_ + + +Bugfixes +-------- + +- Fix tests with newer Python (3.7.6, 3.8.1 and 3.9.0+). + `#409 <https://github.com/aio-libs/yarl/issues/409>`_ +- Fix a bug where query component, passed in a form of mapping or sequence, is unquoted in unexpected way. + `#426 <https://github.com/aio-libs/yarl/issues/426>`_ +- Hide ``Query`` and ``QueryVariable`` type aliases in ``__init__.pyi``, now they are prefixed with underscore. + `#431 <https://github.com/aio-libs/yarl/issues/431>`_ +- Keep IPv6 brackets after updating port/user/password. + `#451 <https://github.com/aio-libs/yarl/issues/451>`_ + + +---- + + +1.4.2 (2019-12-05) +================== + +Features +-------- + +- Workaround for missing ``str.isascii()`` in Python 3.6 + `#389 <https://github.com/aio-libs/yarl/issues/389>`_ + + +---- + + +1.4.1 (2019-11-29) +================== + +* Fix regression, make the library work on Python 3.5 and 3.6 again. + +1.4.0 (2019-11-29) +================== + +* Distinguish an empty password in URL from a password not provided at all (#262) + +* Fixed annotations for optional parameters of ``URL.build`` (#309) + +* Use None as default value of ``user`` parameter of ``URL.build`` (#309) + +* Enforce building C Accelerated modules when installing from source tarball, use + ``YARL_NO_EXTENSIONS`` environment variable for falling back to (slower) Pure Python + implementation (#329) + +* Drop Python 3.5 support + +* Fix quoting of plus in path by pure python version (#339) + +* Don't create a new URL if fragment is unchanged (#292) + +* Included in error message the path that produces starting slash forbidden error (#376) + +* Skip slow IDNA encoding for ASCII-only strings (#387) + + +1.3.0 (2018-12-11) +================== + +* Fix annotations for ``query`` parameter (#207) + +* An incoming query sequence can have int variables (the same as for + Mapping type) (#208) + +* Add ``URL.explicit_port`` property (#218) + +* Give a friendlier error when port can't be converted to int (#168) + +* ``bool(URL())`` now returns ``False`` (#272) + +1.2.6 (2018-06-14) +================== + +* Drop Python 3.4 trove classifier (#205) + +1.2.5 (2018-05-23) +================== + +* Fix annotations for ``build`` (#199) + +1.2.4 (2018-05-08) +================== + +* Fix annotations for ``cached_property`` (#195) + +1.2.3 (2018-05-03) +================== + +* Accept ``str`` subclasses in ``URL`` constructor (#190) + +1.2.2 (2018-05-01) +================== + +* Fix build + +1.2.1 (2018-04-30) +================== + +* Pin minimal required Python to 3.5.3 (#189) + +1.2.0 (2018-04-30) +================== + +* Forbid inheritance, replace ``__init__`` with ``__new__`` (#171) + +* Support PEP-561 (provide type hinting marker) (#182) + +1.1.1 (2018-02-17) +================== + +* Fix performance regression: don't encode empty ``netloc`` (#170) + +1.1.0 (2018-01-21) +================== + +* Make pure Python quoter consistent with Cython version (#162) + +1.0.0 (2018-01-15) +================== + +* Use fast path if quoted string does not need requoting (#154) + +* Speed up quoting/unquoting by ``_Quoter`` and ``_Unquoter`` classes (#155) + +* Drop ``yarl.quote`` and ``yarl.unquote`` public functions (#155) + +* Add custom string writer, reuse static buffer if available (#157) + Code is 50-80 times faster than Pure Python version (was 4-5 times faster) + +* Don't recode IP zone (#144) + +* Support ``encoded=True`` in ``yarl.URL.build()`` (#158) + +* Fix updating query with multiple keys (#160) + +0.18.0 (2018-01-10) +=================== + +* Fallback to IDNA 2003 if domain name is not IDNA 2008 compatible (#152) + +0.17.0 (2017-12-30) +=================== + +* Use IDNA 2008 for domain name processing (#149) + +0.16.0 (2017-12-07) +=================== + +* Fix raising ``TypeError`` by ``url.query_string()`` after + ``url.with_query({})`` (empty mapping) (#141) + +0.15.0 (2017-11-23) +=================== + +* Add ``raw_path_qs`` attribute (#137) + +0.14.2 (2017-11-14) +=================== + +* Restore ``strict`` parameter as no-op in ``quote`` / ``unquote`` + +0.14.1 (2017-11-13) +=================== + +* Restore ``strict`` parameter as no-op for sake of compatibility with + aiohttp 2.2 + +0.14.0 (2017-11-11) +=================== + +* Drop strict mode (#123) + +* Fix ``"ValueError: Unallowed PCT %"`` when there's a ``"%"`` in the URL (#124) + +0.13.0 (2017-10-01) +=================== + +* Document ``encoded`` parameter (#102) + +* Support relative URLs like ``'?key=value'`` (#100) + +* Unsafe encoding for QS fixed. Encode ``;`` character in value parameter (#104) + +* Process passwords without user names (#95) + +0.12.0 (2017-06-26) +=================== + +* Properly support paths without leading slash in ``URL.with_path()`` (#90) + +* Enable type annotation checks + +0.11.0 (2017-06-26) +=================== + +* Normalize path (#86) + +* Clear query and fragment parts in ``.with_path()`` (#85) + +0.10.3 (2017-06-13) +=================== + +* Prevent double URL arguments unquoting (#83) + +0.10.2 (2017-05-05) +=================== + +* Unexpected hash behavior (#75) + + +0.10.1 (2017-05-03) +=================== + +* Unexpected compare behavior (#73) + +* Do not quote or unquote + if not a query string. (#74) + + +0.10.0 (2017-03-14) +=================== + +* Added ``URL.build`` class method (#58) + +* Added ``path_qs`` attribute (#42) + + +0.9.8 (2017-02-16) +================== + +* Do not quote ``:`` in path + + +0.9.7 (2017-02-16) +================== + +* Load from pickle without _cache (#56) + +* Percent-encoded pluses in path variables become spaces (#59) + + +0.9.6 (2017-02-15) +================== + +* Revert backward incompatible change (BaseURL) + + +0.9.5 (2017-02-14) +================== + +* Fix BaseURL rich comparison support + + +0.9.4 (2017-02-14) +================== + +* Use BaseURL + + +0.9.3 (2017-02-14) +================== + +* Added BaseURL + + +0.9.2 (2017-02-08) +================== + +* Remove debug print + + +0.9.1 (2017-02-07) +================== + +* Do not lose tail chars (#45) + + +0.9.0 (2017-02-07) +================== + +* Allow to quote ``%`` in non strict mode (#21) + +* Incorrect parsing of query parameters with %3B (;) inside (#34) + +* Fix core dumps (#41) + +* ``tmpbuf`` - compiling error (#43) + +* Added ``URL.update_path()`` method + +* Added ``URL.update_query()`` method (#47) + + +0.8.1 (2016-12-03) +================== + +* Fix broken aiohttp: revert back ``quote`` / ``unquote``. + + +0.8.0 (2016-12-03) +================== + +* Support more verbose error messages in ``.with_query()`` (#24) + +* Don't percent-encode ``@`` and ``:`` in path (#32) + +* Don't expose ``yarl.quote`` and ``yarl.unquote``, these functions are + part of private API + +0.7.1 (2016-11-18) +================== + +* Accept not only ``str`` but all classes inherited from ``str`` also (#25) + +0.7.0 (2016-11-07) +================== + +* Accept ``int`` as value for ``.with_query()`` + +0.6.0 (2016-11-07) +================== + +* Explicitly use UTF8 encoding in ``setup.py`` (#20) +* Properly unquote non-UTF8 strings (#19) + +0.5.3 (2016-11-02) +================== + +* Don't use ``typing.NamedTuple`` fields but indexes on URL construction + +0.5.2 (2016-11-02) +================== + +* Inline ``_encode`` class method + +0.5.1 (2016-11-02) +================== + +* Make URL construction faster by removing extra classmethod calls + +0.5.0 (2016-11-02) +================== + +* Add Cython optimization for quoting/unquoting +* Provide binary wheels + +0.4.3 (2016-09-29) +================== + +* Fix typing stubs + +0.4.2 (2016-09-29) +================== + +* Expose ``quote()`` and ``unquote()`` as public API + +0.4.1 (2016-09-28) +================== + +* Support empty values in query (``'/path?arg'``) + +0.4.0 (2016-09-27) +================== + +* Introduce ``relative()`` (#16) + +0.3.2 (2016-09-27) +================== + +* Typo fixes #15 + +0.3.1 (2016-09-26) +================== + +* Support sequence of pairs as ``with_query()`` parameter + +0.3.0 (2016-09-26) +================== + +* Introduce ``is_default_port()`` + +0.2.1 (2016-09-26) +================== + +* Raise ValueError for URLs like 'http://:8080/' + +0.2.0 (2016-09-18) +================== + +* Avoid doubling slashes when joining paths (#13) + +* Appending path starting from slash is forbidden (#12) + +0.1.4 (2016-09-09) +================== + +* Add ``kwargs`` support for ``with_query()`` (#10) + +0.1.3 (2016-09-07) +================== + +* Document ``with_query()``, ``with_fragment()`` and ``origin()`` + +* Allow ``None`` for ``with_query()`` and ``with_fragment()`` + +0.1.2 (2016-09-07) +================== + +* Fix links, tune docs theme. + +0.1.1 (2016-09-06) +================== + +* Update README, old version used obsolete API + +0.1.0 (2016-09-06) +================== + +* The library was deeply refactored, bytes are gone away but all + accepted strings are encoded if needed. + +0.0.1 (2016-08-30) +================== + +* The first release. diff --git a/contrib/python/yarl/.dist-info/top_level.txt b/contrib/python/yarl/.dist-info/top_level.txt new file mode 100644 index 0000000000..e93e8bddef --- /dev/null +++ b/contrib/python/yarl/.dist-info/top_level.txt @@ -0,0 +1 @@ +yarl diff --git a/contrib/python/yarl/LICENSE b/contrib/python/yarl/LICENSE new file mode 100644 index 0000000000..d645695673 --- /dev/null +++ b/contrib/python/yarl/LICENSE @@ -0,0 +1,202 @@ + + Apache License + Version 2.0, January 2004 + http://www.apache.org/licenses/ + + TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION + + 1. Definitions. + + "License" shall mean the terms and conditions for use, reproduction, + and distribution as defined by Sections 1 through 9 of this document. + + "Licensor" shall mean the copyright owner or entity authorized by + the copyright owner that is granting the License. + + "Legal Entity" shall mean the union of the acting entity and all + other entities that control, are controlled by, or are under common + control with that entity. For the purposes of this definition, + "control" means (i) the power, direct or indirect, to cause the + direction or management of such entity, whether by contract or + otherwise, or (ii) ownership of fifty percent (50%) or more of the + outstanding shares, or (iii) beneficial ownership of such entity. + + "You" (or "Your") shall mean an individual or Legal Entity + exercising permissions granted by this License. + + "Source" form shall mean the preferred form for making modifications, + including but not limited to software source code, documentation + source, and configuration files. + + "Object" form shall mean any form resulting from mechanical + transformation or translation of a Source form, including but + not limited to compiled object code, generated documentation, + and conversions to other media types. + + "Work" shall mean the work of authorship, whether in Source or + Object form, made available under the License, as indicated by a + copyright notice that is included in or attached to the work + (an example is provided in the Appendix below). + + "Derivative Works" shall mean any work, whether in Source or Object + form, that is based on (or derived from) the Work and for which the + editorial revisions, annotations, elaborations, or other modifications + represent, as a whole, an original work of authorship. For the purposes + of this License, Derivative Works shall not include works that remain + separable from, or merely link (or bind by name) to the interfaces of, + the Work and Derivative Works thereof. + + "Contribution" shall mean any work of authorship, including + the original version of the Work and any modifications or additions + to that Work or Derivative Works thereof, that is intentionally + submitted to Licensor for inclusion in the Work by the copyright owner + or by an individual or Legal Entity authorized to submit on behalf of + the copyright owner. For the purposes of this definition, "submitted" + means any form of electronic, verbal, or written communication sent + to the Licensor or its representatives, including but not limited to + communication on electronic mailing lists, source code control systems, + and issue tracking systems that are managed by, or on behalf of, the + Licensor for the purpose of discussing and improving the Work, but + excluding communication that is conspicuously marked or otherwise + designated in writing by the copyright owner as "Not a Contribution." + + "Contributor" shall mean Licensor and any individual or Legal Entity + on behalf of whom a Contribution has been received by Licensor and + subsequently incorporated within the Work. + + 2. Grant of Copyright License. Subject to the terms and conditions of + this License, each Contributor hereby grants to You a perpetual, + worldwide, non-exclusive, no-charge, royalty-free, irrevocable + copyright license to reproduce, prepare Derivative Works of, + publicly display, publicly perform, sublicense, and distribute the + Work and such Derivative Works in Source or Object form. + + 3. Grant of Patent License. Subject to the terms and conditions of + this License, each Contributor hereby grants to You a perpetual, + worldwide, non-exclusive, no-charge, royalty-free, irrevocable + (except as stated in this section) patent license to make, have made, + use, offer to sell, sell, import, and otherwise transfer the Work, + where such license applies only to those patent claims licensable + by such Contributor that are necessarily infringed by their + Contribution(s) alone or by combination of their Contribution(s) + with the Work to which such Contribution(s) was submitted. If You + institute patent litigation against any entity (including a + cross-claim or counterclaim in a lawsuit) alleging that the Work + or a Contribution incorporated within the Work constitutes direct + or contributory patent infringement, then any patent licenses + granted to You under this License for that Work shall terminate + as of the date such litigation is filed. + + 4. Redistribution. You may reproduce and distribute copies of the + Work or Derivative Works thereof in any medium, with or without + modifications, and in Source or Object form, provided that You + meet the following conditions: + + (a) You must give any other recipients of the Work or + Derivative Works a copy of this License; and + + (b) You must cause any modified files to carry prominent notices + stating that You changed the files; and + + (c) You must retain, in the Source form of any Derivative Works + that You distribute, all copyright, patent, trademark, and + attribution notices from the Source form of the Work, + excluding those notices that do not pertain to any part of + the Derivative Works; and + + (d) If the Work includes a "NOTICE" text file as part of its + distribution, then any Derivative Works that You distribute must + include a readable copy of the attribution notices contained + within such NOTICE file, excluding those notices that do not + pertain to any part of the Derivative Works, in at least one + of the following places: within a NOTICE text file distributed + as part of the Derivative Works; within the Source form or + documentation, if provided along with the Derivative Works; or, + within a display generated by the Derivative Works, if and + wherever such third-party notices normally appear. The contents + of the NOTICE file are for informational purposes only and + do not modify the License. You may add Your own attribution + notices within Derivative Works that You distribute, alongside + or as an addendum to the NOTICE text from the Work, provided + that such additional attribution notices cannot be construed + as modifying the License. + + You may add Your own copyright statement to Your modifications and + may provide additional or different license terms and conditions + for use, reproduction, or distribution of Your modifications, or + for any such Derivative Works as a whole, provided Your use, + reproduction, and distribution of the Work otherwise complies with + the conditions stated in this License. + + 5. Submission of Contributions. Unless You explicitly state otherwise, + any Contribution intentionally submitted for inclusion in the Work + by You to the Licensor shall be under the terms and conditions of + this License, without any additional terms or conditions. + Notwithstanding the above, nothing herein shall supersede or modify + the terms of any separate license agreement you may have executed + with Licensor regarding such Contributions. + + 6. Trademarks. This License does not grant permission to use the trade + names, trademarks, service marks, or product names of the Licensor, + except as required for reasonable and customary use in describing the + origin of the Work and reproducing the content of the NOTICE file. + + 7. Disclaimer of Warranty. Unless required by applicable law or + agreed to in writing, Licensor provides the Work (and each + Contributor provides its Contributions) on an "AS IS" BASIS, + WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or + implied, including, without limitation, any warranties or conditions + of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A + PARTICULAR PURPOSE. You are solely responsible for determining the + appropriateness of using or redistributing the Work and assume any + risks associated with Your exercise of permissions under this License. + + 8. Limitation of Liability. In no event and under no legal theory, + whether in tort (including negligence), contract, or otherwise, + unless required by applicable law (such as deliberate and grossly + negligent acts) or agreed to in writing, shall any Contributor be + liable to You for damages, including any direct, indirect, special, + incidental, or consequential damages of any character arising as a + result of this License or out of the use or inability to use the + Work (including but not limited to damages for loss of goodwill, + work stoppage, computer failure or malfunction, or any and all + other commercial damages or losses), even if such Contributor + has been advised of the possibility of such damages. + + 9. Accepting Warranty or Additional Liability. While redistributing + the Work or Derivative Works thereof, You may choose to offer, + and charge a fee for, acceptance of support, warranty, indemnity, + or other liability obligations and/or rights consistent with this + License. However, in accepting such obligations, You may act only + on Your own behalf and on Your sole responsibility, not on behalf + of any other Contributor, and only if You agree to indemnify, + defend, and hold each Contributor harmless for any liability + incurred by, or claims asserted against, such Contributor by reason + of your accepting any such warranty or additional liability. + + END OF TERMS AND CONDITIONS + + APPENDIX: How to apply the Apache License to your work. + + To apply the Apache License to your work, attach the following + boilerplate notice, with the fields enclosed by brackets "[]" + replaced with your own identifying information. (Don't include + the brackets!) The text should be enclosed in the appropriate + comment syntax for the file format. We also recommend that a + file or class name and description of purpose be included on the + same "printed page" as the copyright notice for easier + identification within third-party archives. + + Copyright [yyyy] [name of copyright owner] + + Licensed under the Apache License, Version 2.0 (the "License"); + you may not use this file except in compliance with the License. + You may obtain a copy of the License at + + http://www.apache.org/licenses/LICENSE-2.0 + + Unless required by applicable law or agreed to in writing, software + distributed under the License is distributed on an "AS IS" BASIS, + WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + See the License for the specific language governing permissions and + limitations under the License. diff --git a/contrib/python/yarl/NOTICE b/contrib/python/yarl/NOTICE new file mode 100644 index 0000000000..fa53b2b138 --- /dev/null +++ b/contrib/python/yarl/NOTICE @@ -0,0 +1,13 @@ + Copyright 2016-2021, Andrew Svetlov and aio-libs team + + Licensed under the Apache License, Version 2.0 (the "License"); + you may not use this file except in compliance with the License. + You may obtain a copy of the License at + + http://www.apache.org/licenses/LICENSE-2.0 + + Unless required by applicable law or agreed to in writing, software + distributed under the License is distributed on an "AS IS" BASIS, + WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + See the License for the specific language governing permissions and + limitations under the License. diff --git a/contrib/python/yarl/README.rst b/contrib/python/yarl/README.rst new file mode 100644 index 0000000000..a1032b206a --- /dev/null +++ b/contrib/python/yarl/README.rst @@ -0,0 +1,209 @@ +yarl +==== + +The module provides handy URL class for URL parsing and changing. + +.. image:: https://github.com/aio-libs/yarl/workflows/CI/badge.svg + :target: https://github.com/aio-libs/yarl/actions?query=workflow%3ACI + :align: right + +.. image:: https://codecov.io/gh/aio-libs/yarl/branch/master/graph/badge.svg + :target: https://codecov.io/gh/aio-libs/yarl + +.. image:: https://badge.fury.io/py/yarl.svg + :target: https://badge.fury.io/py/yarl + + +.. image:: https://readthedocs.org/projects/yarl/badge/?version=latest + :target: https://yarl.aio-libs.org + + +.. image:: https://img.shields.io/pypi/pyversions/yarl.svg + :target: https://pypi.python.org/pypi/yarl + +.. image:: https://img.shields.io/matrix/aio-libs:matrix.org?label=Discuss%20on%20Matrix%20at%20%23aio-libs%3Amatrix.org&logo=matrix&server_fqdn=matrix.org&style=flat + :target: https://matrix.to/#/%23aio-libs:matrix.org + :alt: Matrix Room — #aio-libs:matrix.org + +.. image:: https://img.shields.io/matrix/aio-libs-space:matrix.org?label=Discuss%20on%20Matrix%20at%20%23aio-libs-space%3Amatrix.org&logo=matrix&server_fqdn=matrix.org&style=flat + :target: https://matrix.to/#/%23aio-libs-space:matrix.org + :alt: Matrix Space — #aio-libs-space:matrix.org + +Introduction +------------ + +Url is constructed from ``str``: + +.. code-block:: pycon + + >>> from yarl import URL + >>> url = URL('https://www.python.org/~guido?arg=1#frag') + >>> url + URL('https://www.python.org/~guido?arg=1#frag') + +All url parts: *scheme*, *user*, *password*, *host*, *port*, *path*, +*query* and *fragment* are accessible by properties: + +.. code-block:: pycon + + >>> url.scheme + 'https' + >>> url.host + 'www.python.org' + >>> url.path + '/~guido' + >>> url.query_string + 'arg=1' + >>> url.query + <MultiDictProxy('arg': '1')> + >>> url.fragment + 'frag' + +All url manipulations produce a new url object: + +.. code-block:: pycon + + >>> url = URL('https://www.python.org') + >>> url / 'foo' / 'bar' + URL('https://www.python.org/foo/bar') + >>> url / 'foo' % {'bar': 'baz'} + URL('https://www.python.org/foo?bar=baz') + +Strings passed to constructor and modification methods are +automatically encoded giving canonical representation as result: + +.. code-block:: pycon + + >>> url = URL('https://www.python.org/путь') + >>> url + URL('https://www.python.org/%D0%BF%D1%83%D1%82%D1%8C') + +Regular properties are *percent-decoded*, use ``raw_`` versions for +getting *encoded* strings: + +.. code-block:: pycon + + >>> url.path + '/путь' + + >>> url.raw_path + '/%D0%BF%D1%83%D1%82%D1%8C' + +Human readable representation of URL is available as ``.human_repr()``: + +.. code-block:: pycon + + >>> url.human_repr() + 'https://www.python.org/путь' + +For full documentation please read https://yarl.aio-libs.org. + + +Installation +------------ + +:: + + $ pip install yarl + +The library is Python 3 only! + +PyPI contains binary wheels for Linux, Windows and MacOS. If you want to install +``yarl`` on another operating system (like *Alpine Linux*, which is not +manylinux-compliant because of the missing glibc and therefore, cannot be +used with our wheels) the the tarball will be used to compile the library from +the source code. It requires a C compiler and and Python headers installed. + +To skip the compilation you must explicitly opt-in by using a PEP 517 +configuration setting ``--pure-python``, or setting the ``YARL_NO_EXTENSIONS`` +environment variable to a non-empty value, e.g.: + +.. code-block:: console + + $ pip install yarl --config-settings=--pure-python= + +Please note that the pure-Python (uncompiled) version is much slower. However, +PyPy always uses a pure-Python implementation, and, as such, it is unaffected +by this variable. + +Dependencies +------------ + +YARL requires multidict_ library. + + +API documentation +------------------ + +The documentation is located at https://yarl.aio-libs.org. + + +Why isn't boolean supported by the URL query API? +------------------------------------------------- + +There is no standard for boolean representation of boolean values. + +Some systems prefer ``true``/``false``, others like ``yes``/``no``, ``on``/``off``, +``Y``/``N``, ``1``/``0``, etc. + +``yarl`` cannot make an unambiguous decision on how to serialize ``bool`` values because +it is specific to how the end-user's application is built and would be different for +different apps. The library doesn't accept booleans in the API; a user should convert +bools into strings using own preferred translation protocol. + + +Comparison with other URL libraries +------------------------------------ + +* furl (https://pypi.python.org/pypi/furl) + + The library has rich functionality but the ``furl`` object is mutable. + + I'm afraid to pass this object into foreign code: who knows if the + code will modify my url in a terrible way while I just want to send URL + with handy helpers for accessing URL properties. + + ``furl`` has other non-obvious tricky things but the main objection + is mutability. + +* URLObject (https://pypi.python.org/pypi/URLObject) + + URLObject is immutable, that's pretty good. + + Every URL change generates a new URL object. + + But the library doesn't do any decode/encode transformations leaving the + end user to cope with these gory details. + + +Source code +----------- + +The project is hosted on GitHub_ + +Please file an issue on the `bug tracker +<https://github.com/aio-libs/yarl/issues>`_ if you have found a bug +or have some suggestion in order to improve the library. + +The library uses `Azure Pipelines <https://dev.azure.com/aio-libs/yarl>`_ for +Continuous Integration. + +Discussion list +--------------- + +*aio-libs* google group: https://groups.google.com/forum/#!forum/aio-libs + +Feel free to post your questions and ideas here. + + +Authors and License +------------------- + +The ``yarl`` package is written by Andrew Svetlov. + +It's *Apache 2* licensed and freely available. + + +.. _GitHub: https://github.com/aio-libs/yarl + +.. _multidict: https://github.com/aio-libs/multidict diff --git a/contrib/python/yarl/tests/test_cache.py b/contrib/python/yarl/tests/test_cache.py new file mode 100644 index 0000000000..22141dd085 --- /dev/null +++ b/contrib/python/yarl/tests/test_cache.py @@ -0,0 +1,28 @@ +import yarl + +# Don't check the actual behavior but make sure that calls are allowed + + +def teardown_module(): + yarl.cache_configure() + + +def test_cache_clear() -> None: + yarl.cache_clear() + + +def test_cache_info() -> None: + info = yarl.cache_info() + assert info.keys() == {"idna_encode", "idna_decode"} + + +def test_cache_configure_default() -> None: + yarl.cache_configure() + + +def test_cache_configure_None() -> None: + yarl.cache_configure(idna_encode_size=None, idna_decode_size=None) + + +def test_cache_configure_explicit() -> None: + yarl.cache_configure(idna_encode_size=128, idna_decode_size=128) diff --git a/contrib/python/yarl/tests/test_cached_property.py b/contrib/python/yarl/tests/test_cached_property.py new file mode 100644 index 0000000000..5dcb5ece23 --- /dev/null +++ b/contrib/python/yarl/tests/test_cached_property.py @@ -0,0 +1,45 @@ +import pytest + +from yarl._url import cached_property + + +def test_reify(): + class A: + def __init__(self): + self._cache = {} + + @cached_property + def prop(self): + return 1 + + a = A() + assert 1 == a.prop + + +def test_reify_class(): + class A: + def __init__(self): + self._cache = {} + + @cached_property + def prop(self): + """Docstring.""" + return 1 + + assert isinstance(A.prop, cached_property) + assert "Docstring." == A.prop.__doc__ + + +def test_reify_assignment(): + class A: + def __init__(self): + self._cache = {} + + @cached_property + def prop(self): + return 1 + + a = A() + + with pytest.raises(AttributeError): + a.prop = 123 diff --git a/contrib/python/yarl/tests/test_normalize_path.py b/contrib/python/yarl/tests/test_normalize_path.py new file mode 100644 index 0000000000..defc4d8dd7 --- /dev/null +++ b/contrib/python/yarl/tests/test_normalize_path.py @@ -0,0 +1,34 @@ +import pytest + +from yarl import URL + +PATHS = [ + # No dots + ("", ""), + ("/", "/"), + ("//", "//"), + ("///", "///"), + # Single-dot + ("path/to", "path/to"), + ("././path/to", "path/to"), + ("path/./to", "path/to"), + ("path/././to", "path/to"), + ("path/to/.", "path/to/"), + ("path/to/./.", "path/to/"), + # Double-dots + ("../path/to", "path/to"), + ("path/../to", "to"), + ("path/../../to", "to"), + # absolute path root / is maintained; tests based on two + # tests from web-platform-tests project's urltestdata.json + ("/foo/../../../ton", "/ton"), + ("/foo/../../../..bar", "/..bar"), + # Non-ASCII characters + ("μονοπάτι/../../να/ᴜɴɪ/ᴄᴏᴅᴇ", "να/ᴜɴɪ/ᴄᴏᴅᴇ"), + ("μονοπάτι/../../να/𝕦𝕟𝕚/𝕔𝕠𝕕𝕖/.", "να/𝕦𝕟𝕚/𝕔𝕠𝕕𝕖/"), +] + + +@pytest.mark.parametrize("original,expected", PATHS) +def test__normalize_path(original, expected): + assert URL._normalize_path(original) == expected diff --git a/contrib/python/yarl/tests/test_pickle.py b/contrib/python/yarl/tests/test_pickle.py new file mode 100644 index 0000000000..a1f29ab68c --- /dev/null +++ b/contrib/python/yarl/tests/test_pickle.py @@ -0,0 +1,23 @@ +import pickle + +from yarl import URL + +# serialize + + +def test_pickle(): + u1 = URL("test") + hash(u1) + v = pickle.dumps(u1) + u2 = pickle.loads(v) + assert u1._cache + assert not u2._cache + assert hash(u1) == hash(u2) + + +def test_default_style_state(): + u = URL("test") + hash(u) + u.__setstate__((None, {"_val": "test", "_strict": False, "_cache": {"hash": 1}})) + assert not u._cache + assert u._val == "test" diff --git a/contrib/python/yarl/tests/test_quoting.py b/contrib/python/yarl/tests/test_quoting.py new file mode 100644 index 0000000000..7ebc0f9b04 --- /dev/null +++ b/contrib/python/yarl/tests/test_quoting.py @@ -0,0 +1,450 @@ +import pytest + +from yarl._quoting import NO_EXTENSIONS +from yarl._quoting_py import _Quoter as _PyQuoter +from yarl._quoting_py import _Unquoter as _PyUnquoter + +if not NO_EXTENSIONS: + from yarl._quoting_c import _Quoter as _CQuoter + from yarl._quoting_c import _Unquoter as _CUnquoter + + @pytest.fixture(params=[_PyQuoter, _CQuoter], ids=["py_quoter", "c_quoter"]) + def quoter(request): + return request.param + + @pytest.fixture(params=[_PyUnquoter, _CUnquoter], ids=["py_unquoter", "c_unquoter"]) + def unquoter(request): + return request.param + +else: + + @pytest.fixture(params=[_PyQuoter], ids=["py_quoter"]) + def quoter(request): + return request.param + + @pytest.fixture(params=[_PyUnquoter], ids=["py_unquoter"]) + def unquoter(request): + return request.param + + +def hexescape(char): + """Escape char as RFC 2396 specifies""" + hex_repr = hex(ord(char))[2:].upper() + if len(hex_repr) == 1: + hex_repr = "0%s" % hex_repr + return "%" + hex_repr + + +def test_quote_not_allowed_non_strict(quoter): + assert quoter()("%HH") == "%25HH" + + +def test_quote_unfinished_tail_percent_non_strict(quoter): + assert quoter()("%") == "%25" + + +def test_quote_unfinished_tail_digit_non_strict(quoter): + assert quoter()("%2") == "%252" + + +def test_quote_unfinished_tail_safe_non_strict(quoter): + assert quoter()("%x") == "%25x" + + +def test_quote_unfinished_tail_unsafe_non_strict(quoter): + assert quoter()("%#") == "%25%23" + + +def test_quote_unfinished_tail_non_ascii_non_strict(quoter): + assert quoter()("%ß") == "%25%C3%9F" + + +def test_quote_unfinished_tail_non_ascii2_non_strict(quoter): + assert quoter()("%€") == "%25%E2%82%AC" + + +def test_quote_unfinished_tail_non_ascii3_non_strict(quoter): + assert quoter()("%🐍") == "%25%F0%9F%90%8D" + + +def test_quote_from_bytes(quoter): + assert quoter()("archaeological arcana") == "archaeological%20arcana" + assert quoter()("") == "" + + +def test_quote_ignore_broken_unicode(quoter): + s = quoter()( + "j\u001a\udcf4q\udcda/\udc97g\udcee\udccb\u000ch\udccb" + "\u0018\udce4v\u001b\udce2\udcce\udccecom/y\udccepj\u0016" + ) + + assert s == "j%1Aq%2Fg%0Ch%18v%1Bcom%2Fypj%16" + assert quoter()(s) == s + + +def test_unquote_to_bytes(unquoter): + assert unquoter()("abc%20def") == "abc def" + assert unquoter()("") == "" + + +def test_never_quote(quoter): + # Make sure quote() does not quote letters, digits, and "_,.-~" + do_not_quote = ( + "ABCDEFGHIJKLMNOPQRSTUVWXYZ" "abcdefghijklmnopqrstuvwxyz" "0123456789" "_.-~" + ) + assert quoter()(do_not_quote) == do_not_quote + assert quoter(qs=True)(do_not_quote) == do_not_quote + + +def test_safe(quoter): + # Test setting 'safe' parameter does what it should do + quote_by_default = "<>" + assert quoter(safe=quote_by_default)(quote_by_default) == quote_by_default + + ret = quoter(safe=quote_by_default, qs=True)(quote_by_default) + assert ret == quote_by_default + + +_SHOULD_QUOTE = [chr(num) for num in range(32)] +_SHOULD_QUOTE.append(r'<>#"{}|\^[]`') +_SHOULD_QUOTE.append(chr(127)) # For 0x7F +SHOULD_QUOTE = "".join(_SHOULD_QUOTE) + + +@pytest.mark.parametrize("char", SHOULD_QUOTE) +def test_default_quoting(char, quoter): + # Make sure all characters that should be quoted are by default sans + # space (separate test for that). + result = quoter()(char) + assert hexescape(char) == result + result = quoter(qs=True)(char) + assert hexescape(char) == result + + +# TODO: should it encode percent? +def test_default_quoting_percent(quoter): + result = quoter()("%25") + assert "%25" == result + result = quoter(qs=True)("%25") + assert "%25" == result + result = quoter(requote=False)("%25") + assert "%2525" == result + + +def test_default_quoting_partial(quoter): + partial_quote = "ab[]cd" + expected = "ab%5B%5Dcd" + result = quoter()(partial_quote) + assert expected == result + result = quoter(qs=True)(partial_quote) + assert expected == result + + +def test_quoting_space(quoter): + # Make sure quote() and quote_plus() handle spaces as specified in + # their unique way + result = quoter()(" ") + assert result == hexescape(" ") + result = quoter(qs=True)(" ") + assert result == "+" + + given = "a b cd e f" + expect = given.replace(" ", hexescape(" ")) + result = quoter()(given) + assert expect == result + expect = given.replace(" ", "+") + result = quoter(qs=True)(given) + assert expect == result + + +def test_quoting_plus(quoter): + assert quoter(qs=False)("alpha+beta gamma") == "alpha+beta%20gamma" + assert quoter(qs=True)("alpha+beta gamma") == "alpha%2Bbeta+gamma" + assert quoter(safe="+", qs=True)("alpha+beta gamma") == "alpha+beta+gamma" + + +def test_quote_with_unicode(quoter): + # Characters in Latin-1 range, encoded by default in UTF-8 + given = "\u00a2\u00d8ab\u00ff" + expect = "%C2%A2%C3%98ab%C3%BF" + result = quoter()(given) + assert expect == result + # Characters in BMP, encoded by default in UTF-8 + given = "\u6f22\u5b57" # "Kanji" + expect = "%E6%BC%A2%E5%AD%97" + result = quoter()(given) + assert expect == result + + +def test_quote_plus_with_unicode(quoter): + # Characters in Latin-1 range, encoded by default in UTF-8 + given = "\u00a2\u00d8ab\u00ff" + expect = "%C2%A2%C3%98ab%C3%BF" + result = quoter(qs=True)(given) + assert expect == result + # Characters in BMP, encoded by default in UTF-8 + given = "\u6f22\u5b57" # "Kanji" + expect = "%E6%BC%A2%E5%AD%97" + result = quoter(qs=True)(given) + assert expect == result + + +@pytest.mark.parametrize("num", list(range(128))) +def test_unquoting(num, unquoter): + # Make sure unquoting of all ASCII values works + given = hexescape(chr(num)) + expect = chr(num) + result = unquoter()(given) + assert expect == result + if expect not in "+=&;": + result = unquoter(qs=True)(given) + assert expect == result + + +# Expected value should be the same as given. +# See https://url.spec.whatwg.org/#percent-encoded-bytes +@pytest.mark.parametrize( + ("input", "expected"), + [ + ("%", "%"), + ("%2", "%2"), + ("%x", "%x"), + ("%€", "%€"), + ("%2x", "%2x"), + ("%2 ", "%2 "), + ("% 2", "% 2"), + ("%xa", "%xa"), + ("%%", "%%"), + ("%%3f", "%?"), + ("%2%", "%2%"), + ("%2%3f", "%2?"), + ("%x%3f", "%x?"), + ("%€%3f", "%€?"), + ], +) +def test_unquoting_bad_percent_escapes(unquoter, input, expected): + assert unquoter()(input) == expected + + +@pytest.mark.xfail +# FIXME: After conversion to bytes, should not cause UTF-8 decode fail. +# See https://url.spec.whatwg.org/#percent-encoded-bytes +def test_unquoting_invalid_utf8_sequence(unquoter): + with pytest.raises(ValueError): + unquoter()("%AB") + with pytest.raises(ValueError): + unquoter()("%AB%AB") + + +def test_unquoting_mixed_case_percent_escapes(unquoter): + expected = "𝕦" + assert expected == unquoter()("%F0%9D%95%A6") + assert expected == unquoter()("%F0%9d%95%a6") + assert expected == unquoter()("%f0%9D%95%a6") + assert expected == unquoter()("%f0%9d%95%a6") + + +def test_unquoting_parts(unquoter): + # Make sure unquoting works when have non-quoted characters + # interspersed + given = "ab" + hexescape("c") + "d" + expect = "abcd" + result = unquoter()(given) + assert expect == result + result = unquoter(qs=True)(given) + assert expect == result + + +def test_quote_None(quoter): + assert quoter()(None) is None + + +def test_unquote_None(unquoter): + assert unquoter()(None) is None + + +def test_quote_empty_string(quoter): + assert quoter()("") == "" + + +def test_unquote_empty_string(unquoter): + assert unquoter()("") == "" + + +def test_quote_bad_types(quoter): + with pytest.raises(TypeError): + quoter()(123) + + +def test_unquote_bad_types(unquoter): + with pytest.raises(TypeError): + unquoter()(123) + + +def test_quote_lowercase(quoter): + assert quoter()("%d1%84") == "%D1%84" + + +def test_quote_unquoted(quoter): + assert quoter()("%41") == "A" + + +def test_quote_space(quoter): + assert quoter()(" ") == "%20" # NULL + + +# test to see if this would work to fix +# coverage on this file. +def test_quote_percent_last_character(quoter): + # % is last character in this case. + assert quoter()("%") == "%25" + + +def test_unquote_unsafe(unquoter): + assert unquoter(unsafe="@")("%40") == "%40" + + +def test_unquote_unsafe2(unquoter): + assert unquoter(unsafe="@")("%40abc") == "%40abc" + + +def test_unquote_unsafe3(unquoter): + assert unquoter(qs=True)("a%2Bb=?%3D%2B%26") == "a%2Bb=?%3D%2B%26" + + +def test_unquote_unsafe4(unquoter): + assert unquoter(unsafe="@")("a@b") == "a%40b" + + +@pytest.mark.parametrize( + ("input", "expected"), + [ + ("%e2%82", "%e2%82"), + ("%e2%82ac", "%e2%82ac"), + ("%e2%82%f8", "%e2%82%f8"), + ("%e2%82%2b", "%e2%82+"), + ("%e2%82%e2%82%ac", "%e2%82€"), + ("%e2%82%e2%82", "%e2%82%e2%82"), + ], +) +def test_unquote_non_utf8(unquoter, input, expected): + assert unquoter()(input) == expected + + +def test_unquote_unsafe_non_utf8(unquoter): + assert unquoter(unsafe="\n")("%e2%82%0a") == "%e2%82%0A" + + +def test_unquote_plus_non_utf8(unquoter): + assert unquoter(qs=True)("%e2%82%2b") == "%e2%82%2B" + + +def test_quote_non_ascii(quoter): + assert quoter()("%F8") == "%F8" + + +def test_quote_non_ascii2(quoter): + assert quoter()("a%F8b") == "a%F8b" + + +def test_quote_percent_percent_encoded(quoter): + assert quoter()("%%3f") == "%25%3F" + + +def test_quote_percent_digit_percent_encoded(quoter): + assert quoter()("%2%3f") == "%252%3F" + + +def test_quote_percent_safe_percent_encoded(quoter): + assert quoter()("%x%3f") == "%25x%3F" + + +def test_quote_percent_unsafe_percent_encoded(quoter): + assert quoter()("%#%3f") == "%25%23%3F" + + +def test_quote_percent_non_ascii_percent_encoded(quoter): + assert quoter()("%ß%3f") == "%25%C3%9F%3F" + + +def test_quote_percent_non_ascii2_percent_encoded(quoter): + assert quoter()("%€%3f") == "%25%E2%82%AC%3F" + + +def test_quote_percent_non_ascii3_percent_encoded(quoter): + assert quoter()("%🐍%3f") == "%25%F0%9F%90%8D%3F" + + +class StrLike(str): + pass + + +def test_quote_str_like(quoter): + assert quoter()(StrLike("abc")) == "abc" + + +def test_unquote_str_like(unquoter): + assert unquoter()(StrLike("abc")) == "abc" + + +def test_quote_sub_delims(quoter): + assert quoter()("!$&'()*+,;=") == "!$&'()*+,;=" + + +def test_requote_sub_delims(quoter): + assert quoter()("%21%24%26%27%28%29%2A%2B%2C%3B%3D") == "!$&'()*+,;=" + + +def test_unquoting_plus(unquoter): + assert unquoter(qs=False)("a+b") == "a+b" + + +def test_unquote_plus_to_space(unquoter): + assert unquoter(qs=True)("a+b") == "a b" + + +def test_unquote_plus_to_space_unsafe(unquoter): + assert unquoter(unsafe="+", qs=True)("a+b") == "a+b" + + +def test_quote_qs_with_colon(quoter): + s = quoter(safe="=+&?/:@", qs=True)("next=http%3A//example.com/") + assert s == "next=http://example.com/" + + +def test_quote_protected(quoter): + s = quoter(protected="/")("/path%2fto/three") + assert s == "/path%2Fto/three" + + +def test_quote_fastpath_safe(quoter): + s1 = "/path/to" + s2 = quoter(safe="/")(s1) + assert s1 is s2 + + +def test_quote_fastpath_pct(quoter): + s1 = "abc%A0" + s2 = quoter()(s1) + assert s1 is s2 + + +def test_quote_very_large_string(quoter): + # more than 8 KiB + s = "abcфух%30%0a" * 1024 + assert quoter()(s) == "abc%D1%84%D1%83%D1%850%0A" * 1024 + + +def test_space(quoter): + s = "% A" + assert quoter()(s) == "%25%20A" + + +def test_quoter_path_with_plus(quoter): + s = "/test/x+y%2Bz/:+%2B/" + assert "/test/x+y%2Bz/:+%2B/" == quoter(safe="@:", protected="/+")(s) + + +def test_unquoter_path_with_plus(unquoter): + s = "/test/x+y%2Bz/:+%2B/" + assert "/test/x+y+z/:++/" == unquoter(unsafe="+")(s) diff --git a/contrib/python/yarl/tests/test_update_query.py b/contrib/python/yarl/tests/test_update_query.py new file mode 100644 index 0000000000..e47c468341 --- /dev/null +++ b/contrib/python/yarl/tests/test_update_query.py @@ -0,0 +1,366 @@ +import enum + +import pytest +from multidict import MultiDict + +from yarl import URL + +# with_query + + +def test_with_query(): + url = URL("http://example.com") + assert str(url.with_query({"a": "1"})) == "http://example.com/?a=1" + + +def test_update_query(): + url = URL("http://example.com/") + assert str(url.update_query({"a": "1"})) == "http://example.com/?a=1" + assert str(URL("test").update_query(a=1)) == "test?a=1" + + url = URL("http://example.com/?foo=bar") + expected_url = URL("http://example.com/?foo=bar&baz=foo") + + assert url.update_query({"baz": "foo"}) == expected_url + assert url.update_query(baz="foo") == expected_url + assert url.update_query("baz=foo") == expected_url + + +def test_update_query_with_args_and_kwargs(): + url = URL("http://example.com/") + + with pytest.raises(ValueError): + url.update_query("a", foo="bar") + + +def test_update_query_with_multiple_args(): + url = URL("http://example.com/") + + with pytest.raises(ValueError): + url.update_query("a", "b") + + +def test_update_query_with_none_arg(): + url = URL("http://example.com/?foo=bar&baz=foo") + expected_url = URL("http://example.com/") + assert url.update_query(None) == expected_url + + +def test_update_query_with_empty_dict(): + url = URL("http://example.com/?foo=bar&baz=foo") + assert url.update_query({}) == url + + +def test_with_query_list_of_pairs(): + url = URL("http://example.com") + assert str(url.with_query([("a", "1")])) == "http://example.com/?a=1" + + +def test_with_query_list_non_pairs(): + url = URL("http://example.com") + with pytest.raises(ValueError): + url.with_query(["a=1", "b=2", "c=3"]) + + +def test_with_query_kwargs(): + url = URL("http://example.com") + q = url.with_query(query="1", query2="1").query + assert q == dict(query="1", query2="1") + + +def test_with_query_kwargs_and_args_are_mutually_exclusive(): + url = URL("http://example.com") + with pytest.raises(ValueError): + url.with_query({"a": "2", "b": "4"}, a="1") + + +def test_with_query_only_single_arg_is_supported(): + url = URL("http://example.com") + u1 = url.with_query(b=3) + u2 = URL("http://example.com/?b=3") + assert u1 == u2 + with pytest.raises(ValueError): + url.with_query("a=1", "a=b") + + +def test_with_query_empty_dict(): + url = URL("http://example.com/?a=b") + new_url = url.with_query({}) + assert new_url.query_string == "" + assert str(new_url) == "http://example.com/" + + +def test_with_query_empty_str(): + url = URL("http://example.com/?a=b") + assert str(url.with_query("")) == "http://example.com/" + + +def test_with_query_empty_value(): + url = URL("http://example.com/") + assert str(url.with_query({"a": ""})) == "http://example.com/?a=" + + +def test_with_query_str(): + url = URL("http://example.com") + assert str(url.with_query("a=1&b=2")) == "http://example.com/?a=1&b=2" + + +def test_with_query_str_non_ascii_and_spaces(): + url = URL("http://example.com") + url2 = url.with_query("a=1 2&b=знач") + assert url2.raw_query_string == "a=1+2&b=%D0%B7%D0%BD%D0%B0%D1%87" + assert url2.query_string == "a=1 2&b=знач" + + +def test_with_query_int(): + url = URL("http://example.com") + assert url.with_query({"a": 1}) == URL("http://example.com/?a=1") + + +def test_with_query_kwargs_int(): + url = URL("http://example.com") + assert url.with_query(b=2) == URL("http://example.com/?b=2") + + +def test_with_query_list_int(): + url = URL("http://example.com") + assert str(url.with_query([("a", 1)])) == "http://example.com/?a=1" + + +@pytest.mark.parametrize( + ("query", "expected"), + [ + pytest.param({"a": []}, "", id="empty list"), + pytest.param({"a": ()}, "", id="empty tuple"), + pytest.param({"a": [1]}, "/?a=1", id="single list"), + pytest.param({"a": (1,)}, "/?a=1", id="single tuple"), + pytest.param({"a": [1, 2]}, "/?a=1&a=2", id="list"), + pytest.param({"a": (1, 2)}, "/?a=1&a=2", id="tuple"), + pytest.param({"a[]": [1, 2]}, "/?a%5B%5D=1&a%5B%5D=2", id="key with braces"), + pytest.param({"&": [1, 2]}, "/?%26=1&%26=2", id="quote key"), + pytest.param({"a": ["1", 2]}, "/?a=1&a=2", id="mixed types"), + pytest.param({"&": ["=", 2]}, "/?%26=%3D&%26=2", id="quote key and value"), + pytest.param({"a": 1, "b": [2, 3]}, "/?a=1&b=2&b=3", id="single then list"), + pytest.param({"a": [1, 2], "b": 3}, "/?a=1&a=2&b=3", id="list then single"), + pytest.param({"a": ["1&a=2", 3]}, "/?a=1%26a%3D2&a=3", id="ampersand then int"), + pytest.param({"a": [1, "2&a=3"]}, "/?a=1&a=2%26a%3D3", id="int then ampersand"), + ], +) +def test_with_query_sequence(query, expected): + url = URL("http://example.com") + expected = "http://example.com{expected}".format_map(locals()) + assert str(url.with_query(query)) == expected + + +@pytest.mark.parametrize( + "query", + [ + pytest.param({"a": [[1]]}, id="nested"), + pytest.param([("a", [1, 2])], id="tuple list"), + ], +) +def test_with_query_sequence_invalid_use(query): + url = URL("http://example.com") + with pytest.raises(TypeError, match="Invalid variable type"): + url.with_query(query) + + +class _CStr(str): + pass + + +class _EmptyStrEr: + def __str__(self): + return "" + + +class _CInt(int, _EmptyStrEr): + pass + + +class _CFloat(float, _EmptyStrEr): + pass + + +@pytest.mark.parametrize( + ("value", "expected"), + [ + pytest.param("1", "1", id="str"), + pytest.param(_CStr("1"), "1", id="custom str"), + pytest.param(1, "1", id="int"), + pytest.param(_CInt(1), "1", id="custom int"), + pytest.param(1.1, "1.1", id="float"), + pytest.param(_CFloat(1.1), "1.1", id="custom float"), + ], +) +def test_with_query_valid_type(value, expected): + url = URL("http://example.com") + expected = "http://example.com/?a={expected}".format_map(locals()) + assert str(url.with_query({"a": value})) == expected + + +@pytest.mark.parametrize( + ("value", "exc_type"), + [ + pytest.param(True, TypeError, id="bool"), + pytest.param(None, TypeError, id="none"), + pytest.param(float("inf"), ValueError, id="non-finite float"), + pytest.param(float("nan"), ValueError, id="NaN float"), + ], +) +def test_with_query_invalid_type(value, exc_type): + url = URL("http://example.com") + with pytest.raises(exc_type): + url.with_query({"a": value}) + + +@pytest.mark.parametrize( + ("value", "expected"), + [ + pytest.param("1", "1", id="str"), + pytest.param(_CStr("1"), "1", id="custom str"), + pytest.param(1, "1", id="int"), + pytest.param(_CInt(1), "1", id="custom int"), + pytest.param(1.1, "1.1", id="float"), + pytest.param(_CFloat(1.1), "1.1", id="custom float"), + ], +) +def test_with_query_list_valid_type(value, expected): + url = URL("http://example.com") + expected = "http://example.com/?a={expected}".format_map(locals()) + assert str(url.with_query([("a", value)])) == expected + + +@pytest.mark.parametrize( + ("value"), [pytest.param(True, id="bool"), pytest.param(None, id="none")] +) +def test_with_query_list_invalid_type(value): + url = URL("http://example.com") + with pytest.raises(TypeError): + url.with_query([("a", value)]) + + +def test_with_int_enum(): + class IntEnum(int, enum.Enum): + A = 1 + + url = URL("http://example.com/path") + url2 = url.with_query(a=IntEnum.A) + assert str(url2) == "http://example.com/path?a=1" + + +def test_with_float_enum(): + class FloatEnum(float, enum.Enum): + A = 1.1 + + url = URL("http://example.com/path") + url2 = url.with_query(a=FloatEnum.A) + assert str(url2) == "http://example.com/path?a=1.1" + + +def test_with_query_multidict(): + url = URL("http://example.com/path") + q = MultiDict([("a", "b"), ("c", "d")]) + assert str(url.with_query(q)) == "http://example.com/path?a=b&c=d" + + +def test_with_multidict_with_spaces_and_non_ascii(): + url = URL("http://example.com") + url2 = url.with_query({"a b": "ю б"}) + assert url2.raw_query_string == "a+b=%D1%8E+%D0%B1" + + +def test_with_query_multidict_with_unsafe(): + url = URL("http://example.com/path") + url2 = url.with_query({"a+b": "?=+&;"}) + assert url2.raw_query_string == "a%2Bb=?%3D%2B%26%3B" + assert url2.query_string == "a%2Bb=?%3D%2B%26%3B" + assert url2.query == {"a+b": "?=+&;"} + + +def test_with_query_None(): + url = URL("http://example.com/path?a=b") + assert url.with_query(None).query_string == "" + + +def test_with_query_bad_type(): + url = URL("http://example.com") + with pytest.raises(TypeError): + url.with_query(123) + + +def test_with_query_bytes(): + url = URL("http://example.com") + with pytest.raises(TypeError): + url.with_query(b"123") + + +def test_with_query_bytearray(): + url = URL("http://example.com") + with pytest.raises(TypeError): + url.with_query(bytearray(b"123")) + + +def test_with_query_memoryview(): + url = URL("http://example.com") + with pytest.raises(TypeError): + url.with_query(memoryview(b"123")) + + +@pytest.mark.parametrize( + ("query", "expected"), + [ + pytest.param([("key", "1;2;3")], "?key=1%3B2%3B3", id="tuple list semicolon"), + pytest.param({"key": "1;2;3"}, "?key=1%3B2%3B3", id="mapping semicolon"), + pytest.param([("key", "1&a=2")], "?key=1%26a%3D2", id="tuple list ampersand"), + pytest.param({"key": "1&a=2"}, "?key=1%26a%3D2", id="mapping ampersand"), + pytest.param([("&", "=")], "?%26=%3D", id="tuple list quote key"), + pytest.param({"&": "="}, "?%26=%3D", id="mapping quote key"), + pytest.param( + [("a[]", "3")], + "?a%5B%5D=3", + id="quote one key braces", + ), + pytest.param( + [("a[]", "3"), ("a[]", "4")], + "?a%5B%5D=3&a%5B%5D=4", + id="quote many key braces", + ), + ], +) +def test_with_query_params(query, expected): + url = URL("http://example.com/get") + url2 = url.with_query(query) + assert str(url2) == ("http://example.com/get" + expected) + + +def test_with_query_only(): + url = URL() + url2 = url.with_query(key="value") + assert str(url2) == "?key=value" + + +def test_with_query_complex_url(): + target_url = "http://example.com/?game=bulls+%26+cows" + url = URL("/redir").with_query({"t": target_url}) + assert url.query["t"] == target_url + + +def test_update_query_multiple_keys(): + url = URL("http://example.com/path?a=1&a=2") + u2 = url.update_query([("a", "3"), ("a", "4")]) + + assert str(u2) == "http://example.com/path?a=3&a=4" + + +# mod operator + + +def test_update_query_with_mod_operator(): + url = URL("http://example.com/") + assert str(url % {"a": "1"}) == "http://example.com/?a=1" + assert str(url % [("a", "1")]) == "http://example.com/?a=1" + assert str(url % "a=1&b=2") == "http://example.com/?a=1&b=2" + assert str(url % {"a": "1"} % {"b": "2"}) == "http://example.com/?a=1&b=2" + assert str(url % {"a": "1"} % {"a": "3", "b": "2"}) == "http://example.com/?a=3&b=2" + assert str(url / "foo" % {"a": "1"}) == "http://example.com/foo?a=1" diff --git a/contrib/python/yarl/tests/test_url.py b/contrib/python/yarl/tests/test_url.py new file mode 100644 index 0000000000..af13d0b5d5 --- /dev/null +++ b/contrib/python/yarl/tests/test_url.py @@ -0,0 +1,1732 @@ +from enum import Enum +from urllib.parse import SplitResult + +import pytest + +from yarl import URL + + +def test_inheritance(): + with pytest.raises(TypeError) as ctx: + + class MyURL(URL): # type: ignore[misc] + pass + + assert ( + "Inheriting a class " + "<class '__tests__.test_url.test_inheritance.<locals>.MyURL'> " + "from URL is forbidden" == str(ctx.value) + ) + + +def test_str_subclass(): + class S(str): + pass + + assert str(URL(S("http://example.com"))) == "http://example.com" + + +def test_is(): + u1 = URL("http://example.com") + u2 = URL(u1) + assert u1 is u2 + + +def test_bool(): + assert URL("http://example.com") + assert not URL() + assert not URL("") + + +def test_absolute_url_without_host(): + with pytest.raises(ValueError): + URL("http://:8080/") + + +def test_url_is_not_str(): + url = URL("http://example.com") + assert not isinstance(url, str) + + +def test_str(): + url = URL("http://example.com:8888/path/to?a=1&b=2") + assert str(url) == "http://example.com:8888/path/to?a=1&b=2" + + +def test_repr(): + url = URL("http://example.com") + assert "URL('http://example.com')" == repr(url) + + +def test_origin(): + url = URL("http://user:password@example.com:8888/path/to?a=1&b=2") + assert URL("http://example.com:8888") == url.origin() + + +def test_origin_nonascii(): + url = URL("http://user:password@историк.рф:8888/path/to?a=1&b=2") + assert str(url.origin()) == "http://xn--h1aagokeh.xn--p1ai:8888" + + +def test_origin_ipv6(): + url = URL("http://user:password@[::1]:8888/path/to?a=1&b=2") + assert str(url.origin()) == "http://[::1]:8888" + + +def test_origin_not_absolute_url(): + url = URL("/path/to?a=1&b=2") + with pytest.raises(ValueError): + url.origin() + + +def test_origin_no_scheme(): + url = URL("//user:password@example.com:8888/path/to?a=1&b=2") + with pytest.raises(ValueError): + url.origin() + + +def test_drop_dots(): + u = URL("http://example.com/path/../to") + assert str(u) == "http://example.com/to" + + +def test_abs_cmp(): + assert URL("http://example.com:8888") == URL("http://example.com:8888") + assert URL("http://example.com:8888/") == URL("http://example.com:8888/") + assert URL("http://example.com:8888/") == URL("http://example.com:8888") + assert URL("http://example.com:8888") == URL("http://example.com:8888/") + + +def test_abs_hash(): + url = URL("http://example.com:8888") + url_trailing = URL("http://example.com:8888/") + assert hash(url) == hash(url_trailing) + + +# properties + + +def test_scheme(): + url = URL("http://example.com") + assert "http" == url.scheme + + +def test_raw_user(): + url = URL("http://user@example.com") + assert "user" == url.raw_user + + +def test_raw_user_non_ascii(): + url = URL("http://вася@example.com") + assert "%D0%B2%D0%B0%D1%81%D1%8F" == url.raw_user + + +def test_no_user(): + url = URL("http://example.com") + assert url.user is None + + +def test_user_non_ascii(): + url = URL("http://вася@example.com") + assert "вася" == url.user + + +def test_raw_password(): + url = URL("http://user:password@example.com") + assert "password" == url.raw_password + + +def test_raw_password_non_ascii(): + url = URL("http://user:пароль@example.com") + assert "%D0%BF%D0%B0%D1%80%D0%BE%D0%BB%D1%8C" == url.raw_password + + +def test_password_non_ascii(): + url = URL("http://user:пароль@example.com") + assert "пароль" == url.password + + +def test_password_without_user(): + url = URL("http://:password@example.com") + assert url.user is None + assert "password" == url.password + + +def test_user_empty_password(): + url = URL("http://user:@example.com") + assert "user" == url.user + assert "" == url.password + + +def test_raw_host(): + url = URL("http://example.com") + assert "example.com" == url.raw_host + + +def test_raw_host_non_ascii(): + url = URL("http://историк.рф") + assert "xn--h1aagokeh.xn--p1ai" == url.raw_host + + +def test_host_non_ascii(): + url = URL("http://историк.рф") + assert "историк.рф" == url.host + + +def test_localhost(): + url = URL("http://[::1]") + assert "::1" == url.host + + +def test_host_with_underscore(): + url = URL("http://abc_def.com") + assert "abc_def.com" == url.host + + +def test_raw_host_when_port_is_specified(): + url = URL("http://example.com:8888") + assert "example.com" == url.raw_host + + +def test_raw_host_from_str_with_ipv4(): + url = URL("http://127.0.0.1:80") + assert url.raw_host == "127.0.0.1" + + +def test_raw_host_from_str_with_ipv6(): + url = URL("http://[::1]:80") + assert url.raw_host == "::1" + + +def test_authority_full() -> None: + url = URL("http://user:passwd@host.com:8080/path") + assert url.raw_authority == "user:passwd@host.com:8080" + assert url.authority == "user:passwd@host.com:8080" + + +def test_authority_short() -> None: + url = URL("http://host.com/path") + assert url.raw_authority == "host.com" + + +def test_authority_full_nonasci() -> None: + url = URL("http://ваня:пароль@айдеко.рф:8080/path") + assert url.raw_authority == ( + "%D0%B2%D0%B0%D0%BD%D1%8F:%D0%BF%D0%B0%D1%80%D0%BE%D0%BB%D1%8C@" + "xn--80aidohy.xn--p1ai:8080" + ) + assert url.authority == "ваня:пароль@айдеко.рф:8080" + + +def test_lowercase(): + url = URL("http://gitHUB.com") + assert url.raw_host == "github.com" + assert url.host == url.raw_host + + +def test_lowercase_nonascii(): + url = URL("http://Айдеко.Рф") + assert url.raw_host == "xn--80aidohy.xn--p1ai" + assert url.host == "айдеко.рф" + + +def test_compressed_ipv6(): + url = URL("http://[1DEC:0:0:0::1]") + assert url.raw_host == "1dec::1" + assert url.host == url.raw_host + + +def test_ipv4_zone(): + # I'm unsure if it is correct. + url = URL("http://1.2.3.4%тест%42:123") + assert url.raw_host == "1.2.3.4%тест%42" + assert url.host == url.raw_host + + +def test_port_for_explicit_port(): + url = URL("http://example.com:8888") + assert 8888 == url.port + + +def test_port_for_implicit_port(): + url = URL("http://example.com") + assert 80 == url.port + + +def test_port_for_relative_url(): + url = URL("/path/to") + assert url.port is None + + +def test_port_for_unknown_scheme(): + url = URL("unknown://example.com") + assert url.port is None + + +def test_explicit_port_for_explicit_port(): + url = URL("http://example.com:8888") + assert 8888 == url.explicit_port + + +def test_explicit_port_for_implicit_port(): + url = URL("http://example.com") + assert url.explicit_port is None + + +def test_explicit_port_for_relative_url(): + url = URL("/path/to") + assert url.explicit_port is None + + +def test_explicit_port_for_unknown_scheme(): + url = URL("unknown://example.com") + assert url.explicit_port is None + + +def test_raw_path_string_empty(): + url = URL("http://example.com") + assert "/" == url.raw_path + + +def test_raw_path(): + url = URL("http://example.com/path/to") + assert "/path/to" == url.raw_path + + +def test_raw_path_non_ascii(): + url = URL("http://example.com/путь/сюда") + assert "/%D0%BF%D1%83%D1%82%D1%8C/%D1%81%D1%8E%D0%B4%D0%B0" == url.raw_path + + +def test_path_non_ascii(): + url = URL("http://example.com/путь/сюда") + assert "/путь/сюда" == url.path + + +def test_path_with_spaces(): + url = URL("http://example.com/a b/test") + assert "/a b/test" == url.path + + url = URL("http://example.com/a b") + assert "/a b" == url.path + + +def test_raw_path_for_empty_url(): + url = URL() + assert "" == url.raw_path + + +def test_raw_path_for_colon_and_at(): + url = URL("http://example.com/path:abc@123") + assert url.raw_path == "/path:abc@123" + + +def test_raw_query_string(): + url = URL("http://example.com?a=1&b=2") + assert url.raw_query_string == "a=1&b=2" + + +def test_raw_query_string_non_ascii(): + url = URL("http://example.com?б=в&ю=к") + assert url.raw_query_string == "%D0%B1=%D0%B2&%D1%8E=%D0%BA" + + +def test_query_string_non_ascii(): + url = URL("http://example.com?б=в&ю=к") + assert url.query_string == "б=в&ю=к" + + +def test_path_qs(): + url = URL("http://example.com/") + assert url.path_qs == "/" + url = URL("http://example.com/?б=в&ю=к") + assert url.path_qs == "/?б=в&ю=к" + url = URL("http://example.com/path?б=в&ю=к") + assert url.path_qs == "/path?б=в&ю=к" + + +def test_raw_path_qs(): + url = URL("http://example.com/") + assert url.raw_path_qs == "/" + url = URL("http://example.com/?б=в&ю=к") + assert url.raw_path_qs == "/?%D0%B1=%D0%B2&%D1%8E=%D0%BA" + url = URL("http://example.com/path?б=в&ю=к") + assert url.raw_path_qs == "/path?%D0%B1=%D0%B2&%D1%8E=%D0%BA" + url = URL("http://example.com/путь?a=1&b=2") + assert url.raw_path_qs == "/%D0%BF%D1%83%D1%82%D1%8C?a=1&b=2" + + +def test_query_string_spaces(): + url = URL("http://example.com?a+b=c+d&e=f+g") + assert url.query_string == "a b=c d&e=f g" + + +# raw fragment + + +def test_raw_fragment_empty(): + url = URL("http://example.com") + assert "" == url.raw_fragment + + +def test_raw_fragment(): + url = URL("http://example.com/path#anchor") + assert "anchor" == url.raw_fragment + + +def test_raw_fragment_non_ascii(): + url = URL("http://example.com/path#якорь") + assert "%D1%8F%D0%BA%D0%BE%D1%80%D1%8C" == url.raw_fragment + + +def test_raw_fragment_safe(): + url = URL("http://example.com/path#a?b/c:d@e") + assert "a?b/c:d@e" == url.raw_fragment + + +def test_fragment_non_ascii(): + url = URL("http://example.com/path#якорь") + assert "якорь" == url.fragment + + +def test_raw_parts_empty(): + url = URL("http://example.com") + assert ("/",) == url.raw_parts + + +def test_raw_parts(): + url = URL("http://example.com/path/to") + assert ("/", "path", "to") == url.raw_parts + + +def test_raw_parts_without_path(): + url = URL("http://example.com") + assert ("/",) == url.raw_parts + + +def test_raw_path_parts_with_2F_in_path(): + url = URL("http://example.com/path%2Fto/three") + assert ("/", "path%2Fto", "three") == url.raw_parts + + +def test_raw_path_parts_with_2f_in_path(): + url = URL("http://example.com/path%2fto/three") + assert ("/", "path%2Fto", "three") == url.raw_parts + + +def test_raw_parts_for_relative_path(): + url = URL("path/to") + assert ("path", "to") == url.raw_parts + + +def test_raw_parts_for_relative_path_starting_from_slash(): + url = URL("/path/to") + assert ("/", "path", "to") == url.raw_parts + + +def test_raw_parts_for_relative_double_path(): + url = URL("path/to") + assert ("path", "to") == url.raw_parts + + +def test_parts_for_empty_url(): + url = URL() + assert ("",) == url.raw_parts + + +def test_raw_parts_non_ascii(): + url = URL("http://example.com/путь/сюда") + assert ( + "/", + "%D0%BF%D1%83%D1%82%D1%8C", + "%D1%81%D1%8E%D0%B4%D0%B0", + ) == url.raw_parts + + +def test_parts_non_ascii(): + url = URL("http://example.com/путь/сюда") + assert ("/", "путь", "сюда") == url.parts + + +def test_name_for_empty_url(): + url = URL() + assert "" == url.raw_name + + +def test_raw_name(): + url = URL("http://example.com/path/to#frag") + assert "to" == url.raw_name + + +def test_raw_name_root(): + url = URL("http://example.com/#frag") + assert "" == url.raw_name + + +def test_raw_name_root2(): + url = URL("http://example.com") + assert "" == url.raw_name + + +def test_raw_name_root3(): + url = URL("http://example.com/") + assert "" == url.raw_name + + +def test_relative_raw_name(): + url = URL("path/to") + assert "to" == url.raw_name + + +def test_relative_raw_name_starting_from_slash(): + url = URL("/path/to") + assert "to" == url.raw_name + + +def test_relative_raw_name_slash(): + url = URL("/") + assert "" == url.raw_name + + +def test_name_non_ascii(): + url = URL("http://example.com/путь") + assert url.name == "путь" + + +def test_suffix_for_empty_url(): + url = URL() + assert "" == url.raw_suffix + + +def test_raw_suffix(): + url = URL("http://example.com/path/to.txt#frag") + assert ".txt" == url.raw_suffix + + +def test_raw_suffix_root(): + url = URL("http://example.com/#frag") + assert "" == url.raw_suffix + + +def test_raw_suffix_root2(): + url = URL("http://example.com") + assert "" == url.raw_suffix + + +def test_raw_suffix_root3(): + url = URL("http://example.com/") + assert "" == url.raw_suffix + + +def test_relative_raw_suffix(): + url = URL("path/to") + assert "" == url.raw_suffix + + +def test_relative_raw_suffix_starting_from_slash(): + url = URL("/path/to") + assert "" == url.raw_suffix + + +def test_relative_raw_suffix_dot(): + url = URL(".") + assert "" == url.raw_suffix + + +def test_suffix_non_ascii(): + url = URL("http://example.com/путь.суффикс") + assert url.suffix == ".суффикс" + + +def test_suffix_with_empty_name(): + url = URL("http://example.com/.hgrc") + assert "" == url.raw_suffix + + +def test_suffix_multi_dot(): + url = URL("http://example.com/doc.tar.gz") + assert ".gz" == url.raw_suffix + + +def test_suffix_with_dot_name(): + url = URL("http://example.com/doc.") + assert "" == url.raw_suffix + + +def test_suffixes_for_empty_url(): + url = URL() + assert () == url.raw_suffixes + + +def test_raw_suffixes(): + url = URL("http://example.com/path/to.txt#frag") + assert (".txt",) == url.raw_suffixes + + +def test_raw_suffixes_root(): + url = URL("http://example.com/#frag") + assert () == url.raw_suffixes + + +def test_raw_suffixes_root2(): + url = URL("http://example.com") + assert () == url.raw_suffixes + + +def test_raw_suffixes_root3(): + url = URL("http://example.com/") + assert () == url.raw_suffixes + + +def test_relative_raw_suffixes(): + url = URL("path/to") + assert () == url.raw_suffixes + + +def test_relative_raw_suffixes_starting_from_slash(): + url = URL("/path/to") + assert () == url.raw_suffixes + + +def test_relative_raw_suffixes_dot(): + url = URL(".") + assert () == url.raw_suffixes + + +def test_suffixes_non_ascii(): + url = URL("http://example.com/путь.суффикс") + assert url.suffixes == (".суффикс",) + + +def test_suffixes_with_empty_name(): + url = URL("http://example.com/.hgrc") + assert () == url.raw_suffixes + + +def test_suffixes_multi_dot(): + url = URL("http://example.com/doc.tar.gz") + assert (".tar", ".gz") == url.raw_suffixes + + +def test_suffixes_with_dot_name(): + url = URL("http://example.com/doc.") + assert () == url.raw_suffixes + + +def test_plus_in_path(): + url = URL("http://example.com/test/x+y%2Bz/:+%2B/") + assert "/test/x+y+z/:++/" == url.path + + +def test_nonascii_in_qs(): + url = URL("http://example.com") + url2 = url.with_query({"f\xf8\xf8": "f\xf8\xf8"}) + assert "http://example.com/?f%C3%B8%C3%B8=f%C3%B8%C3%B8" == str(url2) + + +def test_percent_encoded_in_qs(): + url = URL("http://example.com") + url2 = url.with_query({"k%cf%80": "v%cf%80"}) + assert str(url2) == "http://example.com/?k%25cf%2580=v%25cf%2580" + assert url2.raw_query_string == "k%25cf%2580=v%25cf%2580" + assert url2.query_string == "k%cf%80=v%cf%80" + assert url2.query == {"k%cf%80": "v%cf%80"} + + +# modifiers + + +def test_parent_raw_path(): + url = URL("http://example.com/path/to") + assert url.parent.raw_path == "/path" + + +def test_parent_raw_parts(): + url = URL("http://example.com/path/to") + assert url.parent.raw_parts == ("/", "path") + + +def test_double_parent_raw_path(): + url = URL("http://example.com/path/to") + assert url.parent.parent.raw_path == "/" + + +def test_empty_parent_raw_path(): + url = URL("http://example.com/") + assert url.parent.parent.raw_path == "/" + + +def test_empty_parent_raw_path2(): + url = URL("http://example.com") + assert url.parent.parent.raw_path == "/" + + +def test_clear_fragment_on_getting_parent(): + url = URL("http://example.com/path/to#frag") + assert URL("http://example.com/path") == url.parent + + +def test_clear_fragment_on_getting_parent_toplevel(): + url = URL("http://example.com/#frag") + assert URL("http://example.com/") == url.parent + + +def test_clear_query_on_getting_parent(): + url = URL("http://example.com/path/to?a=b") + assert URL("http://example.com/path") == url.parent + + +def test_clear_query_on_getting_parent_toplevel(): + url = URL("http://example.com/?a=b") + assert URL("http://example.com/") == url.parent + + +# truediv + + +def test_div_root(): + url = URL("http://example.com") / "path" / "to" + assert str(url) == "http://example.com/path/to" + assert url.raw_path == "/path/to" + + +def test_div_root_with_slash(): + url = URL("http://example.com/") / "path" / "to" + assert str(url) == "http://example.com/path/to" + assert url.raw_path == "/path/to" + + +def test_div(): + url = URL("http://example.com/path") / "to" + assert str(url) == "http://example.com/path/to" + assert url.raw_path == "/path/to" + + +def test_div_with_slash(): + url = URL("http://example.com/path/") / "to" + assert str(url) == "http://example.com/path/to" + assert url.raw_path == "/path/to" + + +def test_div_path_starting_from_slash_is_forbidden(): + url = URL("http://example.com/path/") + with pytest.raises(ValueError): + url / "/to/others" + + +class StrEnum(str, Enum): + spam = "ham" + + def __str__(self): + return self.value + + +def test_div_path_srting_subclass(): + url = URL("http://example.com/path/") / StrEnum.spam + assert str(url) == "http://example.com/path/ham" + + +def test_div_bad_type(): + url = URL("http://example.com/path/") + with pytest.raises(TypeError): + url / 3 + + +def test_div_cleanup_query_and_fragment(): + url = URL("http://example.com/path?a=1#frag") + assert str(url / "to") == "http://example.com/path/to" + + +def test_div_for_empty_url(): + url = URL() / "a" + assert url.raw_parts == ("a",) + + +def test_div_for_relative_url(): + url = URL("a") / "b" + assert url.raw_parts == ("a", "b") + + +def test_div_for_relative_url_started_with_slash(): + url = URL("/a") / "b" + assert url.raw_parts == ("/", "a", "b") + + +def test_div_non_ascii(): + url = URL("http://example.com/сюда") + url2 = url / "туда" + assert url2.path == "/сюда/туда" + assert url2.raw_path == "/%D1%81%D1%8E%D0%B4%D0%B0/%D1%82%D1%83%D0%B4%D0%B0" + assert url2.parts == ("/", "сюда", "туда") + assert url2.raw_parts == ( + "/", + "%D1%81%D1%8E%D0%B4%D0%B0", + "%D1%82%D1%83%D0%B4%D0%B0", + ) + + +def test_div_percent_encoded(): + url = URL("http://example.com/path") + url2 = url / "%cf%80" + assert url2.path == "/path/%cf%80" + assert url2.raw_path == "/path/%25cf%2580" + assert url2.parts == ("/", "path", "%cf%80") + assert url2.raw_parts == ("/", "path", "%25cf%2580") + + +def test_div_with_colon_and_at(): + url = URL("http://example.com/base") / "path:abc@123" + assert url.raw_path == "/base/path:abc@123" + + +def test_div_with_dots(): + url = URL("http://example.com/base") / "../path/./to" + assert url.raw_path == "/path/to" + + +# joinpath + + +@pytest.mark.parametrize( + "base,to_join,expected", + [ + pytest.param("", ("path", "to"), "http://example.com/path/to", id="root"), + pytest.param( + "/", ("path", "to"), "http://example.com/path/to", id="root-with-slash" + ), + pytest.param("/path", ("to",), "http://example.com/path/to", id="path"), + pytest.param( + "/path/", ("to",), "http://example.com/path/to", id="path-with-slash" + ), + pytest.param( + "/path?a=1#frag", + ("to",), + "http://example.com/path/to", + id="cleanup-query-and-fragment", + ), + pytest.param("", ("path/",), "http://example.com/path/", id="trailing-slash"), + pytest.param( + "", ("path/", "to/"), "http://example.com/path/to/", id="duplicate-slash" + ), + pytest.param("", (), "http://example.com", id="empty-segments"), + pytest.param( + "/", ("path/",), "http://example.com/path/", id="base-slash-trailing-slash" + ), + pytest.param( + "/", + ("path/", "to/"), + "http://example.com/path/to/", + id="base-slash-duplicate-slash", + ), + pytest.param("/", (), "http://example.com", id="base-slash-empty-segments"), + ], +) +def test_joinpath(base, to_join, expected): + url = URL(f"http://example.com{base}") + assert str(url.joinpath(*to_join)) == expected + + +@pytest.mark.parametrize( + "url,to_join,expected", + [ + pytest.param(URL(), ("a",), ("a",), id="empty-url"), + pytest.param(URL("a"), ("b",), ("a", "b"), id="relative-path"), + pytest.param(URL("a"), ("b", "", "c"), ("a", "b", "c"), id="empty-element"), + pytest.param(URL("/a"), ("b"), ("/", "a", "b"), id="absolute-path"), + pytest.param(URL(), ("a/",), ("a", ""), id="trailing-slash"), + pytest.param(URL(), ("a/", "b/"), ("a", "b", ""), id="duplicate-slash"), + pytest.param(URL(), (), ("",), id="empty-segments"), + ], +) +def test_joinpath_relative(url, to_join, expected): + assert url.joinpath(*to_join).raw_parts == expected + + +@pytest.mark.parametrize( + "url,to_join,encoded,e_path,e_raw_path,e_parts,e_raw_parts", + [ + pytest.param( + "http://example.com/сюда", + ("туда",), + False, + "/сюда/туда", + "/%D1%81%D1%8E%D0%B4%D0%B0/%D1%82%D1%83%D0%B4%D0%B0", + ("/", "сюда", "туда"), + ("/", "%D1%81%D1%8E%D0%B4%D0%B0", "%D1%82%D1%83%D0%B4%D0%B0"), + id="non-ascii", + ), + pytest.param( + "http://example.com/path", + ("%cf%80",), + False, + "/path/%cf%80", + "/path/%25cf%2580", + ("/", "path", "%cf%80"), + ("/", "path", "%25cf%2580"), + id="percent-encoded", + ), + pytest.param( + "http://example.com/path", + ("%cf%80",), + True, + "/path/π", + "/path/%cf%80", + ("/", "path", "π"), + ("/", "path", "%cf%80"), + id="encoded-percent-encoded", + ), + ], +) +def test_joinpath_encoding( + url, to_join, encoded, e_path, e_raw_path, e_parts, e_raw_parts +): + joined = URL(url).joinpath(*to_join, encoded=encoded) + assert joined.path == e_path + assert joined.raw_path == e_raw_path + assert joined.parts == e_parts + assert joined.raw_parts == e_raw_parts + + +@pytest.mark.parametrize( + "to_join,expected", + [ + pytest.param(("path:abc@123",), "/base/path:abc@123", id="with-colon-and-at"), + pytest.param(("..", "path", ".", "to"), "/path/to", id="with-dots"), + ], +) +def test_joinpath_edgecases(to_join, expected): + url = URL("http://example.com/base").joinpath(*to_join) + assert url.raw_path == expected + + +def test_joinpath_path_starting_from_slash_is_forbidden(): + url = URL("http://example.com/path/") + with pytest.raises( + ValueError, match="Appending path .* starting from slash is forbidden" + ): + assert url.joinpath("/to/others") + + +# with_path + + +def test_with_path(): + url = URL("http://example.com") + url2 = url.with_path("/test") + assert str(url2) == "http://example.com/test" + assert url2.raw_path == "/test" + assert url2.path == "/test" + + +def test_with_path_nonascii(): + url = URL("http://example.com") + url2 = url.with_path("/π") + assert str(url2) == "http://example.com/%CF%80" + assert url2.raw_path == "/%CF%80" + assert url2.path == "/π" + + +def test_with_path_percent_encoded(): + url = URL("http://example.com") + url2 = url.with_path("/%cf%80") + assert str(url2) == "http://example.com/%25cf%2580" + assert url2.raw_path == "/%25cf%2580" + assert url2.path == "/%cf%80" + + +def test_with_path_encoded(): + url = URL("http://example.com") + url2 = url.with_path("/test", encoded=True) + assert str(url2) == "http://example.com/test" + assert url2.raw_path == "/test" + assert url2.path == "/test" + + +def test_with_path_encoded_nonascii(): + url = URL("http://example.com") + url2 = url.with_path("/π", encoded=True) + assert str(url2) == "http://example.com/π" + assert url2.raw_path == "/π" + assert url2.path == "/π" + + +def test_with_path_encoded_percent_encoded(): + url = URL("http://example.com") + url2 = url.with_path("/%cf%80", encoded=True) + assert str(url2) == "http://example.com/%cf%80" + assert url2.raw_path == "/%cf%80" + assert url2.path == "/π" + + +def test_with_path_dots(): + url = URL("http://example.com") + assert str(url.with_path("/test/.")) == "http://example.com/test/" + + +def test_with_path_relative(): + url = URL("/path") + assert str(url.with_path("/new")) == "/new" + + +def test_with_path_query(): + url = URL("http://example.com?a=b") + assert str(url.with_path("/test")) == "http://example.com/test" + + +def test_with_path_fragment(): + url = URL("http://example.com#frag") + assert str(url.with_path("/test")) == "http://example.com/test" + + +def test_with_path_empty(): + url = URL("http://example.com/test") + assert str(url.with_path("")) == "http://example.com" + + +def test_with_path_leading_slash(): + url = URL("http://example.com") + assert url.with_path("test").path == "/test" + + +# with_fragment + + +def test_with_fragment(): + url = URL("http://example.com") + url2 = url.with_fragment("frag") + assert str(url2) == "http://example.com/#frag" + assert url2.raw_fragment == "frag" + assert url2.fragment == "frag" + + +def test_with_fragment_safe(): + url = URL("http://example.com") + u2 = url.with_fragment("a:b?c@d/e") + assert str(u2) == "http://example.com/#a:b?c@d/e" + + +def test_with_fragment_non_ascii(): + url = URL("http://example.com") + url2 = url.with_fragment("фрагм") + assert url2.raw_fragment == "%D1%84%D1%80%D0%B0%D0%B3%D0%BC" + assert url2.fragment == "фрагм" + + +def test_with_fragment_percent_encoded(): + url = URL("http://example.com") + url2 = url.with_fragment("%cf%80") + assert str(url2) == "http://example.com/#%25cf%2580" + assert url2.raw_fragment == "%25cf%2580" + assert url2.fragment == "%cf%80" + + +def test_with_fragment_None(): + url = URL("http://example.com/path#frag") + url2 = url.with_fragment(None) + assert str(url2) == "http://example.com/path" + + +def test_with_fragment_None_matching(): + url = URL("http://example.com/path") + url2 = url.with_fragment(None) + assert url is url2 + + +def test_with_fragment_matching(): + url = URL("http://example.com/path#frag") + url2 = url.with_fragment("frag") + assert url is url2 + + +def test_with_fragment_bad_type(): + url = URL("http://example.com") + with pytest.raises(TypeError): + url.with_fragment(123) + + +# with_name + + +def test_with_name(): + url = URL("http://example.com/a/b") + assert url.raw_parts == ("/", "a", "b") + url2 = url.with_name("c") + assert url2.raw_parts == ("/", "a", "c") + assert url2.parts == ("/", "a", "c") + assert url2.raw_path == "/a/c" + assert url2.path == "/a/c" + + +def test_with_name_for_naked_path(): + url = URL("http://example.com") + url2 = url.with_name("a") + assert url2.raw_parts == ("/", "a") + + +def test_with_name_for_relative_path(): + url = URL("a") + url2 = url.with_name("b") + assert url2.raw_parts == ("b",) + + +def test_with_name_for_relative_path2(): + url = URL("a/b") + url2 = url.with_name("c") + assert url2.raw_parts == ("a", "c") + + +def test_with_name_for_relative_path_starting_from_slash(): + url = URL("/a") + url2 = url.with_name("b") + assert url2.raw_parts == ("/", "b") + + +def test_with_name_for_relative_path_starting_from_slash2(): + url = URL("/a/b") + url2 = url.with_name("c") + assert url2.raw_parts == ("/", "a", "c") + + +def test_with_name_empty(): + url = URL("http://example.com/path/to").with_name("") + assert str(url) == "http://example.com/path/" + + +def test_with_name_non_ascii(): + url = URL("http://example.com/path").with_name("путь") + assert url.path == "/путь" + assert url.raw_path == "/%D0%BF%D1%83%D1%82%D1%8C" + assert url.parts == ("/", "путь") + assert url.raw_parts == ("/", "%D0%BF%D1%83%D1%82%D1%8C") + + +def test_with_name_percent_encoded(): + url = URL("http://example.com/path") + url2 = url.with_name("%cf%80") + assert url2.raw_parts == ("/", "%25cf%2580") + assert url2.parts == ("/", "%cf%80") + assert url2.raw_path == "/%25cf%2580" + assert url2.path == "/%cf%80" + + +def test_with_name_with_slash(): + with pytest.raises(ValueError): + URL("http://example.com").with_name("a/b") + + +def test_with_name_non_str(): + with pytest.raises(TypeError): + URL("http://example.com").with_name(123) + + +def test_with_name_within_colon_and_at(): + url = URL("http://example.com/oldpath").with_name("path:abc@123") + assert url.raw_path == "/path:abc@123" + + +def test_with_name_dot(): + with pytest.raises(ValueError): + URL("http://example.com").with_name(".") + + +def test_with_name_double_dot(): + with pytest.raises(ValueError): + URL("http://example.com").with_name("..") + + +# with_suffix + + +def test_with_suffix(): + url = URL("http://example.com/a/b") + assert url.raw_parts == ("/", "a", "b") + url2 = url.with_suffix(".c") + assert url2.raw_parts == ("/", "a", "b.c") + assert url2.parts == ("/", "a", "b.c") + assert url2.raw_path == "/a/b.c" + assert url2.path == "/a/b.c" + + +def test_with_suffix_for_naked_path(): + url = URL("http://example.com") + with pytest.raises(ValueError) as excinfo: + url.with_suffix(".a") + (msg,) = excinfo.value.args + assert msg == f"{url!r} has an empty name" + + +def test_with_suffix_for_relative_path(): + url = URL("a") + url2 = url.with_suffix(".b") + assert url2.raw_parts == ("a.b",) + + +def test_with_suffix_for_relative_path2(): + url = URL("a/b") + url2 = url.with_suffix(".c") + assert url2.raw_parts == ("a", "b.c") + + +def test_with_suffix_for_relative_path_starting_from_slash(): + url = URL("/a") + url2 = url.with_suffix(".b") + assert url2.raw_parts == ("/", "a.b") + + +def test_with_suffix_for_relative_path_starting_from_slash2(): + url = URL("/a/b") + url2 = url.with_suffix(".c") + assert url2.raw_parts == ("/", "a", "b.c") + + +def test_with_suffix_empty(): + url = URL("http://example.com/path/to").with_suffix("") + assert str(url) == "http://example.com/path/to" + + +def test_with_suffix_non_ascii(): + url = URL("http://example.com/path").with_suffix(".путь") + assert url.path == "/path.путь" + assert url.raw_path == "/path.%D0%BF%D1%83%D1%82%D1%8C" + assert url.parts == ("/", "path.путь") + assert url.raw_parts == ("/", "path.%D0%BF%D1%83%D1%82%D1%8C") + + +def test_with_suffix_percent_encoded(): + url = URL("http://example.com/path") + url2 = url.with_suffix(".%cf%80") + assert url2.raw_parts == ("/", "path.%25cf%2580") + assert url2.parts == ("/", "path.%cf%80") + assert url2.raw_path == "/path.%25cf%2580" + assert url2.path == "/path.%cf%80" + + +def test_with_suffix_without_dot(): + with pytest.raises(ValueError) as excinfo: + URL("http://example.com/a").with_suffix("b") + (msg,) = excinfo.value.args + assert msg == "Invalid suffix 'b'" + + +def test_with_suffix_non_str(): + with pytest.raises(TypeError) as excinfo: + URL("http://example.com").with_suffix(123) + (msg,) = excinfo.value.args + assert msg == "Invalid suffix type" + + +def test_with_suffix_dot(): + with pytest.raises(ValueError) as excinfo: + URL("http://example.com").with_suffix(".") + (msg,) = excinfo.value.args + assert msg == "Invalid suffix '.'" + + +def test_with_suffix_with_slash(): + with pytest.raises(ValueError) as excinfo: + URL("http://example.com/a").with_suffix("/.b") + (msg,) = excinfo.value.args + assert msg == "Invalid suffix '/.b'" + + +def test_with_suffix_with_slash2(): + with pytest.raises(ValueError) as excinfo: + URL("http://example.com/a").with_suffix(".b/.d") + (msg,) = excinfo.value.args + assert msg == "Slash in name is not allowed" + + +def test_with_suffix_replace(): + url = URL("/a.b") + url2 = url.with_suffix(".c") + assert url2.raw_parts == ("/", "a.c") + + +# is_absolute + + +def test_is_absolute_for_relative_url(): + url = URL("/path/to") + assert not url.is_absolute() + + +def test_is_absolute_for_absolute_url(): + url = URL("http://example.com") + assert url.is_absolute() + + +def test_is_non_absolute_for_empty_url(): + url = URL() + assert not url.is_absolute() + + +def test_is_non_absolute_for_empty_url2(): + url = URL("") + assert not url.is_absolute() + + +def test_is_absolute_path_starting_from_double_slash(): + url = URL("//www.python.org") + assert url.is_absolute() + + +# is_default_port + + +def test_is_default_port_for_relative_url(): + url = URL("/path/to") + assert not url.is_default_port() + + +def test_is_default_port_for_absolute_url_without_port(): + url = URL("http://example.com") + assert url.is_default_port() + + +def test_is_default_port_for_absolute_url_with_default_port(): + url = URL("http://example.com:80") + assert url.is_default_port() + + +def test_is_default_port_for_absolute_url_with_nondefault_port(): + url = URL("http://example.com:8080") + assert not url.is_default_port() + + +def test_is_default_port_for_unknown_scheme(): + url = URL("unknown://example.com:8080") + assert not url.is_default_port() + + +# + + +def test_no_scheme(): + url = URL("example.com") + assert url.raw_host is None + assert url.raw_path == "example.com" + assert str(url) == "example.com" + + +def test_no_scheme2(): + url = URL("example.com/a/b") + assert url.raw_host is None + assert url.raw_path == "example.com/a/b" + assert str(url) == "example.com/a/b" + + +def test_from_non_allowed(): + with pytest.raises(TypeError): + URL(1234) + + +def test_from_idna(): + url = URL("http://xn--jxagkqfkduily1i.eu") + assert "http://xn--jxagkqfkduily1i.eu" == str(url) + url = URL("http://xn--einla-pqa.de/") # needs idna 2008 + assert "http://xn--einla-pqa.de/" == str(url) + + +def test_to_idna(): + url = URL("http://εμπορικόσήμα.eu") + assert "http://xn--jxagkqfkduily1i.eu" == str(url) + url = URL("http://einlaß.de/") + assert "http://xn--einla-pqa.de/" == str(url) + + +def test_from_ascii_login(): + url = URL("http://" "%D0%B2%D0%B0%D1%81%D1%8F" "@host:1234/") + assert ("http://" "%D0%B2%D0%B0%D1%81%D1%8F" "@host:1234/") == str(url) + + +def test_from_non_ascii_login(): + url = URL("http://вася@host:1234/") + assert ("http://" "%D0%B2%D0%B0%D1%81%D1%8F" "@host:1234/") == str(url) + + +def test_from_ascii_login_and_password(): + url = URL( + "http://" + "%D0%B2%D0%B0%D1%81%D1%8F" + ":%D0%BF%D0%B0%D1%80%D0%BE%D0%BB%D1%8C" + "@host:1234/" + ) + assert ( + "http://" + "%D0%B2%D0%B0%D1%81%D1%8F" + ":%D0%BF%D0%B0%D1%80%D0%BE%D0%BB%D1%8C" + "@host:1234/" + ) == str(url) + + +def test_from_non_ascii_login_and_password(): + url = URL("http://вася:пароль@host:1234/") + assert ( + "http://" + "%D0%B2%D0%B0%D1%81%D1%8F" + ":%D0%BF%D0%B0%D1%80%D0%BE%D0%BB%D1%8C" + "@host:1234/" + ) == str(url) + + +def test_from_ascii_path(): + url = URL("http://example.com/" "%D0%BF%D1%83%D1%82%D1%8C/%D1%82%D1%83%D0%B4%D0%B0") + assert ( + "http://example.com/" "%D0%BF%D1%83%D1%82%D1%8C/%D1%82%D1%83%D0%B4%D0%B0" + ) == str(url) + + +def test_from_ascii_path_lower_case(): + url = URL("http://example.com/" "%d0%bf%d1%83%d1%82%d1%8c/%d1%82%d1%83%d0%b4%d0%b0") + assert ( + "http://example.com/" "%D0%BF%D1%83%D1%82%D1%8C/%D1%82%D1%83%D0%B4%D0%B0" + ) == str(url) + + +def test_from_non_ascii_path(): + url = URL("http://example.com/путь/туда") + assert ( + "http://example.com/" "%D0%BF%D1%83%D1%82%D1%8C/%D1%82%D1%83%D0%B4%D0%B0" + ) == str(url) + + +def test_bytes(): + url = URL("http://example.com/путь/туда") + assert ( + b"http://example.com/%D0%BF%D1%83%D1%82%D1%8C/%D1%82%D1%83%D0%B4%D0%B0" + == bytes(url) + ) + + +def test_from_ascii_query_parts(): + url = URL( + "http://example.com/" + "?%D0%BF%D0%B0%D1%80%D0%B0%D0%BC" + "=%D0%B7%D0%BD%D0%B0%D1%87" + ) + assert ( + "http://example.com/" + "?%D0%BF%D0%B0%D1%80%D0%B0%D0%BC" + "=%D0%B7%D0%BD%D0%B0%D1%87" + ) == str(url) + + +def test_from_non_ascii_query_parts(): + url = URL("http://example.com/?парам=знач") + assert ( + "http://example.com/" + "?%D0%BF%D0%B0%D1%80%D0%B0%D0%BC" + "=%D0%B7%D0%BD%D0%B0%D1%87" + ) == str(url) + + +def test_from_non_ascii_query_parts2(): + url = URL("http://example.com/?п=з&ю=б") + assert "http://example.com/?%D0%BF=%D0%B7&%D1%8E=%D0%B1" == str(url) + + +def test_from_ascii_fragment(): + url = URL("http://example.com/" "#%D1%84%D1%80%D0%B0%D0%B3%D0%BC%D0%B5%D0%BD%D1%82") + assert ( + "http://example.com/" "#%D1%84%D1%80%D0%B0%D0%B3%D0%BC%D0%B5%D0%BD%D1%82" + ) == str(url) + + +def test_from_bytes_with_non_ascii_fragment(): + url = URL("http://example.com/#фрагмент") + assert ( + "http://example.com/" "#%D1%84%D1%80%D0%B0%D0%B3%D0%BC%D0%B5%D0%BD%D1%82" + ) == str(url) + + +def test_to_str(): + url = URL("http://εμπορικόσήμα.eu/") + assert "http://xn--jxagkqfkduily1i.eu/" == str(url) + + +def test_to_str_long(): + url = URL( + "https://host-12345678901234567890123456789012345678901234567890" "-name:8888/" + ) + expected = ( + "https://host-" + "12345678901234567890123456789012345678901234567890" + "-name:8888/" + ) + assert expected == str(url) + + +def test_decoding_with_2F_in_path(): + url = URL("http://example.com/path%2Fto") + assert "http://example.com/path%2Fto" == str(url) + assert url == URL(str(url)) + + +def test_decoding_with_26_and_3D_in_query(): + url = URL("http://example.com/?%26=%3D") + assert "http://example.com/?%26=%3D" == str(url) + assert url == URL(str(url)) + + +def test_fragment_only_url(): + url = URL("#frag") + assert str(url) == "#frag" + + +def test_url_from_url(): + url = URL("http://example.com") + assert URL(url) == url + assert URL(url).raw_parts == ("/",) + + +def test_lowercase_scheme(): + url = URL("HTTP://example.com") + assert str(url) == "http://example.com" + + +def test_str_for_empty_url(): + url = URL() + assert "" == str(url) + + +def test_parent_for_empty_url(): + url = URL() + assert url is url.parent + + +def test_empty_value_for_query(): + url = URL("http://example.com/path").with_query({"a": ""}) + assert str(url) == "http://example.com/path?a=" + + +def test_none_value_for_query(): + with pytest.raises(TypeError): + URL("http://example.com/path").with_query({"a": None}) + + +def test_decode_pct_in_path(): + url = URL("http://www.python.org/%7Eguido") + assert "http://www.python.org/~guido" == str(url) + + +def test_decode_pct_in_path_lower_case(): + url = URL("http://www.python.org/%7eguido") + assert "http://www.python.org/~guido" == str(url) + + +# join + + +def test_join(): + base = URL("http://www.cwi.nl/%7Eguido/Python.html") + url = URL("FAQ.html") + url2 = base.join(url) + assert str(url2) == "http://www.cwi.nl/~guido/FAQ.html" + + +def test_join_absolute(): + base = URL("http://www.cwi.nl/%7Eguido/Python.html") + url = URL("//www.python.org/%7Eguido") + url2 = base.join(url) + assert str(url2) == "http://www.python.org/~guido" + + +def test_join_non_url(): + base = URL("http://example.com") + with pytest.raises(TypeError): + base.join("path/to") + + +NORMAL = [ + ("g:h", "g:h"), + ("g", "http://a/b/c/g"), + ("./g", "http://a/b/c/g"), + ("g/", "http://a/b/c/g/"), + ("/g", "http://a/g"), + ("//g", "http://g"), + ("?y", "http://a/b/c/d;p?y"), + ("g?y", "http://a/b/c/g?y"), + ("#s", "http://a/b/c/d;p?q#s"), + ("g#s", "http://a/b/c/g#s"), + ("g?y#s", "http://a/b/c/g?y#s"), + (";x", "http://a/b/c/;x"), + ("g;x", "http://a/b/c/g;x"), + ("g;x?y#s", "http://a/b/c/g;x?y#s"), + ("", "http://a/b/c/d;p?q"), + (".", "http://a/b/c/"), + ("./", "http://a/b/c/"), + ("..", "http://a/b/"), + ("../", "http://a/b/"), + ("../g", "http://a/b/g"), + ("../..", "http://a/"), + ("../../", "http://a/"), + ("../../g", "http://a/g"), +] + + +@pytest.mark.parametrize("url,expected", NORMAL) +def test_join_from_rfc_3986_normal(url, expected): + # test case from https://tools.ietf.org/html/rfc3986.html#section-5.4 + base = URL("http://a/b/c/d;p?q") + url = URL(url) + expected = URL(expected) + assert base.join(url) == expected + + +ABNORMAL = [ + ("../../../g", "http://a/g"), + ("../../../../g", "http://a/g"), + ("/./g", "http://a/g"), + ("/../g", "http://a/g"), + ("g.", "http://a/b/c/g."), + (".g", "http://a/b/c/.g"), + ("g..", "http://a/b/c/g.."), + ("..g", "http://a/b/c/..g"), + ("./../g", "http://a/b/g"), + ("./g/.", "http://a/b/c/g/"), + ("g/./h", "http://a/b/c/g/h"), + ("g/../h", "http://a/b/c/h"), + ("g;x=1/./y", "http://a/b/c/g;x=1/y"), + ("g;x=1/../y", "http://a/b/c/y"), + ("g?y/./x", "http://a/b/c/g?y/./x"), + ("g?y/../x", "http://a/b/c/g?y/../x"), + ("g#s/./x", "http://a/b/c/g#s/./x"), + ("g#s/../x", "http://a/b/c/g#s/../x"), +] + + +@pytest.mark.parametrize("url,expected", ABNORMAL) +def test_join_from_rfc_3986_abnormal(url, expected): + # test case from https://tools.ietf.org/html/rfc3986.html#section-5.4.2 + base = URL("http://a/b/c/d;p?q") + url = URL(url) + expected = URL(expected) + assert base.join(url) == expected + + +def test_split_result_non_decoded(): + with pytest.raises(ValueError): + URL(SplitResult("http", "example.com", "path", "qs", "frag")) + + +def test_human_repr(): + url = URL("http://вася:пароль@хост.домен:8080/путь/сюда?арг=вал#фраг") + s = url.human_repr() + assert URL(s) == url + assert s == "http://вася:пароль@хост.домен:8080/путь/сюда?арг=вал#фраг" + + +def test_human_repr_defaults(): + url = URL("путь") + s = url.human_repr() + assert s == "путь" + + +def test_human_repr_default_port(): + url = URL("http://вася:пароль@хост.домен/путь/сюда?арг=вал#фраг") + s = url.human_repr() + assert URL(s) == url + assert s == "http://вася:пароль@хост.домен/путь/сюда?арг=вал#фраг" + + +def test_human_repr_ipv6(): + url = URL("http://[::1]:8080/path") + s = url.human_repr() + url2 = URL(s) + assert url2 == url + assert url2.host == "::1" + assert s == "http://[::1]:8080/path" + + +def test_human_repr_delimiters(): + url = URL.build( + scheme="http", + user=" !\"#$%&'()*+,-./:;<=>?@[\\]^_`{|}~", + password=" !\"#$%&'()*+,-./:;<=>?@[\\]^_`{|}~", + host="хост.домен", + port=8080, + path="/ !\"#$%&'()*+,-./:;<=>?@[\\]^_`{|}~", + query={ + " !\"#$%&'()*+,-./:;<=>?@[\\]^_`{|}~": " !\"#$%&'()*+,-./:;<=>?@[\\]^_`{|}~" + }, + fragment=" !\"#$%&'()*+,-./:;<=>?@[\\]^_`{|}~", + ) + s = url.human_repr() + assert URL(s) == url + assert ( + s == "http:// !\"%23$%25&'()*+,-.%2F%3A;<=>%3F%40%5B\\%5D^_`{|}~" + ": !\"%23$%25&'()*+,-.%2F%3A;<=>%3F%40%5B\\%5D^_`{|}~" + "@хост.домен:8080" + "/ !\"%23$%25&'()*+,-./:;<=>%3F@[\\]^_`{|}~" + "? !\"%23$%25%26'()*%2B,-./:%3B<%3D>?@[\\]^_`{|}~" + "= !\"%23$%25%26'()*%2B,-./:%3B<%3D>?@[\\]^_`{|}~" + "# !\"#$%25&'()*+,-./:;<=>?@[\\]^_`{|}~" + ) + + +def test_human_repr_non_printable(): + url = URL.build( + scheme="http", + user="вася\n\xad\u200b", + password="пароль\n\xad\u200b", + host="хост.домен", + port=8080, + path="/путь\n\xad\u200b", + query={"арг\n\xad\u200b": "вал\n\xad\u200b"}, + fragment="фраг\n\xad\u200b", + ) + s = url.human_repr() + assert URL(s) == url + assert ( + s == "http://вася%0A%C2%AD%E2%80%8B:пароль%0A%C2%AD%E2%80%8B" + "@хост.домен:8080" + "/путь%0A%C2%AD%E2%80%8B" + "?арг%0A%C2%AD%E2%80%8B=вал%0A%C2%AD%E2%80%8B" + "#фраг%0A%C2%AD%E2%80%8B" + ) + + +# relative + + +def test_relative(): + url = URL("http://user:pass@example.com:8080/path?a=b#frag") + rel = url.relative() + assert str(rel) == "/path?a=b#frag" + + +def test_relative_is_relative(): + url = URL("http://user:pass@example.com:8080/path?a=b#frag") + rel = url.relative() + assert not rel.is_absolute() + + +def test_relative_abs_parts_are_removed(): + url = URL("http://user:pass@example.com:8080/path?a=b#frag") + rel = url.relative() + assert not rel.scheme + assert not rel.user + assert not rel.password + assert not rel.host + assert not rel.port + + +def test_relative_fails_on_rel_url(): + with pytest.raises(ValueError): + URL("/path?a=b#frag").relative() + + +def test_slash_and_question_in_query(): + u = URL("http://example.com/path?http://example.com/p?a#b") + assert u.query_string == "http://example.com/p?a" + + +def test_slash_and_question_in_fragment(): + u = URL("http://example.com/path#http://example.com/p?a") + assert u.fragment == "http://example.com/p?a" + + +def test_requoting(): + u = URL("http://127.0.0.1/?next=http%3A//example.com/") + assert u.raw_query_string == "next=http://example.com/" + assert str(u) == "http://127.0.0.1/?next=http://example.com/" diff --git a/contrib/python/yarl/tests/test_url_build.py b/contrib/python/yarl/tests/test_url_build.py new file mode 100644 index 0000000000..51969fa849 --- /dev/null +++ b/contrib/python/yarl/tests/test_url_build.py @@ -0,0 +1,259 @@ +import pytest + +from yarl import URL + +# build classmethod + + +def test_build_without_arguments(): + u = URL.build() + assert str(u) == "" + + +def test_build_simple(): + u = URL.build(scheme="http", host="127.0.0.1") + assert str(u) == "http://127.0.0.1" + + +def test_build_with_scheme(): + u = URL.build(scheme="blob", path="path") + assert str(u) == "blob:path" + + +def test_build_with_host(): + u = URL.build(host="127.0.0.1") + assert str(u) == "//127.0.0.1" + assert u == URL("//127.0.0.1") + + +def test_build_with_scheme_and_host(): + u = URL.build(scheme="http", host="127.0.0.1") + assert str(u) == "http://127.0.0.1" + assert u == URL("http://127.0.0.1") + + +def test_build_with_port(): + with pytest.raises(ValueError): + URL.build(port=8000) + + u = URL.build(scheme="http", host="127.0.0.1", port=8000) + assert str(u) == "http://127.0.0.1:8000" + + +def test_build_with_user(): + u = URL.build(scheme="http", host="127.0.0.1", user="foo") + assert str(u) == "http://foo@127.0.0.1" + + +def test_build_with_user_password(): + u = URL.build(scheme="http", host="127.0.0.1", user="foo", password="bar") + assert str(u) == "http://foo:bar@127.0.0.1" + + +def test_build_with_query_and_query_string(): + with pytest.raises(ValueError): + URL.build( + scheme="http", + host="127.0.0.1", + user="foo", + password="bar", + port=8000, + path="/index.html", + query=dict(arg="value1"), + query_string="arg=value1", + fragment="top", + ) + + +def test_build_with_all(): + u = URL.build( + scheme="http", + host="127.0.0.1", + user="foo", + password="bar", + port=8000, + path="/index.html", + query_string="arg=value1", + fragment="top", + ) + assert str(u) == "http://foo:bar@127.0.0.1:8000/index.html?arg=value1#top" + + +def test_build_with_authority_and_host(): + with pytest.raises(ValueError): + URL.build(authority="host.com", host="example.com") + + +def test_build_with_authority(): + url = URL.build(scheme="http", authority="ваня:bar@host.com:8000", path="path") + assert str(url) == "http://%D0%B2%D0%B0%D0%BD%D1%8F:bar@host.com:8000/path" + + +def test_build_with_authority_without_encoding(): + url = URL.build( + scheme="http", authority="foo:bar@host.com:8000", path="path", encoded=True + ) + assert str(url) == "http://foo:bar@host.com:8000/path" + + +def test_query_str(): + u = URL.build(scheme="http", host="127.0.0.1", path="/", query_string="arg=value1") + assert str(u) == "http://127.0.0.1/?arg=value1" + + +def test_query_dict(): + u = URL.build(scheme="http", host="127.0.0.1", path="/", query=dict(arg="value1")) + + assert str(u) == "http://127.0.0.1/?arg=value1" + + +def test_build_path_quoting(): + u = URL.build( + scheme="http", host="127.0.0.1", path="/файл.jpg", query=dict(arg="Привет") + ) + + assert u == URL("http://127.0.0.1/файл.jpg?arg=Привет") + assert str(u) == ( + "http://127.0.0.1/%D1%84%D0%B0%D0%B9%D0%BB.jpg?" + "arg=%D0%9F%D1%80%D0%B8%D0%B2%D0%B5%D1%82" + ) + + +def test_build_query_quoting(): + u = URL.build(scheme="http", host="127.0.0.1", path="/файл.jpg", query="arg=Привет") + + assert u == URL("http://127.0.0.1/файл.jpg?arg=Привет") + assert str(u) == ( + "http://127.0.0.1/%D1%84%D0%B0%D0%B9%D0%BB.jpg?" + "arg=%D0%9F%D1%80%D0%B8%D0%B2%D0%B5%D1%82" + ) + + +def test_build_query_only(): + u = URL.build(query={"key": "value"}) + + assert str(u) == "?key=value" + + +def test_build_drop_dots(): + u = URL.build(scheme="http", host="example.com", path="/path/../to") + assert str(u) == "http://example.com/to" + + +def test_build_encode(): + u = URL.build( + scheme="http", + host="историк.рф", + path="/путь/файл", + query_string="ключ=знач", + fragment="фраг", + ) + expected = ( + "http://xn--h1aagokeh.xn--p1ai" + "/%D0%BF%D1%83%D1%82%D1%8C/%D1%84%D0%B0%D0%B9%D0%BB" + "?%D0%BA%D0%BB%D1%8E%D1%87=%D0%B7%D0%BD%D0%B0%D1%87" + "#%D1%84%D1%80%D0%B0%D0%B3" + ) + assert str(u) == expected + + +def test_build_already_encoded(): + # resulting URL is invalid but not encoded + u = URL.build( + scheme="http", + host="историк.рф", + path="/путь/файл", + query_string="ключ=знач", + fragment="фраг", + encoded=True, + ) + assert str(u) == "http://историк.рф/путь/файл?ключ=знач#фраг" + + +def test_build_percent_encoded(): + u = URL.build( + scheme="http", + host="%2d.org", + user="u%2d", + password="p%2d", + path="/%2d", + query_string="k%2d=v%2d", + fragment="f%2d", + ) + assert str(u) == "http://u%252d:p%252d@%2d.org/%252d?k%252d=v%252d#f%252d" + assert u.raw_host == "%2d.org" + assert u.host == "%2d.org" + assert u.raw_user == "u%252d" + assert u.user == "u%2d" + assert u.raw_password == "p%252d" + assert u.password == "p%2d" + assert u.raw_authority == "u%252d:p%252d@%2d.org" + assert u.authority == "u%2d:p%2d@%2d.org:80" + assert u.raw_path == "/%252d" + assert u.path == "/%2d" + assert u.query == {"k%2d": "v%2d"} + assert u.raw_query_string == "k%252d=v%252d" + assert u.query_string == "k%2d=v%2d" + assert u.raw_fragment == "f%252d" + assert u.fragment == "f%2d" + + +def test_build_with_authority_percent_encoded(): + u = URL.build(scheme="http", authority="u%2d:p%2d@%2d.org") + assert str(u) == "http://u%252d:p%252d@%2d.org" + assert u.raw_host == "%2d.org" + assert u.host == "%2d.org" + assert u.raw_user == "u%252d" + assert u.user == "u%2d" + assert u.raw_password == "p%252d" + assert u.password == "p%2d" + assert u.raw_authority == "u%252d:p%252d@%2d.org" + assert u.authority == "u%2d:p%2d@%2d.org:80" + + +def test_build_with_authority_percent_encoded_already_encoded(): + u = URL.build(scheme="http", authority="u%2d:p%2d@%2d.org", encoded=True) + assert str(u) == "http://u%2d:p%2d@%2d.org" + assert u.raw_host == "%2d.org" + assert u.host == "%2d.org" + assert u.user == "u-" + assert u.raw_user == "u%2d" + assert u.password == "p-" + assert u.raw_password == "p%2d" + assert u.authority == "u-:p-@%2d.org:80" + assert u.raw_authority == "u%2d:p%2d@%2d.org" + + +def test_build_with_authority_with_path_with_leading_slash(): + u = URL.build(scheme="http", host="example.com", path="/path_with_leading_slash") + assert str(u) == "http://example.com/path_with_leading_slash" + + +def test_build_with_authority_with_empty_path(): + u = URL.build(scheme="http", host="example.com", path="") + assert str(u) == "http://example.com" + + +def test_build_with_authority_with_path_without_leading_slash(): + with pytest.raises(ValueError): + URL.build(scheme="http", host="example.com", path="path_without_leading_slash") + + +def test_build_with_none_host(): + with pytest.raises(TypeError, match="NoneType is illegal for.*host"): + URL.build(scheme="http", host=None) + + +def test_build_with_none_path(): + with pytest.raises(TypeError): + URL.build(scheme="http", host="example.com", path=None) + + +def test_build_with_none_query_string(): + with pytest.raises(TypeError): + URL.build(scheme="http", host="example.com", query_string=None) + + +def test_build_with_none_fragment(): + with pytest.raises(TypeError): + URL.build(scheme="http", host="example.com", fragment=None) diff --git a/contrib/python/yarl/tests/test_url_cmp_and_hash.py b/contrib/python/yarl/tests/test_url_cmp_and_hash.py new file mode 100644 index 0000000000..17c42e3566 --- /dev/null +++ b/contrib/python/yarl/tests/test_url_cmp_and_hash.py @@ -0,0 +1,88 @@ +from yarl import URL + +# comparison and hashing + + +def test_ne_str(): + url = URL("http://example.com/") + assert url != "http://example.com/" + + +def test_eq(): + url = URL("http://example.com/") + assert url == URL("http://example.com/") + + +def test_hash(): + assert hash(URL("http://example.com/")) == hash(URL("http://example.com/")) + + +def test_hash_double_call(): + url = URL("http://example.com/") + assert hash(url) == hash(url) + + +def test_le_less(): + url1 = URL("http://example1.com/") + url2 = URL("http://example2.com/") + + assert url1 <= url2 + + +def test_le_eq(): + url1 = URL("http://example.com/") + url2 = URL("http://example.com/") + + assert url1 <= url2 + + +def test_le_not_implemented(): + url = URL("http://example1.com/") + + assert url.__le__(123) is NotImplemented + + +def test_lt(): + url1 = URL("http://example1.com/") + url2 = URL("http://example2.com/") + + assert url1 < url2 + + +def test_lt_not_implemented(): + url = URL("http://example1.com/") + + assert url.__lt__(123) is NotImplemented + + +def test_ge_more(): + url1 = URL("http://example1.com/") + url2 = URL("http://example2.com/") + + assert url2 >= url1 + + +def test_ge_eq(): + url1 = URL("http://example.com/") + url2 = URL("http://example.com/") + + assert url2 >= url1 + + +def test_ge_not_implemented(): + url = URL("http://example1.com/") + + assert url.__ge__(123) is NotImplemented + + +def test_gt(): + url1 = URL("http://example1.com/") + url2 = URL("http://example2.com/") + + assert url2 > url1 + + +def test_gt_not_implemented(): + url = URL("http://example1.com/") + + assert url.__gt__(123) is NotImplemented diff --git a/contrib/python/yarl/tests/test_url_parsing.py b/contrib/python/yarl/tests/test_url_parsing.py new file mode 100644 index 0000000000..cc753fcd0c --- /dev/null +++ b/contrib/python/yarl/tests/test_url_parsing.py @@ -0,0 +1,582 @@ +import sys + +import pytest + +from yarl import URL + + +class TestScheme: + def test_scheme_path(self): + u = URL("scheme:path") + assert u.scheme == "scheme" + assert u.host is None + assert u.path == "path" + assert u.query_string == "" + assert u.fragment == "" + + def test_scheme_path_other(self): + u = URL("scheme:path:other") + assert u.scheme == "scheme" + assert u.host is None + assert u.path == "path:other" + assert u.query_string == "" + assert u.fragment == "" + + def test_complex_scheme(self): + u = URL("allow+chars-33.:path") + assert u.scheme == "allow+chars-33." + assert u.host is None + assert u.path == "path" + assert u.query_string == "" + assert u.fragment == "" + + def test_scheme_only(self): + u = URL("simple:") + assert u.scheme == "simple" + assert u.host is None + assert u.path == "" + assert u.query_string == "" + assert u.fragment == "" + + def test_no_scheme1(self): + u = URL("google.com:80") + # See: https://bugs.python.org/issue27657 + if ( + sys.version_info[:3] == (3, 7, 6) + or sys.version_info[:3] == (3, 8, 1) + or sys.version_info >= (3, 9, 0) + ): + assert u.scheme == "google.com" + assert u.host is None + assert u.path == "80" + else: + assert u.scheme == "" + assert u.host is None + assert u.path == "google.com:80" + assert u.query_string == "" + assert u.fragment == "" + + def test_no_scheme2(self): + u = URL("google.com:80/root") + assert u.scheme == "google.com" + assert u.host is None + assert u.path == "80/root" + assert u.query_string == "" + assert u.fragment == "" + + def test_not_a_scheme1(self): + u = URL("not_cheme:path") + assert u.scheme == "" + assert u.host is None + assert u.path == "not_cheme:path" + assert u.query_string == "" + assert u.fragment == "" + + def test_not_a_scheme2(self): + u = URL("signals37:book") + assert u.scheme == "signals37" + assert u.host is None + assert u.path == "book" + assert u.query_string == "" + assert u.fragment == "" + + def test_scheme_rel_path1(self): + u = URL(":relative-path") + assert u.scheme == "" + assert u.host is None + assert u.path == ":relative-path" + assert u.query_string == "" + assert u.fragment == "" + + def test_scheme_rel_path2(self): + u = URL(":relative/path") + assert u.scheme == "" + assert u.host is None + assert u.path == ":relative/path" + assert u.query_string == "" + assert u.fragment == "" + + def test_scheme_weird(self): + u = URL("://and-this") + assert u.scheme == "" + assert u.host is None + assert u.path == "://and-this" + assert u.query_string == "" + assert u.fragment == "" + + +class TestHost: + def test_canonical(self): + u = URL("scheme://host/path") + assert u.scheme == "scheme" + assert u.host == "host" + assert u.path == "/path" + assert u.query_string == "" + assert u.fragment == "" + + def test_absolute_no_scheme(self): + u = URL("//host/path") + assert u.scheme == "" + assert u.host == "host" + assert u.path == "/path" + assert u.query_string == "" + assert u.fragment == "" + + def test_absolute_no_scheme_complex_host(self): + u = URL("//host+path") + assert u.scheme == "" + assert u.host == "host+path" + assert u.path == "/" + assert u.query_string == "" + assert u.fragment == "" + + def test_absolute_no_scheme_simple_host(self): + u = URL("//host") + assert u.scheme == "" + assert u.host == "host" + assert u.path == "/" + assert u.query_string == "" + assert u.fragment == "" + + def test_weird_host(self): + u = URL("//this+is$also&host!") + assert u.scheme == "" + assert u.host == "this+is$also&host!" + assert u.path == "/" + assert u.query_string == "" + assert u.fragment == "" + + def test_scheme_no_host(self): + u = URL("scheme:/host/path") + assert u.scheme == "scheme" + assert u.host is None + assert u.path == "/host/path" + assert u.query_string == "" + assert u.fragment == "" + + def test_scheme_no_host2(self): + u = URL("scheme:///host/path") + assert u.scheme == "scheme" + assert u.host is None + assert u.path == "/host/path" + assert u.query_string == "" + assert u.fragment == "" + + def test_no_scheme_no_host(self): + u = URL("scheme//host/path") + assert u.scheme == "" + assert u.host is None + assert u.path == "scheme//host/path" + assert u.query_string == "" + assert u.fragment == "" + + def test_ipv4(self): + u = URL("//127.0.0.1/") + assert u.scheme == "" + assert u.host == "127.0.0.1" + assert u.path == "/" + assert u.query_string == "" + assert u.fragment == "" + + def test_ipv6(self): + u = URL("//[::1]/") + assert u.scheme == "" + assert u.host == "::1" + assert u.path == "/" + assert u.query_string == "" + assert u.fragment == "" + + def test_ipvfuture_address(self): + u = URL("//[v1.-1]/") + assert u.scheme == "" + assert u.host == "v1.-1" + assert u.path == "/" + assert u.query_string == "" + assert u.fragment == "" + + +class TestPort: + def test_canonical(self): + u = URL("//host:80/path") + assert u.scheme == "" + assert u.host == "host" + assert u.port == 80 + assert u.path == "/path" + assert u.query_string == "" + assert u.fragment == "" + + def test_no_path(self): + u = URL("//host:80") + assert u.scheme == "" + assert u.host == "host" + assert u.port == 80 + assert u.path == "/" + assert u.query_string == "" + assert u.fragment == "" + + @pytest.mark.xfail(reason="https://github.com/aio-libs/yarl/issues/821") + def test_no_host(self): + u = URL("//:80") + assert u.scheme == "" + assert u.host == "" + assert u.port == 80 + assert u.path == "/" + assert u.query_string == "" + assert u.fragment == "" + + def test_double_port(self): + with pytest.raises(ValueError): + URL("//h:22:80/") + + def test_bad_port(self): + with pytest.raises(ValueError): + URL("//h:no/path") + + def test_another_bad_port(self): + with pytest.raises(ValueError): + URL("//h:22:no/path") + + def test_bad_port_again(self): + with pytest.raises(ValueError): + URL("//h:-80/path") + + +class TestUserInfo: + def test_canonical(self): + u = URL("sch://user@host/") + assert u.scheme == "sch" + assert u.user == "user" + assert u.host == "host" + assert u.path == "/" + assert u.query_string == "" + assert u.fragment == "" + + def test_user_pass(self): + u = URL("//user:pass@host") + assert u.scheme == "" + assert u.user == "user" + assert u.password == "pass" + assert u.host == "host" + assert u.path == "/" + assert u.query_string == "" + assert u.fragment == "" + + def test_complex_userinfo(self): + u = URL("//user:pas:and:more@host") + assert u.scheme == "" + assert u.user == "user" + assert u.password == "pas:and:more" + assert u.host == "host" + assert u.path == "/" + assert u.query_string == "" + assert u.fragment == "" + + def test_no_user(self): + u = URL("//:pas:@host") + assert u.scheme == "" + assert u.user is None + assert u.password == "pas:" + assert u.host == "host" + assert u.path == "/" + assert u.query_string == "" + assert u.fragment == "" + + def test_weird_user(self): + u = URL("//!($&')*+,;=@host") + assert u.scheme == "" + assert u.user == "!($&')*+,;=" + assert u.password is None + assert u.host == "host" + assert u.path == "/" + assert u.query_string == "" + assert u.fragment == "" + + def test_weird_user2(self): + u = URL("//user@info@ya.ru") + assert u.scheme == "" + assert u.user == "user@info" + assert u.password is None + assert u.host == "ya.ru" + assert u.path == "/" + assert u.query_string == "" + assert u.fragment == "" + + def test_weird_user3(self): + u = URL("//%5Bsome%5D@host") + assert u.scheme == "" + assert u.user == "[some]" + assert u.password is None + assert u.host == "host" + assert u.path == "/" + assert u.query_string == "" + assert u.fragment == "" + + +class TestQuery_String: + def test_simple(self): + u = URL("?query") + assert u.scheme == "" + assert u.user is None + assert u.password is None + assert u.host is None + assert u.path == "" + assert u.query_string == "query" + assert u.fragment == "" + + def test_scheme_query(self): + u = URL("http:?query") + assert u.scheme == "http" + assert u.user is None + assert u.password is None + assert u.host is None + assert u.path == "" + assert u.query_string == "query" + assert u.fragment == "" + + def test_abs_url_query(self): + u = URL("//host?query") + assert u.scheme == "" + assert u.user is None + assert u.password is None + assert u.host == "host" + assert u.path == "/" + assert u.query_string == "query" + assert u.fragment == "" + + def test_abs_url_path_query(self): + u = URL("//host/path?query") + assert u.scheme == "" + assert u.user is None + assert u.password is None + assert u.host == "host" + assert u.path == "/path" + assert u.query_string == "query" + assert u.fragment == "" + + def test_double_question_mark(self): + u = URL("//ho?st/path?query") + assert u.scheme == "" + assert u.user is None + assert u.password is None + assert u.host == "ho" + assert u.path == "/" + assert u.query_string == "st/path?query" + assert u.fragment == "" + + def test_complex_query(self): + u = URL("?a://b:c@d.e/f?g#h") + assert u.scheme == "" + assert u.user is None + assert u.password is None + assert u.host is None + assert u.path == "" + assert u.query_string == "a://b:c@d.e/f?g" + assert u.fragment == "h" + + def test_query_in_fragment(self): + u = URL("#?query") + assert u.scheme == "" + assert u.user is None + assert u.password is None + assert u.host is None + assert u.path == "" + assert u.query_string == "" + assert u.fragment == "?query" + + +class TestFragment: + def test_simple(self): + u = URL("#frag") + assert u.scheme == "" + assert u.user is None + assert u.password is None + assert u.host is None + assert u.path == "" + assert u.query_string == "" + assert u.fragment == "frag" + + def test_scheme_frag(self): + u = URL("http:#frag") + assert u.scheme == "http" + assert u.user is None + assert u.password is None + assert u.host is None + assert u.path == "" + assert u.query_string == "" + assert u.fragment == "frag" + + def test_host_frag(self): + u = URL("//host#frag") + assert u.scheme == "" + assert u.user is None + assert u.password is None + assert u.host == "host" + assert u.path == "/" + assert u.query_string == "" + assert u.fragment == "frag" + + def test_scheme_path_frag(self): + u = URL("//host/path#frag") + assert u.scheme == "" + assert u.user is None + assert u.password is None + assert u.host == "host" + assert u.path == "/path" + assert u.query_string == "" + assert u.fragment == "frag" + + def test_scheme_query_frag(self): + u = URL("//host?query#frag") + assert u.scheme == "" + assert u.user is None + assert u.password is None + assert u.host == "host" + assert u.path == "/" + assert u.query_string == "query" + assert u.fragment == "frag" + + def test_host_frag_query(self): + u = URL("//ho#st/path?query") + assert u.scheme == "" + assert u.user is None + assert u.password is None + assert u.host == "ho" + assert u.path == "/" + assert u.query_string == "" + assert u.fragment == "st/path?query" + + def test_complex_frag(self): + u = URL("#a://b:c@d.e/f?g#h") + assert u.scheme == "" + assert u.user is None + assert u.password is None + assert u.host is None + assert u.path == "" + assert u.query_string == "" + assert u.fragment == "a://b:c@d.e/f?g#h" + + +class TestStripEmptyParts: + def test_all_empty(self): + with pytest.raises(ValueError): + URL("//@:?#") + + def test_path_only(self): + u = URL("///path") + assert u.scheme == "" + assert u.user is None + assert u.password is None + assert u.host is None + assert u.path == "/path" + assert u.query_string == "" + assert u.fragment == "" + + def test_empty_user(self): + u = URL("//@host") + assert u.scheme == "" + assert u.user is None + assert u.password is None + assert u.host == "host" + assert u.path == "/" + assert u.query_string == "" + assert u.fragment == "" + + def test_empty_port(self): + u = URL("//host:") + assert u.scheme == "" + assert u.user is None + assert u.password is None + assert u.host == "host" + assert u.path == "/" + assert u.query_string == "" + assert u.fragment == "" + + def test_empty_port_and_path(self): + u = URL("//host:/") + assert u.scheme == "" + assert u.user is None + assert u.password is None + assert u.host == "host" + assert u.path == "/" + assert u.query_string == "" + assert u.fragment == "" + + def test_empty_path_only(self): + u = URL("/") + assert u.scheme == "" + assert u.user is None + assert u.password is None + assert u.host is None + assert u.path == "/" + assert u.query_string == "" + assert u.fragment == "" + + def test_relative_path_only(self): + u = URL("path") + assert u.scheme == "" + assert u.user is None + assert u.password is None + assert u.host is None + assert u.path == "path" + assert u.query_string == "" + assert u.fragment == "" + + def test_path(self): + u = URL("/path") + assert u.scheme == "" + assert u.user is None + assert u.password is None + assert u.host is None + assert u.path == "/path" + assert u.query_string == "" + assert u.fragment == "" + + def test_empty_query_with_path(self): + u = URL("/path?") + assert u.scheme == "" + assert u.user is None + assert u.password is None + assert u.host is None + assert u.path == "/path" + assert u.query_string == "" + assert u.fragment == "" + + def test_empty_query(self): + u = URL("?") + assert u.scheme == "" + assert u.user is None + assert u.password is None + assert u.host is None + assert u.path == "" + assert u.query_string == "" + assert u.fragment == "" + + def test_empty_query_with_frag(self): + u = URL("?#frag") + assert u.scheme == "" + assert u.user is None + assert u.password is None + assert u.host is None + assert u.path == "" + assert u.query_string == "" + assert u.fragment == "frag" + + def test_path_empty_frag(self): + u = URL("/path#") + assert u.scheme == "" + assert u.user is None + assert u.password is None + assert u.host is None + assert u.path == "/path" + assert u.query_string == "" + assert u.fragment == "" + + def test_empty_path(self): + u = URL("#") + assert u.scheme == "" + assert u.user is None + assert u.password is None + assert u.host is None + assert u.path == "" + assert u.query_string == "" + assert u.fragment == "" diff --git a/contrib/python/yarl/tests/test_url_query.py b/contrib/python/yarl/tests/test_url_query.py new file mode 100644 index 0000000000..bcd2433cbc --- /dev/null +++ b/contrib/python/yarl/tests/test_url_query.py @@ -0,0 +1,173 @@ +from typing import List, Tuple +from urllib.parse import parse_qs, urlencode + +import pytest +from multidict import MultiDict, MultiDictProxy + +from yarl import URL + +# ======================================== +# Basic chars in query values +# ======================================== + +URLS_WITH_BASIC_QUERY_VALUES: List[Tuple[URL, MultiDict]] = [ + # Empty strings, keys and values + ( + URL("http://example.com"), + MultiDict(), + ), + ( + URL("http://example.com?a="), + MultiDict([("a", "")]), + ), + # ASCII chars + ( + URL("http://example.com?a+b=c+d"), + MultiDict({"a b": "c d"}), + ), + ( + URL("http://example.com?a=1&b=2"), + MultiDict([("a", "1"), ("b", "2")]), + ), + ( + URL("http://example.com?a=1&b=2&a=3"), + MultiDict([("a", "1"), ("b", "2"), ("a", "3")]), + ), + # Non-ASCI BMP chars + ( + URL("http://example.com?ключ=знач"), + MultiDict({"ключ": "знач"}), + ), + ( + URL("http://example.com?foo=ᴜɴɪᴄᴏᴅᴇ"), + MultiDict({"foo": "ᴜɴɪᴄᴏᴅᴇ"}), + ), + # Non-BMP chars + ( + URL("http://example.com?bar=𝕦𝕟𝕚𝕔𝕠𝕕𝕖"), + MultiDict({"bar": "𝕦𝕟𝕚𝕔𝕠𝕕𝕖"}), + ), +] + + +@pytest.mark.parametrize( + "original_url, expected_query", + URLS_WITH_BASIC_QUERY_VALUES, +) +def test_query_basic_parsing(original_url, expected_query): + assert isinstance(original_url.query, MultiDictProxy) + assert original_url.query == expected_query + + +@pytest.mark.parametrize( + "original_url, expected_query", + URLS_WITH_BASIC_QUERY_VALUES, +) +def test_query_basic_update_query(original_url, expected_query): + new_url = original_url.update_query({}) + assert new_url == original_url + + +def test_query_dont_unqoute_twice(): + sample_url = "http://base.place?" + urlencode({"a": "/////"}) + query = urlencode({"url": sample_url}) + full_url = "http://test_url.aha?" + query + + url = URL(full_url) + assert url.query["url"] == sample_url + + +# ======================================== +# Reserved chars in query values +# ======================================== + +# See https://github.com/python/cpython#87133, which introduced a new +# `separator` keyword argument to `urllib.parse.parse_qs` (among others). +# If the name doesn't exist as a variable in the function bytecode, the +# test is expected to fail. +_SEMICOLON_XFAIL = pytest.mark.xfail( + condition="separator" not in parse_qs.__code__.co_varnames, + reason=( + "Python versions < 3.7.10, < 3.8.8 and < 3.9.2 lack a fix for " + 'CVE-2021-23336 dropping ";" as a valid query parameter separator, ' + "making this test fail." + ), + strict=True, +) + + +URLS_WITH_RESERVED_CHARS_IN_QUERY_VALUES = [ + # Ampersand + (URL("http://127.0.0.1/?a=10&b=20"), 2, "10"), + (URL("http://127.0.0.1/?a=10%26b=20"), 1, "10&b=20"), + (URL("http://127.0.0.1/?a=10%3Bb=20"), 1, "10;b=20"), + # Semicolon, which is *not* a query parameter separator as of RFC3986 + (URL("http://127.0.0.1/?a=10;b=20"), 1, "10;b=20"), + (URL("http://127.0.0.1/?a=10%26b=20"), 1, "10&b=20"), + (URL("http://127.0.0.1/?a=10%3Bb=20"), 1, "10;b=20"), +] +URLS_WITH_RESERVED_CHARS_IN_QUERY_VALUES_W_XFAIL = [ + # Ampersand + *URLS_WITH_RESERVED_CHARS_IN_QUERY_VALUES[:3], + # Semicolon, which is *not* a query parameter separator as of RFC3986 + # Mark the first of these as expecting to fail on old Python patch releases. + pytest.param(*URLS_WITH_RESERVED_CHARS_IN_QUERY_VALUES[3], marks=_SEMICOLON_XFAIL), + *URLS_WITH_RESERVED_CHARS_IN_QUERY_VALUES[4:], +] + + +@pytest.mark.parametrize( + "original_url, expected_query_len, expected_value_a", + URLS_WITH_RESERVED_CHARS_IN_QUERY_VALUES_W_XFAIL, +) +def test_query_separators_from_parsing( + original_url, + expected_query_len, + expected_value_a, +): + assert len(original_url.query) == expected_query_len + assert original_url.query["a"] == expected_value_a + + +@pytest.mark.parametrize( + "original_url, expected_query_len, expected_value_a", + URLS_WITH_RESERVED_CHARS_IN_QUERY_VALUES_W_XFAIL, +) +def test_query_separators_from_update_query( + original_url, + expected_query_len, + expected_value_a, +): + new_url = original_url.update_query({"c": expected_value_a}) + assert new_url.query["a"] == expected_value_a + assert new_url.query["c"] == expected_value_a + + +@pytest.mark.parametrize( + "original_url, expected_query_len, expected_value_a", + URLS_WITH_RESERVED_CHARS_IN_QUERY_VALUES, +) +def test_query_separators_from_with_query( + original_url, + expected_query_len, + expected_value_a, +): + new_url = original_url.with_query({"c": expected_value_a}) + assert new_url.query["c"] == expected_value_a + + +@pytest.mark.parametrize( + "original_url, expected_query_len, expected_value_a", + URLS_WITH_RESERVED_CHARS_IN_QUERY_VALUES, +) +def test_query_from_empty_update_query( + original_url, + expected_query_len, + expected_value_a, +): + new_url = original_url.update_query({}) + + assert new_url.query["a"] == original_url.query["a"] + + if "b" in original_url.query: + assert new_url.query["b"] == original_url.query["b"] diff --git a/contrib/python/yarl/tests/test_url_update_netloc.py b/contrib/python/yarl/tests/test_url_update_netloc.py new file mode 100644 index 0000000000..cf0cc1c44c --- /dev/null +++ b/contrib/python/yarl/tests/test_url_update_netloc.py @@ -0,0 +1,228 @@ +import pytest + +from yarl import URL + +# with_* + + +def test_with_scheme(): + url = URL("http://example.com") + assert str(url.with_scheme("https")) == "https://example.com" + + +def test_with_scheme_uppercased(): + url = URL("http://example.com") + assert str(url.with_scheme("HTTPS")) == "https://example.com" + + +def test_with_scheme_for_relative_url(): + with pytest.raises(ValueError): + URL("path/to").with_scheme("http") + + +def test_with_scheme_invalid_type(): + url = URL("http://example.com") + with pytest.raises(TypeError): + assert str(url.with_scheme(123)) + + +def test_with_user(): + url = URL("http://example.com") + assert str(url.with_user("john")) == "http://john@example.com" + + +def test_with_user_non_ascii(): + url = URL("http://example.com") + url2 = url.with_user("вася") + assert url2.raw_user == "%D0%B2%D0%B0%D1%81%D1%8F" + assert url2.user == "вася" + assert url2.raw_authority == "%D0%B2%D0%B0%D1%81%D1%8F@example.com" + assert url2.authority == "вася@example.com:80" + + +def test_with_user_percent_encoded(): + url = URL("http://example.com") + url2 = url.with_user("%cf%80") + assert url2.raw_user == "%25cf%2580" + assert url2.user == "%cf%80" + assert url2.raw_authority == "%25cf%2580@example.com" + assert url2.authority == "%cf%80@example.com:80" + + +def test_with_user_for_relative_url(): + with pytest.raises(ValueError): + URL("path/to").with_user("user") + + +def test_with_user_invalid_type(): + url = URL("http://example.com:123") + with pytest.raises(TypeError): + url.with_user(123) + + +def test_with_user_None(): + url = URL("http://john@example.com") + assert str(url.with_user(None)) == "http://example.com" + + +def test_with_user_ipv6(): + url = URL("http://john:pass@[::1]:8080/") + assert str(url.with_user(None)) == "http://[::1]:8080/" + + +def test_with_user_None_when_password_present(): + url = URL("http://john:pass@example.com") + assert str(url.with_user(None)) == "http://example.com" + + +def test_with_password(): + url = URL("http://john@example.com") + assert str(url.with_password("pass")) == "http://john:pass@example.com" + + +def test_with_password_ipv6(): + url = URL("http://john:pass@[::1]:8080/") + assert str(url.with_password(None)) == "http://john@[::1]:8080/" + + +def test_with_password_non_ascii(): + url = URL("http://john@example.com") + url2 = url.with_password("пароль") + assert url2.raw_password == "%D0%BF%D0%B0%D1%80%D0%BE%D0%BB%D1%8C" + assert url2.password == "пароль" + assert url2.raw_authority == "john:%D0%BF%D0%B0%D1%80%D0%BE%D0%BB%D1%8C@example.com" + assert url2.authority == "john:пароль@example.com:80" + + +def test_with_password_percent_encoded(): + url = URL("http://john@example.com") + url2 = url.with_password("%cf%80") + assert url2.raw_password == "%25cf%2580" + assert url2.password == "%cf%80" + assert url2.raw_authority == "john:%25cf%2580@example.com" + assert url2.authority == "john:%cf%80@example.com:80" + + +def test_with_password_non_ascii_with_colon(): + url = URL("http://john@example.com") + url2 = url.with_password("п:а") + assert url2.raw_password == "%D0%BF%3A%D0%B0" + assert url2.password == "п:а" + + +def test_with_password_for_relative_url(): + with pytest.raises(ValueError): + URL("path/to").with_password("pass") + + +def test_with_password_None(): + url = URL("http://john:pass@example.com") + assert str(url.with_password(None)) == "http://john@example.com" + + +def test_with_password_invalid_type(): + url = URL("http://example.com:123") + with pytest.raises(TypeError): + url.with_password(123) + + +def test_with_password_and_empty_user(): + url = URL("http://example.com") + url2 = url.with_password("pass") + assert url2.password == "pass" + assert url2.user is None + assert str(url2) == "http://:pass@example.com" + + +def test_from_str_with_host_ipv4(): + url = URL("http://host:80") + url = url.with_host("192.168.1.1") + assert url.raw_host == "192.168.1.1" + + +def test_from_str_with_host_ipv6(): + url = URL("http://host:80") + url = url.with_host("::1") + assert url.raw_host == "::1" + + +def test_with_host(): + url = URL("http://example.com:123") + assert str(url.with_host("example.org")) == "http://example.org:123" + + +def test_with_host_empty(): + url = URL("http://example.com:123") + with pytest.raises(ValueError): + url.with_host("") + + +def test_with_host_non_ascii(): + url = URL("http://example.com:123") + url2 = url.with_host("историк.рф") + assert url2.raw_host == "xn--h1aagokeh.xn--p1ai" + assert url2.host == "историк.рф" + assert url2.raw_authority == "xn--h1aagokeh.xn--p1ai:123" + assert url2.authority == "историк.рф:123" + + +def test_with_host_percent_encoded(): + url = URL("http://%25cf%2580%cf%80:%25cf%2580%cf%80@example.com:123") + url2 = url.with_host("%cf%80.org") + assert url2.raw_host == "%cf%80.org" + assert url2.host == "%cf%80.org" + assert url2.raw_authority == "%25cf%2580%CF%80:%25cf%2580%CF%80@%cf%80.org:123" + assert url2.authority == "%cf%80π:%cf%80π@%cf%80.org:123" + + +def test_with_host_for_relative_url(): + with pytest.raises(ValueError): + URL("path/to").with_host("example.com") + + +def test_with_host_invalid_type(): + url = URL("http://example.com:123") + with pytest.raises(TypeError): + url.with_host(None) + + +def test_with_port(): + url = URL("http://example.com") + assert str(url.with_port(8888)) == "http://example.com:8888" + + +def test_with_port_with_no_port(): + url = URL("http://example.com") + assert str(url.with_port(None)) == "http://example.com" + + +def test_with_port_ipv6(): + url = URL("http://[::1]:8080/") + assert str(url.with_port(80)) == "http://[::1]:80/" + + +def test_with_port_keeps_query_and_fragment(): + url = URL("http://example.com/?a=1#frag") + assert str(url.with_port(8888)) == "http://example.com:8888/?a=1#frag" + + +def test_with_port_percent_encoded(): + url = URL("http://user%name:pass%word@example.com/") + assert str(url.with_port(808)) == "http://user%25name:pass%25word@example.com:808/" + + +def test_with_port_for_relative_url(): + with pytest.raises(ValueError): + URL("path/to").with_port(1234) + + +def test_with_port_invalid_type(): + with pytest.raises(TypeError): + URL("http://example.com").with_port("123") + with pytest.raises(TypeError): + URL("http://example.com").with_port(True) + + +def test_with_port_invalid_range(): + with pytest.raises(ValueError): + URL("http://example.com").with_port(-1) diff --git a/contrib/python/yarl/tests/ya.make b/contrib/python/yarl/tests/ya.make new file mode 100644 index 0000000000..f86b0f6380 --- /dev/null +++ b/contrib/python/yarl/tests/ya.make @@ -0,0 +1,24 @@ +PY3TEST() + +PEERDIR( + contrib/python/yarl +) + +TEST_SRCS( + test_cache.py + test_cached_property.py + test_normalize_path.py + test_pickle.py + test_quoting.py + test_update_query.py + test_url.py + test_url_build.py + test_url_cmp_and_hash.py + test_url_parsing.py + test_url_query.py + test_url_update_netloc.py +) + +NO_LINT() + +END() diff --git a/contrib/python/yarl/ya.make b/contrib/python/yarl/ya.make new file mode 100644 index 0000000000..0c3d0ce434 --- /dev/null +++ b/contrib/python/yarl/ya.make @@ -0,0 +1,41 @@ +# Generated by devtools/yamaker (pypi). + +PY3_LIBRARY() + +VERSION(1.9.3) + +LICENSE(Apache-2.0) + +PEERDIR( + contrib/python/idna + contrib/python/multidict +) + +NO_COMPILER_WARNINGS() + +NO_LINT() + +PY_SRCS( + TOP_LEVEL + yarl/__init__.py + yarl/__init__.pyi + yarl/_quoting.py + yarl/_quoting_c.pyi + yarl/_quoting_py.py + yarl/_url.py + CYTHON_C + yarl/_quoting_c.pyx +) + +RESOURCE_FILES( + PREFIX contrib/python/yarl/ + .dist-info/METADATA + .dist-info/top_level.txt + yarl/py.typed +) + +END() + +RECURSE_FOR_TESTS( + tests +) diff --git a/contrib/python/yarl/yarl/__init__.py b/contrib/python/yarl/yarl/__init__.py new file mode 100644 index 0000000000..f43aecbc92 --- /dev/null +++ b/contrib/python/yarl/yarl/__init__.py @@ -0,0 +1,5 @@ +from ._url import URL, cache_clear, cache_configure, cache_info + +__version__ = "1.9.3" + +__all__ = ("URL", "cache_clear", "cache_configure", "cache_info") diff --git a/contrib/python/yarl/yarl/_quoting.py b/contrib/python/yarl/yarl/_quoting.py new file mode 100644 index 0000000000..8d1c705ff2 --- /dev/null +++ b/contrib/python/yarl/yarl/_quoting.py @@ -0,0 +1,18 @@ +import os +import sys + +__all__ = ("_Quoter", "_Unquoter") + + +NO_EXTENSIONS = bool(os.environ.get("YARL_NO_EXTENSIONS")) # type: bool +if sys.implementation.name != "cpython": + NO_EXTENSIONS = True + + +if not NO_EXTENSIONS: # pragma: no branch + try: + from ._quoting_c import _Quoter, _Unquoter # type: ignore[assignment] + except ImportError: # pragma: no cover + from ._quoting_py import _Quoter, _Unquoter # type: ignore[assignment] +else: + from ._quoting_py import _Quoter, _Unquoter # type: ignore[assignment] diff --git a/contrib/python/yarl/yarl/_quoting_c.pyx b/contrib/python/yarl/yarl/_quoting_c.pyx new file mode 100644 index 0000000000..5335d17365 --- /dev/null +++ b/contrib/python/yarl/yarl/_quoting_c.pyx @@ -0,0 +1,371 @@ +# cython: language_level=3 + +from cpython.exc cimport PyErr_NoMemory +from cpython.mem cimport PyMem_Free, PyMem_Malloc, PyMem_Realloc +from cpython.unicode cimport PyUnicode_DecodeASCII, PyUnicode_DecodeUTF8Stateful +from libc.stdint cimport uint8_t, uint64_t +from libc.string cimport memcpy, memset + +from string import ascii_letters, digits + + +cdef str GEN_DELIMS = ":/?#[]@" +cdef str SUB_DELIMS_WITHOUT_QS = "!$'()*," +cdef str SUB_DELIMS = SUB_DELIMS_WITHOUT_QS + '+?=;' +cdef str RESERVED = GEN_DELIMS + SUB_DELIMS +cdef str UNRESERVED = ascii_letters + digits + '-._~' +cdef str ALLOWED = UNRESERVED + SUB_DELIMS_WITHOUT_QS +cdef str QS = '+&=;' + +DEF BUF_SIZE = 8 * 1024 # 8KiB +cdef char BUFFER[BUF_SIZE] + +cdef inline Py_UCS4 _to_hex(uint8_t v): + if v < 10: + return <Py_UCS4>(v+0x30) # ord('0') == 0x30 + else: + return <Py_UCS4>(v+0x41-10) # ord('A') == 0x41 + + +cdef inline int _from_hex(Py_UCS4 v): + if '0' <= v <= '9': + return <int>(v) - 0x30 # ord('0') == 0x30 + elif 'A' <= v <= 'F': + return <int>(v) - 0x41 + 10 # ord('A') == 0x41 + elif 'a' <= v <= 'f': + return <int>(v) - 0x61 + 10 # ord('a') == 0x61 + else: + return -1 + + +cdef inline int _is_lower_hex(Py_UCS4 v): + return 'a' <= v <= 'f' + + +cdef inline Py_UCS4 _restore_ch(Py_UCS4 d1, Py_UCS4 d2): + cdef int digit1 = _from_hex(d1) + if digit1 < 0: + return <Py_UCS4>-1 + cdef int digit2 = _from_hex(d2) + if digit2 < 0: + return <Py_UCS4>-1 + return <Py_UCS4>(digit1 << 4 | digit2) + + +cdef uint8_t ALLOWED_TABLE[16] +cdef uint8_t ALLOWED_NOTQS_TABLE[16] + + +cdef inline bint bit_at(uint8_t array[], uint64_t ch): + return array[ch >> 3] & (1 << (ch & 7)) + + +cdef inline void set_bit(uint8_t array[], uint64_t ch): + array[ch >> 3] |= (1 << (ch & 7)) + + +memset(ALLOWED_TABLE, 0, sizeof(ALLOWED_TABLE)) +memset(ALLOWED_NOTQS_TABLE, 0, sizeof(ALLOWED_NOTQS_TABLE)) + +for i in range(128): + if chr(i) in ALLOWED: + set_bit(ALLOWED_TABLE, i) + set_bit(ALLOWED_NOTQS_TABLE, i) + if chr(i) in QS: + set_bit(ALLOWED_NOTQS_TABLE, i) + +# ----------------- writer --------------------------- + +cdef struct Writer: + char *buf + Py_ssize_t size + Py_ssize_t pos + bint changed + + +cdef inline void _init_writer(Writer* writer): + writer.buf = &BUFFER[0] + writer.size = BUF_SIZE + writer.pos = 0 + writer.changed = 0 + + +cdef inline void _release_writer(Writer* writer): + if writer.buf != BUFFER: + PyMem_Free(writer.buf) + + +cdef inline int _write_char(Writer* writer, Py_UCS4 ch, bint changed): + cdef char * buf + cdef Py_ssize_t size + + if writer.pos == writer.size: + # reallocate + size = writer.size + BUF_SIZE + if writer.buf == BUFFER: + buf = <char*>PyMem_Malloc(size) + if buf == NULL: + PyErr_NoMemory() + return -1 + memcpy(buf, writer.buf, writer.size) + else: + buf = <char*>PyMem_Realloc(writer.buf, size) + if buf == NULL: + PyErr_NoMemory() + return -1 + writer.buf = buf + writer.size = size + writer.buf[writer.pos] = <char>ch + writer.pos += 1 + writer.changed |= changed + return 0 + + +cdef inline int _write_pct(Writer* writer, uint8_t ch, bint changed): + if _write_char(writer, '%', changed) < 0: + return -1 + if _write_char(writer, _to_hex(<uint8_t>ch >> 4), changed) < 0: + return -1 + return _write_char(writer, _to_hex(<uint8_t>ch & 0x0f), changed) + + +cdef inline int _write_utf8(Writer* writer, Py_UCS4 symbol): + cdef uint64_t utf = <uint64_t> symbol + + if utf < 0x80: + return _write_pct(writer, <uint8_t>utf, True) + elif utf < 0x800: + if _write_pct(writer, <uint8_t>(0xc0 | (utf >> 6)), True) < 0: + return -1 + return _write_pct(writer, <uint8_t>(0x80 | (utf & 0x3f)), True) + elif 0xD800 <= utf <= 0xDFFF: + # surogate pair, ignored + return 0 + elif utf < 0x10000: + if _write_pct(writer, <uint8_t>(0xe0 | (utf >> 12)), True) < 0: + return -1 + if _write_pct(writer, <uint8_t>(0x80 | ((utf >> 6) & 0x3f)), + True) < 0: + return -1 + return _write_pct(writer, <uint8_t>(0x80 | (utf & 0x3f)), True) + elif utf > 0x10FFFF: + # symbol is too large + return 0 + else: + if _write_pct(writer, <uint8_t>(0xf0 | (utf >> 18)), True) < 0: + return -1 + if _write_pct(writer, <uint8_t>(0x80 | ((utf >> 12) & 0x3f)), + True) < 0: + return -1 + if _write_pct(writer, <uint8_t>(0x80 | ((utf >> 6) & 0x3f)), + True) < 0: + return -1 + return _write_pct(writer, <uint8_t>(0x80 | (utf & 0x3f)), True) + + +# --------------------- end writer -------------------------- + + +cdef class _Quoter: + cdef bint _qs + cdef bint _requote + + cdef uint8_t _safe_table[16] + cdef uint8_t _protected_table[16] + + def __init__( + self, *, str safe='', str protected='', bint qs=False, bint requote=True, + ): + cdef Py_UCS4 ch + + self._qs = qs + self._requote = requote + + if not self._qs: + memcpy(self._safe_table, + ALLOWED_NOTQS_TABLE, + sizeof(self._safe_table)) + else: + memcpy(self._safe_table, + ALLOWED_TABLE, + sizeof(self._safe_table)) + for ch in safe: + if ord(ch) > 127: + raise ValueError("Only safe symbols with ORD < 128 are allowed") + set_bit(self._safe_table, ch) + + memset(self._protected_table, 0, sizeof(self._protected_table)) + for ch in protected: + if ord(ch) > 127: + raise ValueError("Only safe symbols with ORD < 128 are allowed") + set_bit(self._safe_table, ch) + set_bit(self._protected_table, ch) + + def __call__(self, val): + cdef Writer writer + if val is None: + return None + if type(val) is not str: + if isinstance(val, str): + # derived from str + val = str(val) + else: + raise TypeError("Argument should be str") + _init_writer(&writer) + try: + return self._do_quote(<str>val, &writer) + finally: + _release_writer(&writer) + + cdef str _do_quote(self, str val, Writer *writer): + cdef Py_UCS4 ch + cdef int changed + cdef int idx = 0 + cdef int length = len(val) + + while idx < length: + ch = val[idx] + idx += 1 + if ch == '%' and self._requote and idx <= length - 2: + ch = _restore_ch(val[idx], val[idx + 1]) + if ch != <Py_UCS4>-1: + idx += 2 + if ch < 128: + if bit_at(self._protected_table, ch): + if _write_pct(writer, ch, True) < 0: + raise + continue + + if bit_at(self._safe_table, ch): + if _write_char(writer, ch, True) < 0: + raise + continue + + changed = (_is_lower_hex(val[idx - 2]) or + _is_lower_hex(val[idx - 1])) + if _write_pct(writer, ch, changed) < 0: + raise + continue + else: + ch = '%' + + if self._write(writer, ch) < 0: + raise + + if not writer.changed: + return val + else: + return PyUnicode_DecodeASCII(writer.buf, writer.pos, "strict") + + cdef inline int _write(self, Writer *writer, Py_UCS4 ch): + if self._qs: + if ch == ' ': + return _write_char(writer, '+', True) + + if ch < 128 and bit_at(self._safe_table, ch): + return _write_char(writer, ch, False) + + return _write_utf8(writer, ch) + + +cdef class _Unquoter: + cdef str _unsafe + cdef bint _qs + cdef _Quoter _quoter + cdef _Quoter _qs_quoter + + def __init__(self, *, unsafe='', qs=False): + self._unsafe = unsafe + self._qs = qs + self._quoter = _Quoter() + self._qs_quoter = _Quoter(qs=True) + + def __call__(self, val): + if val is None: + return None + if type(val) is not str: + if isinstance(val, str): + # derived from str + val = str(val) + else: + raise TypeError("Argument should be str") + return self._do_unquote(<str>val) + + cdef str _do_unquote(self, str val): + if len(val) == 0: + return val + cdef list ret = [] + cdef char buffer[4] + cdef Py_ssize_t buflen = 0 + cdef Py_ssize_t consumed + cdef str unquoted + cdef Py_UCS4 ch = 0 + cdef Py_ssize_t idx = 0 + cdef Py_ssize_t length = len(val) + cdef Py_ssize_t start_pct + + while idx < length: + ch = val[idx] + idx += 1 + if ch == '%' and idx <= length - 2: + ch = _restore_ch(val[idx], val[idx + 1]) + if ch != <Py_UCS4>-1: + idx += 2 + assert buflen < 4 + buffer[buflen] = ch + buflen += 1 + try: + unquoted = PyUnicode_DecodeUTF8Stateful(buffer, buflen, + NULL, &consumed) + except UnicodeDecodeError: + start_pct = idx - buflen * 3 + buffer[0] = ch + buflen = 1 + ret.append(val[start_pct : idx - 3]) + try: + unquoted = PyUnicode_DecodeUTF8Stateful(buffer, buflen, + NULL, &consumed) + except UnicodeDecodeError: + buflen = 0 + ret.append(val[idx - 3 : idx]) + continue + if not unquoted: + assert consumed == 0 + continue + assert consumed == buflen + buflen = 0 + if self._qs and unquoted in '+=&;': + ret.append(self._qs_quoter(unquoted)) + elif unquoted in self._unsafe: + ret.append(self._quoter(unquoted)) + else: + ret.append(unquoted) + continue + else: + ch = '%' + + if buflen: + start_pct = idx - 1 - buflen * 3 + ret.append(val[start_pct : idx - 1]) + buflen = 0 + + if ch == '+': + if not self._qs or ch in self._unsafe: + ret.append('+') + else: + ret.append(' ') + continue + + if ch in self._unsafe: + ret.append('%') + h = hex(ord(ch)).upper()[2:] + for ch in h: + ret.append(ch) + continue + + ret.append(ch) + + if buflen: + ret.append(val[length - buflen * 3 : length]) + + return ''.join(ret) diff --git a/contrib/python/yarl/yarl/_quoting_py.py b/contrib/python/yarl/yarl/_quoting_py.py new file mode 100644 index 0000000000..585a1da804 --- /dev/null +++ b/contrib/python/yarl/yarl/_quoting_py.py @@ -0,0 +1,197 @@ +import codecs +import re +from string import ascii_letters, ascii_lowercase, digits +from typing import Optional, cast + +BASCII_LOWERCASE = ascii_lowercase.encode("ascii") +BPCT_ALLOWED = {f"%{i:02X}".encode("ascii") for i in range(256)} +GEN_DELIMS = ":/?#[]@" +SUB_DELIMS_WITHOUT_QS = "!$'()*," +SUB_DELIMS = SUB_DELIMS_WITHOUT_QS + "+&=;" +RESERVED = GEN_DELIMS + SUB_DELIMS +UNRESERVED = ascii_letters + digits + "-._~" +ALLOWED = UNRESERVED + SUB_DELIMS_WITHOUT_QS + + +_IS_HEX = re.compile(b"[A-Z0-9][A-Z0-9]") +_IS_HEX_STR = re.compile("[A-Fa-f0-9][A-Fa-f0-9]") + +utf8_decoder = codecs.getincrementaldecoder("utf-8") + + +class _Quoter: + def __init__( + self, + *, + safe: str = "", + protected: str = "", + qs: bool = False, + requote: bool = True, + ) -> None: + self._safe = safe + self._protected = protected + self._qs = qs + self._requote = requote + + def __call__(self, val: Optional[str]) -> Optional[str]: + if val is None: + return None + if not isinstance(val, str): + raise TypeError("Argument should be str") + if not val: + return "" + bval = cast(str, val).encode("utf8", errors="ignore") + ret = bytearray() + pct = bytearray() + safe = self._safe + safe += ALLOWED + if not self._qs: + safe += "+&=;" + safe += self._protected + bsafe = safe.encode("ascii") + idx = 0 + while idx < len(bval): + ch = bval[idx] + idx += 1 + + if pct: + if ch in BASCII_LOWERCASE: + ch = ch - 32 # convert to uppercase + pct.append(ch) + if len(pct) == 3: # pragma: no branch # peephole optimizer + buf = pct[1:] + if not _IS_HEX.match(buf): + ret.extend(b"%25") + pct.clear() + idx -= 2 + continue + try: + unquoted = chr(int(pct[1:].decode("ascii"), base=16)) + except ValueError: + ret.extend(b"%25") + pct.clear() + idx -= 2 + continue + + if unquoted in self._protected: + ret.extend(pct) + elif unquoted in safe: + ret.append(ord(unquoted)) + else: + ret.extend(pct) + pct.clear() + + # special case, if we have only one char after "%" + elif len(pct) == 2 and idx == len(bval): + ret.extend(b"%25") + pct.clear() + idx -= 1 + + continue + + elif ch == ord("%") and self._requote: + pct.clear() + pct.append(ch) + + # special case if "%" is last char + if idx == len(bval): + ret.extend(b"%25") + + continue + + if self._qs: + if ch == ord(" "): + ret.append(ord("+")) + continue + if ch in bsafe: + ret.append(ch) + continue + + ret.extend((f"%{ch:02X}").encode("ascii")) + + ret2 = ret.decode("ascii") + if ret2 == val: + return val + return ret2 + + +class _Unquoter: + def __init__(self, *, unsafe: str = "", qs: bool = False) -> None: + self._unsafe = unsafe + self._qs = qs + self._quoter = _Quoter() + self._qs_quoter = _Quoter(qs=True) + + def __call__(self, val: Optional[str]) -> Optional[str]: + if val is None: + return None + if not isinstance(val, str): + raise TypeError("Argument should be str") + if not val: + return "" + decoder = cast(codecs.BufferedIncrementalDecoder, utf8_decoder()) + ret = [] + idx = 0 + while idx < len(val): + ch = val[idx] + idx += 1 + if ch == "%" and idx <= len(val) - 2: + pct = val[idx : idx + 2] + if _IS_HEX_STR.fullmatch(pct): + b = bytes([int(pct, base=16)]) + idx += 2 + try: + unquoted = decoder.decode(b) + except UnicodeDecodeError: + start_pct = idx - 3 - len(decoder.buffer) * 3 + ret.append(val[start_pct : idx - 3]) + decoder.reset() + try: + unquoted = decoder.decode(b) + except UnicodeDecodeError: + ret.append(val[idx - 3 : idx]) + continue + if not unquoted: + continue + if self._qs and unquoted in "+=&;": + to_add = self._qs_quoter(unquoted) + if to_add is None: # pragma: no cover + raise RuntimeError("Cannot quote None") + ret.append(to_add) + elif unquoted in self._unsafe: + to_add = self._quoter(unquoted) + if to_add is None: # pragma: no cover + raise RuntimeError("Cannot quote None") + ret.append(to_add) + else: + ret.append(unquoted) + continue + + if decoder.buffer: + start_pct = idx - 1 - len(decoder.buffer) * 3 + ret.append(val[start_pct : idx - 1]) + decoder.reset() + + if ch == "+": + if not self._qs or ch in self._unsafe: + ret.append("+") + else: + ret.append(" ") + continue + + if ch in self._unsafe: + ret.append("%") + h = hex(ord(ch)).upper()[2:] + for ch in h: + ret.append(ch) + continue + + ret.append(ch) + + if decoder.buffer: + ret.append(val[-len(decoder.buffer) * 3 :]) + + ret2 = "".join(ret) + if ret2 == val: + return val + return ret2 diff --git a/contrib/python/yarl/yarl/_url.py b/contrib/python/yarl/yarl/_url.py new file mode 100644 index 0000000000..c8f2acb39b --- /dev/null +++ b/contrib/python/yarl/yarl/_url.py @@ -0,0 +1,1198 @@ +import functools +import math +import warnings +from collections.abc import Mapping, Sequence +from contextlib import suppress +from ipaddress import ip_address +from urllib.parse import SplitResult, parse_qsl, quote, urljoin, urlsplit, urlunsplit + +import idna +from multidict import MultiDict, MultiDictProxy + +from ._quoting import _Quoter, _Unquoter + +DEFAULT_PORTS = {"http": 80, "https": 443, "ws": 80, "wss": 443} + +sentinel = object() + + +def rewrite_module(obj: object) -> object: + obj.__module__ = "yarl" + return obj + + +class cached_property: + """Use as a class method decorator. It operates almost exactly like + the Python `@property` decorator, but it puts the result of the + method it decorates into the instance dict after the first call, + effectively replacing the function it decorates with an instance + variable. It is, in Python parlance, a data descriptor. + + """ + + def __init__(self, wrapped): + self.wrapped = wrapped + try: + self.__doc__ = wrapped.__doc__ + except AttributeError: # pragma: no cover + self.__doc__ = "" + self.name = wrapped.__name__ + + def __get__(self, inst, owner, _sentinel=sentinel): + if inst is None: + return self + val = inst._cache.get(self.name, _sentinel) + if val is not _sentinel: + return val + val = self.wrapped(inst) + inst._cache[self.name] = val + return val + + def __set__(self, inst, value): + raise AttributeError("cached property is read-only") + + +def _normalize_path_segments(segments): + """Drop '.' and '..' from a sequence of str segments""" + + resolved_path = [] + + for seg in segments: + if seg == "..": + # ignore any .. segments that would otherwise cause an + # IndexError when popped from resolved_path if + # resolving for rfc3986 + with suppress(IndexError): + resolved_path.pop() + elif seg != ".": + resolved_path.append(seg) + + if segments and segments[-1] in (".", ".."): + # do some post-processing here. + # if the last segment was a relative dir, + # then we need to append the trailing '/' + resolved_path.append("") + + return resolved_path + + +@rewrite_module +class URL: + # Don't derive from str + # follow pathlib.Path design + # probably URL will not suffer from pathlib problems: + # it's intended for libraries like aiohttp, + # not to be passed into standard library functions like os.open etc. + + # URL grammar (RFC 3986) + # pct-encoded = "%" HEXDIG HEXDIG + # reserved = gen-delims / sub-delims + # gen-delims = ":" / "/" / "?" / "#" / "[" / "]" / "@" + # sub-delims = "!" / "$" / "&" / "'" / "(" / ")" + # / "*" / "+" / "," / ";" / "=" + # unreserved = ALPHA / DIGIT / "-" / "." / "_" / "~" + # URI = scheme ":" hier-part [ "?" query ] [ "#" fragment ] + # hier-part = "//" authority path-abempty + # / path-absolute + # / path-rootless + # / path-empty + # scheme = ALPHA *( ALPHA / DIGIT / "+" / "-" / "." ) + # authority = [ userinfo "@" ] host [ ":" port ] + # userinfo = *( unreserved / pct-encoded / sub-delims / ":" ) + # host = IP-literal / IPv4address / reg-name + # IP-literal = "[" ( IPv6address / IPvFuture ) "]" + # IPvFuture = "v" 1*HEXDIG "." 1*( unreserved / sub-delims / ":" ) + # IPv6address = 6( h16 ":" ) ls32 + # / "::" 5( h16 ":" ) ls32 + # / [ h16 ] "::" 4( h16 ":" ) ls32 + # / [ *1( h16 ":" ) h16 ] "::" 3( h16 ":" ) ls32 + # / [ *2( h16 ":" ) h16 ] "::" 2( h16 ":" ) ls32 + # / [ *3( h16 ":" ) h16 ] "::" h16 ":" ls32 + # / [ *4( h16 ":" ) h16 ] "::" ls32 + # / [ *5( h16 ":" ) h16 ] "::" h16 + # / [ *6( h16 ":" ) h16 ] "::" + # ls32 = ( h16 ":" h16 ) / IPv4address + # ; least-significant 32 bits of address + # h16 = 1*4HEXDIG + # ; 16 bits of address represented in hexadecimal + # IPv4address = dec-octet "." dec-octet "." dec-octet "." dec-octet + # dec-octet = DIGIT ; 0-9 + # / %x31-39 DIGIT ; 10-99 + # / "1" 2DIGIT ; 100-199 + # / "2" %x30-34 DIGIT ; 200-249 + # / "25" %x30-35 ; 250-255 + # reg-name = *( unreserved / pct-encoded / sub-delims ) + # port = *DIGIT + # path = path-abempty ; begins with "/" or is empty + # / path-absolute ; begins with "/" but not "//" + # / path-noscheme ; begins with a non-colon segment + # / path-rootless ; begins with a segment + # / path-empty ; zero characters + # path-abempty = *( "/" segment ) + # path-absolute = "/" [ segment-nz *( "/" segment ) ] + # path-noscheme = segment-nz-nc *( "/" segment ) + # path-rootless = segment-nz *( "/" segment ) + # path-empty = 0<pchar> + # segment = *pchar + # segment-nz = 1*pchar + # segment-nz-nc = 1*( unreserved / pct-encoded / sub-delims / "@" ) + # ; non-zero-length segment without any colon ":" + # pchar = unreserved / pct-encoded / sub-delims / ":" / "@" + # query = *( pchar / "/" / "?" ) + # fragment = *( pchar / "/" / "?" ) + # URI-reference = URI / relative-ref + # relative-ref = relative-part [ "?" query ] [ "#" fragment ] + # relative-part = "//" authority path-abempty + # / path-absolute + # / path-noscheme + # / path-empty + # absolute-URI = scheme ":" hier-part [ "?" query ] + __slots__ = ("_cache", "_val") + + _QUOTER = _Quoter(requote=False) + _REQUOTER = _Quoter() + _PATH_QUOTER = _Quoter(safe="@:", protected="/+", requote=False) + _PATH_REQUOTER = _Quoter(safe="@:", protected="/+") + _QUERY_QUOTER = _Quoter(safe="?/:@", protected="=+&;", qs=True, requote=False) + _QUERY_REQUOTER = _Quoter(safe="?/:@", protected="=+&;", qs=True) + _QUERY_PART_QUOTER = _Quoter(safe="?/:@", qs=True, requote=False) + _FRAGMENT_QUOTER = _Quoter(safe="?/:@", requote=False) + _FRAGMENT_REQUOTER = _Quoter(safe="?/:@") + + _UNQUOTER = _Unquoter() + _PATH_UNQUOTER = _Unquoter(unsafe="+") + _QS_UNQUOTER = _Unquoter(qs=True) + + def __new__(cls, val="", *, encoded=False, strict=None): + if strict is not None: # pragma: no cover + warnings.warn("strict parameter is ignored") + if type(val) is cls: + return val + if type(val) is str: + val = urlsplit(val) + elif type(val) is SplitResult: + if not encoded: + raise ValueError("Cannot apply decoding to SplitResult") + elif isinstance(val, str): + val = urlsplit(str(val)) + else: + raise TypeError("Constructor parameter should be str") + + if not encoded: + if not val[1]: # netloc + netloc = "" + host = "" + else: + host = val.hostname + if host is None: + raise ValueError("Invalid URL: host is required for absolute urls") + + try: + port = val.port + except ValueError as e: + raise ValueError( + "Invalid URL: port can't be converted to integer" + ) from e + + netloc = cls._make_netloc( + val.username, val.password, host, port, encode=True, requote=True + ) + path = cls._PATH_REQUOTER(val[2]) + if netloc: + path = cls._normalize_path(path) + + cls._validate_authority_uri_abs_path(host=host, path=path) + query = cls._QUERY_REQUOTER(val[3]) + fragment = cls._FRAGMENT_REQUOTER(val[4]) + val = SplitResult(val[0], netloc, path, query, fragment) + + self = object.__new__(cls) + self._val = val + self._cache = {} + return self + + @classmethod + def build( + cls, + *, + scheme="", + authority="", + user=None, + password=None, + host="", + port=None, + path="", + query=None, + query_string="", + fragment="", + encoded=False, + ): + """Creates and returns a new URL""" + + if authority and (user or password or host or port): + raise ValueError( + 'Can\'t mix "authority" with "user", "password", "host" or "port".' + ) + if port and not host: + raise ValueError('Can\'t build URL with "port" but without "host".') + if query and query_string: + raise ValueError('Only one of "query" or "query_string" should be passed') + if ( + scheme is None + or authority is None + or host is None + or path is None + or query_string is None + or fragment is None + ): + raise TypeError( + 'NoneType is illegal for "scheme", "authority", "host", "path", ' + '"query_string", and "fragment" args, use empty string instead.' + ) + + if authority: + if encoded: + netloc = authority + else: + tmp = SplitResult("", authority, "", "", "") + netloc = cls._make_netloc( + tmp.username, tmp.password, tmp.hostname, tmp.port, encode=True + ) + elif not user and not password and not host and not port: + netloc = "" + else: + netloc = cls._make_netloc( + user, password, host, port, encode=not encoded, encode_host=not encoded + ) + if not encoded: + path = cls._PATH_QUOTER(path) + if netloc: + path = cls._normalize_path(path) + + cls._validate_authority_uri_abs_path(host=host, path=path) + query_string = cls._QUERY_QUOTER(query_string) + fragment = cls._FRAGMENT_QUOTER(fragment) + + url = cls( + SplitResult(scheme, netloc, path, query_string, fragment), encoded=True + ) + + if query: + return url.with_query(query) + else: + return url + + def __init_subclass__(cls): + raise TypeError(f"Inheriting a class {cls!r} from URL is forbidden") + + def __str__(self): + val = self._val + if not val.path and self.is_absolute() and (val.query or val.fragment): + val = val._replace(path="/") + return urlunsplit(val) + + def __repr__(self): + return f"{self.__class__.__name__}('{str(self)}')" + + def __bytes__(self): + return str(self).encode("ascii") + + def __eq__(self, other): + if not type(other) is URL: + return NotImplemented + + val1 = self._val + if not val1.path and self.is_absolute(): + val1 = val1._replace(path="/") + + val2 = other._val + if not val2.path and other.is_absolute(): + val2 = val2._replace(path="/") + + return val1 == val2 + + def __hash__(self): + ret = self._cache.get("hash") + if ret is None: + val = self._val + if not val.path and self.is_absolute(): + val = val._replace(path="/") + ret = self._cache["hash"] = hash(val) + return ret + + def __le__(self, other): + if not type(other) is URL: + return NotImplemented + return self._val <= other._val + + def __lt__(self, other): + if not type(other) is URL: + return NotImplemented + return self._val < other._val + + def __ge__(self, other): + if not type(other) is URL: + return NotImplemented + return self._val >= other._val + + def __gt__(self, other): + if not type(other) is URL: + return NotImplemented + return self._val > other._val + + def __truediv__(self, name): + if not isinstance(name, str): + return NotImplemented + return self._make_child((str(name),)) + + def __mod__(self, query): + return self.update_query(query) + + def __bool__(self) -> bool: + return bool( + self._val.netloc or self._val.path or self._val.query or self._val.fragment + ) + + def __getstate__(self): + return (self._val,) + + def __setstate__(self, state): + if state[0] is None and isinstance(state[1], dict): + # default style pickle + self._val = state[1]["_val"] + else: + self._val, *unused = state + self._cache = {} + + def is_absolute(self): + """A check for absolute URLs. + + Return True for absolute ones (having scheme or starting + with //), False otherwise. + + """ + return self.raw_host is not None + + def is_default_port(self): + """A check for default port. + + Return True if port is default for specified scheme, + e.g. 'http://python.org' or 'http://python.org:80', False + otherwise. + + """ + if self.port is None: + return False + default = DEFAULT_PORTS.get(self.scheme) + if default is None: + return False + return self.port == default + + def origin(self): + """Return an URL with scheme, host and port parts only. + + user, password, path, query and fragment are removed. + + """ + # TODO: add a keyword-only option for keeping user/pass maybe? + if not self.is_absolute(): + raise ValueError("URL should be absolute") + if not self._val.scheme: + raise ValueError("URL should have scheme") + v = self._val + netloc = self._make_netloc(None, None, v.hostname, v.port) + val = v._replace(netloc=netloc, path="", query="", fragment="") + return URL(val, encoded=True) + + def relative(self): + """Return a relative part of the URL. + + scheme, user, password, host and port are removed. + + """ + if not self.is_absolute(): + raise ValueError("URL should be absolute") + val = self._val._replace(scheme="", netloc="") + return URL(val, encoded=True) + + @property + def scheme(self): + """Scheme for absolute URLs. + + Empty string for relative URLs or URLs starting with // + + """ + return self._val.scheme + + @property + def raw_authority(self): + """Encoded authority part of URL. + + Empty string for relative URLs. + + """ + return self._val.netloc + + @cached_property + def authority(self): + """Decoded authority part of URL. + + Empty string for relative URLs. + + """ + return self._make_netloc( + self.user, self.password, self.host, self.port, encode_host=False + ) + + @property + def raw_user(self): + """Encoded user part of URL. + + None if user is missing. + + """ + # not .username + ret = self._val.username + if not ret: + return None + return ret + + @cached_property + def user(self): + """Decoded user part of URL. + + None if user is missing. + + """ + return self._UNQUOTER(self.raw_user) + + @property + def raw_password(self): + """Encoded password part of URL. + + None if password is missing. + + """ + return self._val.password + + @cached_property + def password(self): + """Decoded password part of URL. + + None if password is missing. + + """ + return self._UNQUOTER(self.raw_password) + + @property + def raw_host(self): + """Encoded host part of URL. + + None for relative URLs. + + """ + # Use host instead of hostname for sake of shortness + # May add .hostname prop later + return self._val.hostname + + @cached_property + def host(self): + """Decoded host part of URL. + + None for relative URLs. + + """ + raw = self.raw_host + if raw is None: + return None + if "%" in raw: + # Hack for scoped IPv6 addresses like + # fe80::2%Проверка + # presence of '%' sign means only IPv6 address, so idna is useless. + return raw + return _idna_decode(raw) + + @property + def port(self): + """Port part of URL, with scheme-based fallback. + + None for relative URLs or URLs without explicit port and + scheme without default port substitution. + + """ + return self._val.port or DEFAULT_PORTS.get(self._val.scheme) + + @property + def explicit_port(self): + """Port part of URL, without scheme-based fallback. + + None for relative URLs or URLs without explicit port. + + """ + return self._val.port + + @property + def raw_path(self): + """Encoded path of URL. + + / for absolute URLs without path part. + + """ + ret = self._val.path + if not ret and self.is_absolute(): + ret = "/" + return ret + + @cached_property + def path(self): + """Decoded path of URL. + + / for absolute URLs without path part. + + """ + return self._PATH_UNQUOTER(self.raw_path) + + @cached_property + def query(self): + """A MultiDictProxy representing parsed query parameters in decoded + representation. + + Empty value if URL has no query part. + + """ + ret = MultiDict(parse_qsl(self.raw_query_string, keep_blank_values=True)) + return MultiDictProxy(ret) + + @property + def raw_query_string(self): + """Encoded query part of URL. + + Empty string if query is missing. + + """ + return self._val.query + + @cached_property + def query_string(self): + """Decoded query part of URL. + + Empty string if query is missing. + + """ + return self._QS_UNQUOTER(self.raw_query_string) + + @cached_property + def path_qs(self): + """Decoded path of URL with query.""" + if not self.query_string: + return self.path + return f"{self.path}?{self.query_string}" + + @cached_property + def raw_path_qs(self): + """Encoded path of URL with query.""" + if not self.raw_query_string: + return self.raw_path + return f"{self.raw_path}?{self.raw_query_string}" + + @property + def raw_fragment(self): + """Encoded fragment part of URL. + + Empty string if fragment is missing. + + """ + return self._val.fragment + + @cached_property + def fragment(self): + """Decoded fragment part of URL. + + Empty string if fragment is missing. + + """ + return self._UNQUOTER(self.raw_fragment) + + @cached_property + def raw_parts(self): + """A tuple containing encoded *path* parts. + + ('/',) for absolute URLs if *path* is missing. + + """ + path = self._val.path + if self.is_absolute(): + if not path: + parts = ["/"] + else: + parts = ["/"] + path[1:].split("/") + else: + if path.startswith("/"): + parts = ["/"] + path[1:].split("/") + else: + parts = path.split("/") + return tuple(parts) + + @cached_property + def parts(self): + """A tuple containing decoded *path* parts. + + ('/',) for absolute URLs if *path* is missing. + + """ + return tuple(self._UNQUOTER(part) for part in self.raw_parts) + + @cached_property + def parent(self): + """A new URL with last part of path removed and cleaned up query and + fragment. + + """ + path = self.raw_path + if not path or path == "/": + if self.raw_fragment or self.raw_query_string: + return URL(self._val._replace(query="", fragment=""), encoded=True) + return self + parts = path.split("/") + val = self._val._replace(path="/".join(parts[:-1]), query="", fragment="") + return URL(val, encoded=True) + + @cached_property + def raw_name(self): + """The last part of raw_parts.""" + parts = self.raw_parts + if self.is_absolute(): + parts = parts[1:] + if not parts: + return "" + else: + return parts[-1] + else: + return parts[-1] + + @cached_property + def name(self): + """The last part of parts.""" + return self._UNQUOTER(self.raw_name) + + @cached_property + def raw_suffix(self): + name = self.raw_name + i = name.rfind(".") + if 0 < i < len(name) - 1: + return name[i:] + else: + return "" + + @cached_property + def suffix(self): + return self._UNQUOTER(self.raw_suffix) + + @cached_property + def raw_suffixes(self): + name = self.raw_name + if name.endswith("."): + return () + name = name.lstrip(".") + return tuple("." + suffix for suffix in name.split(".")[1:]) + + @cached_property + def suffixes(self): + return tuple(self._UNQUOTER(suffix) for suffix in self.raw_suffixes) + + @staticmethod + def _validate_authority_uri_abs_path(host, path): + """Ensure that path in URL with authority starts with a leading slash. + + Raise ValueError if not. + """ + if len(host) > 0 and len(path) > 0 and not path.startswith("/"): + raise ValueError( + "Path in a URL with authority should start with a slash ('/') if set" + ) + + def _make_child(self, segments, encoded=False): + """add segments to self._val.path, accounting for absolute vs relative paths""" + # keep the trailing slash if the last segment ends with / + parsed = [""] if segments and segments[-1][-1:] == "/" else [] + for seg in reversed(segments): + if not seg: + continue + if seg[0] == "/": + raise ValueError( + f"Appending path {seg!r} starting from slash is forbidden" + ) + seg = seg if encoded else self._PATH_QUOTER(seg) + if "/" in seg: + parsed += ( + sub for sub in reversed(seg.split("/")) if sub and sub != "." + ) + elif seg != ".": + parsed.append(seg) + parsed.reverse() + old_path = self._val.path + if old_path: + parsed = [*old_path.rstrip("/").split("/"), *parsed] + if self.is_absolute(): + parsed = _normalize_path_segments(parsed) + if parsed and parsed[0] != "": + # inject a leading slash when adding a path to an absolute URL + # where there was none before + parsed = ["", *parsed] + new_path = "/".join(parsed) + return URL( + self._val._replace(path=new_path, query="", fragment=""), encoded=True + ) + + @classmethod + def _normalize_path(cls, path): + # Drop '.' and '..' from str path + + prefix = "" + if path.startswith("/"): + # preserve the "/" root element of absolute paths, copying it to the + # normalised output as per sections 5.2.4 and 6.2.2.3 of rfc3986. + prefix = "/" + path = path[1:] + + segments = path.split("/") + return prefix + "/".join(_normalize_path_segments(segments)) + + @classmethod + def _encode_host(cls, host, human=False): + try: + ip, sep, zone = host.partition("%") + ip = ip_address(ip) + except ValueError: + host = host.lower() + # IDNA encoding is slow, + # skip it for ASCII-only strings + # Don't move the check into _idna_encode() helper + # to reduce the cache size + if human or host.isascii(): + return host + host = _idna_encode(host) + else: + host = ip.compressed + if sep: + host += "%" + zone + if ip.version == 6: + host = "[" + host + "]" + return host + + @classmethod + def _make_netloc( + cls, user, password, host, port, encode=False, encode_host=True, requote=False + ): + quoter = cls._REQUOTER if requote else cls._QUOTER + if encode_host: + ret = cls._encode_host(host) + else: + ret = host + if port is not None: + ret = ret + ":" + str(port) + if password is not None: + if not user: + user = "" + else: + if encode: + user = quoter(user) + if encode: + password = quoter(password) + user = user + ":" + password + elif user and encode: + user = quoter(user) + if user: + ret = user + "@" + ret + return ret + + def with_scheme(self, scheme): + """Return a new URL with scheme replaced.""" + # N.B. doesn't cleanup query/fragment + if not isinstance(scheme, str): + raise TypeError("Invalid scheme type") + if not self.is_absolute(): + raise ValueError("scheme replacement is not allowed for relative URLs") + return URL(self._val._replace(scheme=scheme.lower()), encoded=True) + + def with_user(self, user): + """Return a new URL with user replaced. + + Autoencode user if needed. + + Clear user/password if user is None. + + """ + # N.B. doesn't cleanup query/fragment + val = self._val + if user is None: + password = None + elif isinstance(user, str): + user = self._QUOTER(user) + password = val.password + else: + raise TypeError("Invalid user type") + if not self.is_absolute(): + raise ValueError("user replacement is not allowed for relative URLs") + return URL( + self._val._replace( + netloc=self._make_netloc(user, password, val.hostname, val.port) + ), + encoded=True, + ) + + def with_password(self, password): + """Return a new URL with password replaced. + + Autoencode password if needed. + + Clear password if argument is None. + + """ + # N.B. doesn't cleanup query/fragment + if password is None: + pass + elif isinstance(password, str): + password = self._QUOTER(password) + else: + raise TypeError("Invalid password type") + if not self.is_absolute(): + raise ValueError("password replacement is not allowed for relative URLs") + val = self._val + return URL( + self._val._replace( + netloc=self._make_netloc(val.username, password, val.hostname, val.port) + ), + encoded=True, + ) + + def with_host(self, host): + """Return a new URL with host replaced. + + Autoencode host if needed. + + Changing host for relative URLs is not allowed, use .join() + instead. + + """ + # N.B. doesn't cleanup query/fragment + if not isinstance(host, str): + raise TypeError("Invalid host type") + if not self.is_absolute(): + raise ValueError("host replacement is not allowed for relative URLs") + if not host: + raise ValueError("host removing is not allowed") + val = self._val + return URL( + self._val._replace( + netloc=self._make_netloc(val.username, val.password, host, val.port) + ), + encoded=True, + ) + + def with_port(self, port): + """Return a new URL with port replaced. + + Clear port to default if None is passed. + + """ + # N.B. doesn't cleanup query/fragment + if port is not None: + if isinstance(port, bool) or not isinstance(port, int): + raise TypeError(f"port should be int or None, got {type(port)}") + if port < 0 or port > 65535: + raise ValueError(f"port must be between 0 and 65535, got {port}") + if not self.is_absolute(): + raise ValueError("port replacement is not allowed for relative URLs") + val = self._val + return URL( + self._val._replace( + netloc=self._make_netloc(val.username, val.password, val.hostname, port) + ), + encoded=True, + ) + + def with_path(self, path, *, encoded=False): + """Return a new URL with path replaced.""" + if not encoded: + path = self._PATH_QUOTER(path) + if self.is_absolute(): + path = self._normalize_path(path) + if len(path) > 0 and path[0] != "/": + path = "/" + path + return URL(self._val._replace(path=path, query="", fragment=""), encoded=True) + + @classmethod + def _query_seq_pairs(cls, quoter, pairs): + for key, val in pairs: + if isinstance(val, (list, tuple)): + for v in val: + yield quoter(key) + "=" + quoter(cls._query_var(v)) + else: + yield quoter(key) + "=" + quoter(cls._query_var(val)) + + @staticmethod + def _query_var(v): + cls = type(v) + if issubclass(cls, str): + return v + if issubclass(cls, float): + if math.isinf(v): + raise ValueError("float('inf') is not supported") + if math.isnan(v): + raise ValueError("float('nan') is not supported") + return str(float(v)) + if issubclass(cls, int) and cls is not bool: + return str(int(v)) + raise TypeError( + "Invalid variable type: value " + "should be str, int or float, got {!r} " + "of type {}".format(v, cls) + ) + + def _get_str_query(self, *args, **kwargs): + if kwargs: + if len(args) > 0: + raise ValueError( + "Either kwargs or single query parameter must be present" + ) + query = kwargs + elif len(args) == 1: + query = args[0] + else: + raise ValueError("Either kwargs or single query parameter must be present") + + if query is None: + query = None + elif isinstance(query, Mapping): + quoter = self._QUERY_PART_QUOTER + query = "&".join(self._query_seq_pairs(quoter, query.items())) + elif isinstance(query, str): + query = self._QUERY_QUOTER(query) + elif isinstance(query, (bytes, bytearray, memoryview)): + raise TypeError( + "Invalid query type: bytes, bytearray and memoryview are forbidden" + ) + elif isinstance(query, Sequence): + quoter = self._QUERY_PART_QUOTER + # We don't expect sequence values if we're given a list of pairs + # already; only mappings like builtin `dict` which can't have the + # same key pointing to multiple values are allowed to use + # `_query_seq_pairs`. + query = "&".join( + quoter(k) + "=" + quoter(self._query_var(v)) for k, v in query + ) + else: + raise TypeError( + "Invalid query type: only str, mapping or " + "sequence of (key, value) pairs is allowed" + ) + + return query + + def with_query(self, *args, **kwargs): + """Return a new URL with query part replaced. + + Accepts any Mapping (e.g. dict, multidict.MultiDict instances) + or str, autoencode the argument if needed. + + A sequence of (key, value) pairs is supported as well. + + It also can take an arbitrary number of keyword arguments. + + Clear query if None is passed. + + """ + # N.B. doesn't cleanup query/fragment + + new_query = self._get_str_query(*args, **kwargs) or "" + return URL( + self._val._replace(path=self._val.path, query=new_query), encoded=True + ) + + def update_query(self, *args, **kwargs): + """Return a new URL with query part updated.""" + s = self._get_str_query(*args, **kwargs) + query = None + if s is not None: + new_query = MultiDict(parse_qsl(s, keep_blank_values=True)) + query = MultiDict(self.query) + query.update(new_query) + + return URL( + self._val._replace(query=self._get_str_query(query) or ""), encoded=True + ) + + def with_fragment(self, fragment): + """Return a new URL with fragment replaced. + + Autoencode fragment if needed. + + Clear fragment to default if None is passed. + + """ + # N.B. doesn't cleanup query/fragment + if fragment is None: + raw_fragment = "" + elif not isinstance(fragment, str): + raise TypeError("Invalid fragment type") + else: + raw_fragment = self._FRAGMENT_QUOTER(fragment) + if self.raw_fragment == raw_fragment: + return self + return URL(self._val._replace(fragment=raw_fragment), encoded=True) + + def with_name(self, name): + """Return a new URL with name (last part of path) replaced. + + Query and fragment parts are cleaned up. + + Name is encoded if needed. + + """ + # N.B. DOES cleanup query/fragment + if not isinstance(name, str): + raise TypeError("Invalid name type") + if "/" in name: + raise ValueError("Slash in name is not allowed") + name = self._PATH_QUOTER(name) + if name in (".", ".."): + raise ValueError(". and .. values are forbidden") + parts = list(self.raw_parts) + if self.is_absolute(): + if len(parts) == 1: + parts.append(name) + else: + parts[-1] = name + parts[0] = "" # replace leading '/' + else: + parts[-1] = name + if parts[0] == "/": + parts[0] = "" # replace leading '/' + return URL( + self._val._replace(path="/".join(parts), query="", fragment=""), + encoded=True, + ) + + def with_suffix(self, suffix): + """Return a new URL with suffix (file extension of name) replaced. + + Query and fragment parts are cleaned up. + + suffix is encoded if needed. + """ + if not isinstance(suffix, str): + raise TypeError("Invalid suffix type") + if suffix and not suffix.startswith(".") or suffix == ".": + raise ValueError(f"Invalid suffix {suffix!r}") + name = self.raw_name + if not name: + raise ValueError(f"{self!r} has an empty name") + old_suffix = self.raw_suffix + if not old_suffix: + name = name + suffix + else: + name = name[: -len(old_suffix)] + suffix + return self.with_name(name) + + def join(self, url): + """Join URLs + + Construct a full (“absolute”) URL by combining a “base URL” + (self) with another URL (url). + + Informally, this uses components of the base URL, in + particular the addressing scheme, the network location and + (part of) the path, to provide missing components in the + relative URL. + + """ + # See docs for urllib.parse.urljoin + if not isinstance(url, URL): + raise TypeError("url should be URL") + return URL(urljoin(str(self), str(url)), encoded=True) + + def joinpath(self, *other, encoded=False): + """Return a new URL with the elements in other appended to the path.""" + return self._make_child(other, encoded=encoded) + + def human_repr(self): + """Return decoded human readable string for URL representation.""" + user = _human_quote(self.user, "#/:?@[]") + password = _human_quote(self.password, "#/:?@[]") + host = self.host + if host: + host = self._encode_host(self.host, human=True) + path = _human_quote(self.path, "#?") + query_string = "&".join( + "{}={}".format(_human_quote(k, "#&+;="), _human_quote(v, "#&+;=")) + for k, v in self.query.items() + ) + fragment = _human_quote(self.fragment, "") + return urlunsplit( + SplitResult( + self.scheme, + self._make_netloc( + user, + password, + host, + self._val.port, + encode_host=False, + ), + path, + query_string, + fragment, + ) + ) + + +def _human_quote(s, unsafe): + if not s: + return s + for c in "%" + unsafe: + if c in s: + s = s.replace(c, f"%{ord(c):02X}") + if s.isprintable(): + return s + return "".join(c if c.isprintable() else quote(c) for c in s) + + +_MAXCACHE = 256 + + +@functools.lru_cache(_MAXCACHE) +def _idna_decode(raw): + try: + return idna.decode(raw.encode("ascii")) + except UnicodeError: # e.g. '::1' + return raw.encode("ascii").decode("idna") + + +@functools.lru_cache(_MAXCACHE) +def _idna_encode(host): + try: + return idna.encode(host, uts46=True).decode("ascii") + except UnicodeError: + return host.encode("idna").decode("ascii") + + +@rewrite_module +def cache_clear(): + _idna_decode.cache_clear() + _idna_encode.cache_clear() + + +@rewrite_module +def cache_info(): + return { + "idna_encode": _idna_encode.cache_info(), + "idna_decode": _idna_decode.cache_info(), + } + + +@rewrite_module +def cache_configure(*, idna_encode_size=_MAXCACHE, idna_decode_size=_MAXCACHE): + global _idna_decode, _idna_encode + + _idna_encode = functools.lru_cache(idna_encode_size)(_idna_encode.__wrapped__) + _idna_decode = functools.lru_cache(idna_decode_size)(_idna_decode.__wrapped__) diff --git a/contrib/python/yarl/yarl/py.typed b/contrib/python/yarl/yarl/py.typed new file mode 100644 index 0000000000..dcf2c804da --- /dev/null +++ b/contrib/python/yarl/yarl/py.typed @@ -0,0 +1 @@ +# Placeholder diff --git a/contrib/python/ydb/py2/.dist-info/METADATA b/contrib/python/ydb/py2/.dist-info/METADATA new file mode 100644 index 0000000000..1efcb264c4 --- /dev/null +++ b/contrib/python/ydb/py2/.dist-info/METADATA @@ -0,0 +1,63 @@ +Metadata-Version: 2.1 +Name: ydb +Version: 2.15.1 +Summary: YDB Python SDK +Home-page: http://github.com/ydb-platform/ydb-python-sdk +Author: Yandex LLC +Author-email: ydb@yandex-team.ru +License: Apache 2.0 +Classifier: Programming Language :: Python +Classifier: Programming Language :: Python :: 2 +Classifier: Programming Language :: Python :: 2.7 +Classifier: Programming Language :: Python :: 3 +Classifier: Programming Language :: Python :: 3.6 +Description-Content-Type: text/markdown +License-File: LICENSE +License-File: AUTHORS +Requires-Dist: aiohttp ==3.7.4 +Requires-Dist: enum-compat >=0.0.1 +Requires-Dist: grpcio >=1.5.0 +Requires-Dist: packaging +Requires-Dist: protobuf <5.0.0,>3.13.0 +Requires-Dist: pytest ==6.2.4 +Requires-Dist: six <2 +Provides-Extra: yc +Requires-Dist: yandexcloud ; extra == 'yc' + +YDB Python SDK +--- +[](https://github.com/ydb-platform/ydb/blob/main/LICENSE) +[](https://badge.fury.io/py/ydb) +[](https://github.com/ydb-platform/ydb-python-sdk/actions/workflows/tests.yaml) +[](https://github.com/ydb-platform/ydb-python-sdk/actions/workflows/style.yaml) + +Officially supported Python client for YDB. + +## Quickstart + +### Prerequisites + +- Python 3.8 or higher +- `pip` version 9.0.1 or higher + +If necessary, upgrade your version of `pip`: + +```sh +$ python -m pip install --upgrade pip +``` + +If you cannot upgrade `pip` due to a system-owned installation, you can +run the example in a virtualenv: + +```sh +$ python -m pip install virtualenv +$ virtualenv venv +$ source venv/bin/activate +$ python -m pip install --upgrade pip +``` + +Install YDB python sdk: + +```sh +$ python -m pip install ydb +``` diff --git a/contrib/python/ydb/py2/.dist-info/top_level.txt b/contrib/python/ydb/py2/.dist-info/top_level.txt new file mode 100644 index 0000000000..7298732379 --- /dev/null +++ b/contrib/python/ydb/py2/.dist-info/top_level.txt @@ -0,0 +1,2 @@ +kikimr +ydb diff --git a/contrib/python/ydb/py2/AUTHORS b/contrib/python/ydb/py2/AUTHORS new file mode 100644 index 0000000000..200343e364 --- /dev/null +++ b/contrib/python/ydb/py2/AUTHORS @@ -0,0 +1,4 @@ +The following authors have created the source code of "YDB Python SDK" +published and distributed by YANDEX LLC as the owner: + +Vitalii Gridnev <gridnevvvit@gmail.com> diff --git a/contrib/python/ydb/py2/LICENSE b/contrib/python/ydb/py2/LICENSE new file mode 100644 index 0000000000..cabac2dec9 --- /dev/null +++ b/contrib/python/ydb/py2/LICENSE @@ -0,0 +1,202 @@ +Copyright 2022 YANDEX LLC + Apache License + Version 2.0, January 2004 + http://www.apache.org/licenses/ + + TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION + + 1. Definitions. + + "License" shall mean the terms and conditions for use, reproduction, + and distribution as defined by Sections 1 through 9 of this document. + + "Licensor" shall mean the copyright owner or entity authorized by + the copyright owner that is granting the License. + + "Legal Entity" shall mean the union of the acting entity and all + other entities that control, are controlled by, or are under common + control with that entity. For the purposes of this definition, + "control" means (i) the power, direct or indirect, to cause the + direction or management of such entity, whether by contract or + otherwise, or (ii) ownership of fifty percent (50%) or more of the + outstanding shares, or (iii) beneficial ownership of such entity. + + "You" (or "Your") shall mean an individual or Legal Entity + exercising permissions granted by this License. + + "Source" form shall mean the preferred form for making modifications, + including but not limited to software source code, documentation + source, and configuration files. + + "Object" form shall mean any form resulting from mechanical + transformation or translation of a Source form, including but + not limited to compiled object code, generated documentation, + and conversions to other media types. + + "Work" shall mean the work of authorship, whether in Source or + Object form, made available under the License, as indicated by a + copyright notice that is included in or attached to the work + (an example is provided in the Appendix below). + + "Derivative Works" shall mean any work, whether in Source or Object + form, that is based on (or derived from) the Work and for which the + editorial revisions, annotations, elaborations, or other modifications + represent, as a whole, an original work of authorship. For the purposes + of this License, Derivative Works shall not include works that remain + separable from, or merely link (or bind by name) to the interfaces of, + the Work and Derivative Works thereof. + + "Contribution" shall mean any work of authorship, including + the original version of the Work and any modifications or additions + to that Work or Derivative Works thereof, that is intentionally + submitted to Licensor for inclusion in the Work by the copyright owner + or by an individual or Legal Entity authorized to submit on behalf of + the copyright owner. For the purposes of this definition, "submitted" + means any form of electronic, verbal, or written communication sent + to the Licensor or its representatives, including but not limited to + communication on electronic mailing lists, source code control systems, + and issue tracking systems that are managed by, or on behalf of, the + Licensor for the purpose of discussing and improving the Work, but + excluding communication that is conspicuously marked or otherwise + designated in writing by the copyright owner as "Not a Contribution." + + "Contributor" shall mean Licensor and any individual or Legal Entity + on behalf of whom a Contribution has been received by Licensor and + subsequently incorporated within the Work. + + 2. Grant of Copyright License. Subject to the terms and conditions of + this License, each Contributor hereby grants to You a perpetual, + worldwide, non-exclusive, no-charge, royalty-free, irrevocable + copyright license to reproduce, prepare Derivative Works of, + publicly display, publicly perform, sublicense, and distribute the + Work and such Derivative Works in Source or Object form. + + 3. Grant of Patent License. Subject to the terms and conditions of + this License, each Contributor hereby grants to You a perpetual, + worldwide, non-exclusive, no-charge, royalty-free, irrevocable + (except as stated in this section) patent license to make, have made, + use, offer to sell, sell, import, and otherwise transfer the Work, + where such license applies only to those patent claims licensable + by such Contributor that are necessarily infringed by their + Contribution(s) alone or by combination of their Contribution(s) + with the Work to which such Contribution(s) was submitted. If You + institute patent litigation against any entity (including a + cross-claim or counterclaim in a lawsuit) alleging that the Work + or a Contribution incorporated within the Work constitutes direct + or contributory patent infringement, then any patent licenses + granted to You under this License for that Work shall terminate + as of the date such litigation is filed. + + 4. Redistribution. You may reproduce and distribute copies of the + Work or Derivative Works thereof in any medium, with or without + modifications, and in Source or Object form, provided that You + meet the following conditions: + + (a) You must give any other recipients of the Work or + Derivative Works a copy of this License; and + + (b) You must cause any modified files to carry prominent notices + stating that You changed the files; and + + (c) You must retain, in the Source form of any Derivative Works + that You distribute, all copyright, patent, trademark, and + attribution notices from the Source form of the Work, + excluding those notices that do not pertain to any part of + the Derivative Works; and + + (d) If the Work includes a "NOTICE" text file as part of its + distribution, then any Derivative Works that You distribute must + include a readable copy of the attribution notices contained + within such NOTICE file, excluding those notices that do not + pertain to any part of the Derivative Works, in at least one + of the following places: within a NOTICE text file distributed + as part of the Derivative Works; within the Source form or + documentation, if provided along with the Derivative Works; or, + within a display generated by the Derivative Works, if and + wherever such third-party notices normally appear. The contents + of the NOTICE file are for informational purposes only and + do not modify the License. You may add Your own attribution + notices within Derivative Works that You distribute, alongside + or as an addendum to the NOTICE text from the Work, provided + that such additional attribution notices cannot be construed + as modifying the License. + + You may add Your own copyright statement to Your modifications and + may provide additional or different license terms and conditions + for use, reproduction, or distribution of Your modifications, or + for any such Derivative Works as a whole, provided Your use, + reproduction, and distribution of the Work otherwise complies with + the conditions stated in this License. + + 5. Submission of Contributions. Unless You explicitly state otherwise, + any Contribution intentionally submitted for inclusion in the Work + by You to the Licensor shall be under the terms and conditions of + this License, without any additional terms or conditions. + Notwithstanding the above, nothing herein shall supersede or modify + the terms of any separate license agreement you may have executed + with Licensor regarding such Contributions. + + 6. Trademarks. This License does not grant permission to use the trade + names, trademarks, service marks, or product names of the Licensor, + except as required for reasonable and customary use in describing the + origin of the Work and reproducing the content of the NOTICE file. + + 7. Disclaimer of Warranty. Unless required by applicable law or + agreed to in writing, Licensor provides the Work (and each + Contributor provides its Contributions) on an "AS IS" BASIS, + WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or + implied, including, without limitation, any warranties or conditions + of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A + PARTICULAR PURPOSE. You are solely responsible for determining the + appropriateness of using or redistributing the Work and assume any + risks associated with Your exercise of permissions under this License. + + 8. Limitation of Liability. In no event and under no legal theory, + whether in tort (including negligence), contract, or otherwise, + unless required by applicable law (such as deliberate and grossly + negligent acts) or agreed to in writing, shall any Contributor be + liable to You for damages, including any direct, indirect, special, + incidental, or consequential damages of any character arising as a + result of this License or out of the use or inability to use the + Work (including but not limited to damages for loss of goodwill, + work stoppage, computer failure or malfunction, or any and all + other commercial damages or losses), even if such Contributor + has been advised of the possibility of such damages. + + 9. Accepting Warranty or Additional Liability. While redistributing + the Work or Derivative Works thereof, You may choose to offer, + and charge a fee for, acceptance of support, warranty, indemnity, + or other liability obligations and/or rights consistent with this + License. However, in accepting such obligations, You may act only + on Your own behalf and on Your sole responsibility, not on behalf + of any other Contributor, and only if You agree to indemnify, + defend, and hold each Contributor harmless for any liability + incurred by, or claims asserted against, such Contributor by reason + of your accepting any such warranty or additional liability. + + END OF TERMS AND CONDITIONS + + APPENDIX: How to apply the Apache License to your work. + + To apply the Apache License to your work, attach the following + boilerplate notice, with the fields enclosed by brackets "[]" + replaced with your own identifying information. (Don't include + the brackets!) The text should be enclosed in the appropriate + comment syntax for the file format. We also recommend that a + file or class name and description of purpose be included on the + same "printed page" as the copyright notice for easier + identification within third-party archives. + +Copyright 2022 YANDEX LLC + + Licensed under the Apache License, Version 2.0 (the "License"); + you may not use this file except in compliance with the License. + You may obtain a copy of the License at + + http://www.apache.org/licenses/LICENSE-2.0 + + Unless required by applicable law or agreed to in writing, software + distributed under the License is distributed on an "AS IS" BASIS, + WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + See the License for the specific language governing permissions and + limitations under the License. diff --git a/ydb/public/sdk/python3/README.md b/contrib/python/ydb/py2/README.md index cfc57eb276..cfc57eb276 100644 --- a/ydb/public/sdk/python3/README.md +++ b/contrib/python/ydb/py2/README.md diff --git a/ydb/public/sdk/python2/ya.make b/contrib/python/ydb/py2/ya.make index 8b983ba5c5..f88f935402 100644 --- a/ydb/public/sdk/python2/ya.make +++ b/contrib/python/ydb/py2/ya.make @@ -1,9 +1,32 @@ +# Generated by devtools/yamaker (pypi). + PY2_LIBRARY() +VERSION(2.15.1) + +LICENSE(Apache-2.0) + +PEERDIR( + contrib/deprecated/python/enum34 + contrib/python/grpcio + contrib/python/packaging + contrib/python/protobuf + contrib/python/pytest + contrib/python/six +) + +NO_LINT() + +NO_CHECK_IMPORTS( + ydb.public.api.grpc + ydb.public.api.grpc.* +) + PY_SRCS( TOP_LEVEL ydb/__init__.py ydb/_apis.py + ydb/_errors.py ydb/_session_impl.py ydb/_sp_impl.py ydb/_tx_ctx_impl.py @@ -19,6 +42,7 @@ PY_SRCS( ydb/default_pem.py ydb/driver.py ydb/export.py + ydb/global_settings.py ydb/iam/__init__.py ydb/iam/auth.py ydb/import_client.py @@ -38,15 +62,10 @@ PY_SRCS( ydb/ydb_version.py ) -PEERDIR( - contrib/deprecated/python/enum34 - contrib/deprecated/python/futures - contrib/python/grpcio - contrib/python/protobuf - ydb/public/api/grpc - ydb/public/api/grpc/draft - ydb/public/api/protos - +RESOURCE_FILES( + PREFIX contrib/python/ydb/py2/ + .dist-info/METADATA + .dist-info/top_level.txt ) END() diff --git a/ydb/public/sdk/python2/ydb/__init__.py b/contrib/python/ydb/py2/ydb/__init__.py index d8c23fee4c..56b73478eb 100644 --- a/ydb/public/sdk/python2/ydb/__init__.py +++ b/contrib/python/ydb/py2/ydb/__init__.py @@ -1,5 +1,6 @@ from .credentials import * # noqa from .driver import * # noqa +from .global_settings import * # noqa from .table import * # noqa from .issues import * # noqa from .types import * # noqa diff --git a/ydb/public/sdk/python2/ydb/_apis.py b/contrib/python/ydb/py2/ydb/_apis.py index aa496d6b29..89efac6041 100644 --- a/ydb/public/sdk/python2/ydb/_apis.py +++ b/contrib/python/ydb/py2/ydb/_apis.py @@ -1,20 +1,40 @@ # -*- coding: utf-8 -*- -from ydb.public.api.grpc import ( - ydb_cms_v1_pb2_grpc, - ydb_discovery_v1_pb2_grpc, - ydb_scheme_v1_pb2_grpc, - ydb_table_v1_pb2_grpc, -) -from ydb.public.api.protos import ( - ydb_status_codes_pb2, - ydb_discovery_pb2, - ydb_scheme_pb2, - ydb_table_pb2, - ydb_value_pb2, -) -from ydb.public.api.protos import ydb_operation_pb2 -from ydb.public.api.protos import ydb_common_pb2 -from ydb.public.api.grpc import ydb_operation_v1_pb2_grpc +try: + from ydb.public.api.grpc import ( + ydb_cms_v1_pb2_grpc, + ydb_discovery_v1_pb2_grpc, + ydb_scheme_v1_pb2_grpc, + ydb_table_v1_pb2_grpc, + ydb_operation_v1_pb2_grpc, + ) + + from ydb.public.api.protos import ( + ydb_status_codes_pb2, + ydb_discovery_pb2, + ydb_scheme_pb2, + ydb_table_pb2, + ydb_value_pb2, + ydb_operation_pb2, + ydb_common_pb2, + ) +except ImportError: + from contrib.ydb.public.api.grpc import ( + ydb_cms_v1_pb2_grpc, + ydb_discovery_v1_pb2_grpc, + ydb_scheme_v1_pb2_grpc, + ydb_table_v1_pb2_grpc, + ydb_operation_v1_pb2_grpc, + ) + + from contrib.ydb.public.api.protos import ( + ydb_status_codes_pb2, + ydb_discovery_pb2, + ydb_scheme_pb2, + ydb_table_pb2, + ydb_value_pb2, + ydb_operation_pb2, + ydb_common_pb2, + ) StatusIds = ydb_status_codes_pb2.StatusIds diff --git a/contrib/python/ydb/py2/ydb/_errors.py b/contrib/python/ydb/py2/ydb/_errors.py new file mode 100644 index 0000000000..84ac23db76 --- /dev/null +++ b/contrib/python/ydb/py2/ydb/_errors.py @@ -0,0 +1,60 @@ +from . import issues + +_errors_retriable_fast_backoff_types = [ + issues.Unavailable, +] +_errors_retriable_slow_backoff_types = [ + issues.Aborted, + issues.BadSession, + issues.Overloaded, + issues.SessionPoolEmpty, + issues.ConnectionError, +] +_errors_retriable_slow_backoff_idempotent_types = [ + issues.Undetermined, +] + + +def check_retriable_error(err, retry_settings, attempt): + if isinstance(err, issues.NotFound): + if retry_settings.retry_not_found: + return ErrorRetryInfo( + True, retry_settings.fast_backoff.calc_timeout(attempt) + ) + else: + return ErrorRetryInfo(False, None) + + if isinstance(err, issues.InternalError): + if retry_settings.retry_internal_error: + return ErrorRetryInfo( + True, retry_settings.slow_backoff.calc_timeout(attempt) + ) + else: + return ErrorRetryInfo(False, None) + + for t in _errors_retriable_fast_backoff_types: + if isinstance(err, t): + return ErrorRetryInfo( + True, retry_settings.fast_backoff.calc_timeout(attempt) + ) + + for t in _errors_retriable_slow_backoff_types: + if isinstance(err, t): + return ErrorRetryInfo( + True, retry_settings.slow_backoff.calc_timeout(attempt) + ) + + if retry_settings.idempotent: + for t in _errors_retriable_slow_backoff_idempotent_types: + if isinstance(err, t): + return ErrorRetryInfo( + True, retry_settings.slow_backoff.calc_timeout(attempt) + ) + + return ErrorRetryInfo(False, None) + + +class ErrorRetryInfo: + def __init__(self, is_retriable, sleep_timeout_seconds=None): + self.is_retriable = is_retriable + self.sleep_timeout_seconds = sleep_timeout_seconds diff --git a/ydb/public/sdk/python2/ydb/_session_impl.py b/contrib/python/ydb/py2/ydb/_session_impl.py index 27d7a3f6a0..27d7a3f6a0 100644 --- a/ydb/public/sdk/python2/ydb/_session_impl.py +++ b/contrib/python/ydb/py2/ydb/_session_impl.py diff --git a/ydb/public/sdk/python2/ydb/_sp_impl.py b/contrib/python/ydb/py2/ydb/_sp_impl.py index a8529d7321..a8529d7321 100644 --- a/ydb/public/sdk/python2/ydb/_sp_impl.py +++ b/contrib/python/ydb/py2/ydb/_sp_impl.py diff --git a/ydb/public/sdk/python2/ydb/_tx_ctx_impl.py b/contrib/python/ydb/py2/ydb/_tx_ctx_impl.py index 0f12865920..925d74b441 100644 --- a/ydb/public/sdk/python2/ydb/_tx_ctx_impl.py +++ b/contrib/python/ydb/py2/ydb/_tx_ctx_impl.py @@ -116,7 +116,9 @@ def _construct_tx_settings(tx_state): @wrap_tx_factory_handler -def execute_request_factory(session_state, tx_state, query, parameters, commit_tx, settings): +def execute_request_factory( + session_state, tx_state, query, parameters, commit_tx, settings +): data_query, query_id = session_state.lookup(query) parameters_types = {} @@ -134,7 +136,9 @@ def execute_request_factory(session_state, tx_state, query, parameters, commit_t else: yql_text = query query_pb = _apis.ydb_table.Query(yql_text=yql_text) - request = _apis.ydb_table.ExecuteDataQueryRequest(parameters=convert.parameters_to_pb(parameters_types, parameters)) + request = _apis.ydb_table.ExecuteDataQueryRequest( + parameters=convert.parameters_to_pb(parameters_types, parameters) + ) if query_id is not None: # SDK not send query text and nothing save to cache diff --git a/ydb/public/sdk/python2/ydb/_utilities.py b/contrib/python/ydb/py2/ydb/_utilities.py index 32419b1bf9..32419b1bf9 100644 --- a/ydb/public/sdk/python2/ydb/_utilities.py +++ b/contrib/python/ydb/py2/ydb/_utilities.py diff --git a/ydb/public/sdk/python2/ydb/auth_helpers.py b/contrib/python/ydb/py2/ydb/auth_helpers.py index 5d889555df..5d889555df 100644 --- a/ydb/public/sdk/python2/ydb/auth_helpers.py +++ b/contrib/python/ydb/py2/ydb/auth_helpers.py diff --git a/ydb/public/sdk/python2/ydb/connection.py b/contrib/python/ydb/py2/ydb/connection.py index 95db084a3c..95db084a3c 100644 --- a/ydb/public/sdk/python2/ydb/connection.py +++ b/contrib/python/ydb/py2/ydb/connection.py diff --git a/ydb/public/sdk/python2/ydb/convert.py b/contrib/python/ydb/py2/ydb/convert.py index 07a3658b31..97d51cf1b6 100644 --- a/ydb/public/sdk/python2/ydb/convert.py +++ b/contrib/python/ydb/py2/ydb/convert.py @@ -13,6 +13,7 @@ _DecimalNanRepr = 10**35 + 1 _DecimalInfRepr = 10**35 _DecimalSignedInfRepr = -(10**35) _primitive_type_by_id = {} +_default_allow_truncated_result = True def _initialize(): @@ -216,9 +217,10 @@ def _dict_to_pb(type_pb, value): for key, payload in value.items(): kv_pair = value_pb.pairs.add() kv_pair.key.MergeFrom(_from_native_value(type_pb.dict_type.key, key)) - kv_pair.payload.MergeFrom( - _from_native_value(type_pb.dict_type.payload, payload) - ) + if payload: + kv_pair.payload.MergeFrom( + _from_native_value(type_pb.dict_type.payload, payload) + ) return value_pb @@ -493,10 +495,20 @@ class ResultSets(list): if table_client_settings is None else table_client_settings._make_result_sets_lazy ) + + allow_truncated_result = _default_allow_truncated_result + if table_client_settings: + allow_truncated_result = table_client_settings._allow_truncated_result + result_sets = [] initializer = ( _ResultSet.from_message if not make_lazy else _ResultSet.lazy_from_message ) for result_set in result_sets_pb: - result_sets.append(initializer(result_set, table_client_settings)) + result_set = initializer(result_set, table_client_settings) + if result_set.truncated and not allow_truncated_result: + raise issues.TruncatedResponseError( + "Response for the request was truncated by server" + ) + result_sets.append(result_set) super(ResultSets, self).__init__(result_sets) diff --git a/ydb/public/sdk/python2/ydb/credentials.py b/contrib/python/ydb/py2/ydb/credentials.py index 96247ef32e..8547fbbd7b 100644 --- a/ydb/public/sdk/python2/ydb/credentials.py +++ b/contrib/python/ydb/py2/ydb/credentials.py @@ -7,8 +7,13 @@ import threading from concurrent import futures import logging import time -from ydb.public.api.protos import ydb_auth_pb2 -from ydb.public.api.grpc import ydb_auth_v1_pb2_grpc + +try: + from ydb.public.api.protos import ydb_auth_pb2 + from ydb.public.api.grpc import ydb_auth_v1_pb2_grpc +except ImportError: + from contrib.ydb.public.api.protos import ydb_auth_pb2 + from contrib.ydb.public.api.grpc import ydb_auth_v1_pb2_grpc YDB_AUTH_TICKET_HEADER = "x-ydb-auth-ticket" diff --git a/ydb/public/sdk/python2/ydb/dbapi/__init__.py b/contrib/python/ydb/py2/ydb/dbapi/__init__.py index 7363921192..7363921192 100644 --- a/ydb/public/sdk/python2/ydb/dbapi/__init__.py +++ b/contrib/python/ydb/py2/ydb/dbapi/__init__.py diff --git a/ydb/public/sdk/python2/ydb/dbapi/connection.py b/contrib/python/ydb/py2/ydb/dbapi/connection.py index c69381a92f..c69381a92f 100644 --- a/ydb/public/sdk/python2/ydb/dbapi/connection.py +++ b/contrib/python/ydb/py2/ydb/dbapi/connection.py diff --git a/ydb/public/sdk/python2/ydb/dbapi/cursor.py b/contrib/python/ydb/py2/ydb/dbapi/cursor.py index 71175abf4e..71175abf4e 100644 --- a/ydb/public/sdk/python2/ydb/dbapi/cursor.py +++ b/contrib/python/ydb/py2/ydb/dbapi/cursor.py diff --git a/ydb/public/sdk/python2/ydb/dbapi/errors.py b/contrib/python/ydb/py2/ydb/dbapi/errors.py index f4309f3812..f4309f3812 100644 --- a/ydb/public/sdk/python2/ydb/dbapi/errors.py +++ b/contrib/python/ydb/py2/ydb/dbapi/errors.py diff --git a/ydb/public/sdk/python2/ydb/default_pem.py b/contrib/python/ydb/py2/ydb/default_pem.py index 92286ba237..92286ba237 100644 --- a/ydb/public/sdk/python2/ydb/default_pem.py +++ b/contrib/python/ydb/py2/ydb/default_pem.py diff --git a/ydb/public/sdk/python2/ydb/driver.py b/contrib/python/ydb/py2/ydb/driver.py index 9b3fa99cfa..9b3fa99cfa 100644 --- a/ydb/public/sdk/python2/ydb/driver.py +++ b/contrib/python/ydb/py2/ydb/driver.py diff --git a/ydb/public/sdk/python2/ydb/export.py b/contrib/python/ydb/py2/ydb/export.py index 43d326e1eb..aac31315ca 100644 --- a/ydb/public/sdk/python2/ydb/export.py +++ b/contrib/python/ydb/py2/ydb/export.py @@ -3,8 +3,14 @@ import enum from . import _apis from . import settings_impl as s_impl -from ydb.public.api.protos import ydb_export_pb2 -from ydb.public.api.grpc import ydb_export_v1_pb2_grpc + +try: + from ydb.public.api.protos import ydb_export_pb2 + from ydb.public.api.grpc import ydb_export_v1_pb2_grpc +except ImportError: + from contrib.ydb.public.api.protos import ydb_export_pb2 + from contrib.ydb.public.api.grpc import ydb_export_v1_pb2_grpc + from . import operation _ExportToYt = "ExportToYt" diff --git a/contrib/python/ydb/py2/ydb/global_settings.py b/contrib/python/ydb/py2/ydb/global_settings.py new file mode 100644 index 0000000000..8a3ec9535e --- /dev/null +++ b/contrib/python/ydb/py2/ydb/global_settings.py @@ -0,0 +1,16 @@ +from . import convert +from . import table + + +def global_allow_truncated_result(enabled=True): + """ + call global_allow_truncated_result(False) for more safe execution and compatible with future changes + """ + convert._default_allow_truncated_result = enabled + + +def global_allow_split_transactions(enabled): + """ + call global_allow_truncated_result(False) for more safe execution and compatible with future changes + """ + table._allow_split_transaction = enabled diff --git a/ydb/public/sdk/python2/ydb/iam/__init__.py b/contrib/python/ydb/py2/ydb/iam/__init__.py index 7167efe13e..7167efe13e 100644 --- a/ydb/public/sdk/python2/ydb/iam/__init__.py +++ b/contrib/python/ydb/py2/ydb/iam/__init__.py diff --git a/ydb/public/sdk/python2/ydb/iam/auth.py b/contrib/python/ydb/py2/ydb/iam/auth.py index 06b07e917e..06b07e917e 100644 --- a/ydb/public/sdk/python2/ydb/iam/auth.py +++ b/contrib/python/ydb/py2/ydb/iam/auth.py diff --git a/ydb/public/sdk/python2/ydb/import_client.py b/contrib/python/ydb/py2/ydb/import_client.py index 160b0a6bed..a11d77a0c6 100644 --- a/ydb/public/sdk/python2/ydb/import_client.py +++ b/contrib/python/ydb/py2/ydb/import_client.py @@ -3,8 +3,15 @@ import enum from . import _apis from . import settings_impl as s_impl -from ydb.public.api.protos import ydb_import_pb2 -from ydb.public.api.grpc import ydb_import_v1_pb2_grpc + +try: + from ydb.public.api.protos import ydb_import_pb2 + from ydb.public.api.grpc import ydb_import_v1_pb2_grpc +except ImportError: + from contrib.ydb.public.api.protos import ydb_import_pb2 + from contrib.ydb.public.api.grpc import ydb_import_v1_pb2_grpc + + from . import operation _ImportFromS3 = "ImportFromS3" diff --git a/ydb/public/sdk/python2/ydb/interceptor.py b/contrib/python/ydb/py2/ydb/interceptor.py index 5405177a8e..5405177a8e 100644 --- a/ydb/public/sdk/python2/ydb/interceptor.py +++ b/contrib/python/ydb/py2/ydb/interceptor.py diff --git a/ydb/public/sdk/python2/ydb/issues.py b/contrib/python/ydb/py2/ydb/issues.py index fd301e32b6..611f15ecad 100644 --- a/ydb/public/sdk/python2/ydb/issues.py +++ b/contrib/python/ydb/py2/ydb/issues.py @@ -53,6 +53,10 @@ class Error(Exception): self.message = message +class TruncatedResponseError(Error): + status = None + + class ConnectionError(Error): status = None @@ -160,11 +164,18 @@ class SessionPoolEmpty(Error, queue.Empty): def _format_issues(issues): if not issues: return "" + return " ,".join( - [text_format.MessageToString(issue, False, True) for issue in issues] + text_format.MessageToString(issue, as_utf8=False, as_one_line=True) + for issue in issues ) +def _format_response(response): + fmt_issues = _format_issues(response.issues) + return "{0} (server_code: {1})".format(fmt_issues, response.status) + + _success_status_codes = {StatusCode.STATUS_CODE_UNSPECIFIED, StatusCode.SUCCESS} _server_side_error_map = { StatusCode.BAD_REQUEST: BadRequest, @@ -192,4 +203,4 @@ _server_side_error_map = { def _process_response(response_proto): if response_proto.status not in _success_status_codes: exc_obj = _server_side_error_map.get(response_proto.status) - raise exc_obj(_format_issues(response_proto.issues), response_proto.issues) + raise exc_obj(_format_response(response_proto), response_proto.issues) diff --git a/ydb/public/sdk/python2/ydb/operation.py b/contrib/python/ydb/py2/ydb/operation.py index 6084ef0f18..6084ef0f18 100644 --- a/ydb/public/sdk/python2/ydb/operation.py +++ b/contrib/python/ydb/py2/ydb/operation.py diff --git a/ydb/public/sdk/python2/ydb/pool.py b/contrib/python/ydb/py2/ydb/pool.py index dfda0adff2..dfda0adff2 100644 --- a/ydb/public/sdk/python2/ydb/pool.py +++ b/contrib/python/ydb/py2/ydb/pool.py diff --git a/ydb/public/sdk/python2/ydb/resolver.py b/contrib/python/ydb/py2/ydb/resolver.py index b40ae984dc..b40ae984dc 100644 --- a/ydb/public/sdk/python2/ydb/resolver.py +++ b/contrib/python/ydb/py2/ydb/resolver.py diff --git a/ydb/public/sdk/python2/ydb/scheme.py b/contrib/python/ydb/py2/ydb/scheme.py index a78357e030..4e8b716d5f 100644 --- a/ydb/public/sdk/python2/ydb/scheme.py +++ b/contrib/python/ydb/py2/ydb/scheme.py @@ -12,6 +12,7 @@ class SchemeEntryType(enum.IntEnum): Enumerates all available entry types. """ + TYPE_UNSPECIFIED = 0 DIRECTORY = 1 TABLE = 2 PERS_QUEUE_GROUP = 3 @@ -299,7 +300,7 @@ def _wrap_scheme_entry(entry_pb, scheme_entry_cls=None, *args, **kwargs): return scheme_entry_cls( entry_pb.name, entry_pb.owner, - getattr(SchemeEntryType, _apis.ydb_scheme.Entry.Type.Name(entry_pb.type)), + SchemeEntryType(entry_pb.type), _wrap_permissions(entry_pb.effective_permissions), _wrap_permissions(entry_pb.permissions), entry_pb.size_bytes, @@ -318,7 +319,7 @@ def _wrap_list_directory_response(rpc_state, response): message = _apis.ydb_scheme.ListDirectoryResult() response.operation.result.Unpack(message) children = [] - supported_items = set([i.value for i in SchemeEntryType]) + supported_items = set(i.value for i in SchemeEntryType) for children_item in message.children: if children_item.type not in supported_items: continue @@ -328,12 +329,10 @@ def _wrap_list_directory_response(rpc_state, response): return Directory( message.self.name, message.self.owner, - getattr(SchemeEntryType, _apis.ydb_scheme.Entry.Type.Name(message.self.type)), + SchemeEntryType(message.self.type), _wrap_permissions(message.self.effective_permissions), _wrap_permissions(message.self.permissions), - tuple( - children, - ), + tuple(children), ) diff --git a/ydb/public/sdk/python2/ydb/scripting.py b/contrib/python/ydb/py2/ydb/scripting.py index 587630f324..27c4be5386 100644 --- a/ydb/public/sdk/python2/ydb/scripting.py +++ b/contrib/python/ydb/py2/ydb/scripting.py @@ -1,5 +1,11 @@ -from ydb.public.api.protos import ydb_scripting_pb2 -from ydb.public.api.grpc import ydb_scripting_v1_pb2_grpc +try: + from ydb.public.api.protos import ydb_scripting_pb2 + from ydb.public.api.grpc import ydb_scripting_v1_pb2_grpc +except ImportError: + from contrib.ydb.public.api.protos import ydb_scripting_pb2 + from contrib.ydb.public.api.grpc import ydb_scripting_v1_pb2_grpc + + from . import issues, convert, settings diff --git a/ydb/public/sdk/python2/ydb/settings.py b/contrib/python/ydb/py2/ydb/settings.py index 6739a46fab..6739a46fab 100644 --- a/ydb/public/sdk/python2/ydb/settings.py +++ b/contrib/python/ydb/py2/ydb/settings.py diff --git a/ydb/public/sdk/python2/ydb/sqlalchemy/__init__.py b/contrib/python/ydb/py2/ydb/sqlalchemy/__init__.py index 9e065d8f0e..aa9b2d006c 100644 --- a/ydb/public/sdk/python2/ydb/sqlalchemy/__init__.py +++ b/contrib/python/ydb/py2/ydb/sqlalchemy/__init__.py @@ -191,11 +191,16 @@ try: ydb.PrimitiveType.DyNumber: sa.TEXT, } - def _get_column_type(t): - if isinstance(t.item, ydb.DecimalType): - return sa.DECIMAL(precision=t.item.precision, scale=t.item.scale) + def _get_column_info(t): + nullable = False + if isinstance(t, ydb.OptionalType): + nullable = True + t = t.item - return COLUMN_TYPES[t.item] + if isinstance(t, ydb.DecimalType): + return sa.DECIMAL(precision=t.precision, scale=t.scale), nullable + + return COLUMN_TYPES[t], nullable class YqlDialect(DefaultDialect): name = "yql" @@ -250,11 +255,12 @@ try: columns = raw_conn.describe(qt) as_compatible = [] for column in columns: + col_type, nullable = _get_column_info(column.type) as_compatible.append( { "name": column.name, - "type": _get_column_type(column.type), - "nullable": True, + "type": col_type, + "nullable": nullable, } ) diff --git a/ydb/public/sdk/python2/ydb/sqlalchemy/types.py b/contrib/python/ydb/py2/ydb/sqlalchemy/types.py index f6e10ccd6f..f6e10ccd6f 100644 --- a/ydb/public/sdk/python2/ydb/sqlalchemy/types.py +++ b/contrib/python/ydb/py2/ydb/sqlalchemy/types.py diff --git a/ydb/public/sdk/python2/ydb/table.py b/contrib/python/ydb/py2/ydb/table.py index 2241de3590..c06652610a 100644 --- a/ydb/public/sdk/python2/ydb/table.py +++ b/contrib/python/ydb/py2/ydb/table.py @@ -21,12 +21,15 @@ from . import ( _tx_ctx_impl, tracing, ) +from ._errors import check_retriable_error try: from . import interceptor except ImportError: interceptor = None +_allow_split_transaction = True + logger = logging.getLogger(__name__) ################################################################## @@ -780,6 +783,22 @@ class AbstractTransactionModeBuilder(object): pass +class SnapshotReadOnly(AbstractTransactionModeBuilder): + __slots__ = ("_pb", "_name") + + def __init__(self): + self._pb = _apis.ydb_table.SnapshotModeSettings() + self._name = "snapshot_read_only" + + @property + def settings(self): + return self._pb + + @property + def name(self): + return self._name + + class SerializableReadWrite(AbstractTransactionModeBuilder): __slots__ = ("_pb", "_name") @@ -900,12 +919,28 @@ class YdbRetryOperationSleepOpt(object): def __init__(self, timeout): self.timeout = timeout + def __eq__(self, other): + return type(self) == type(other) and self.timeout == other.timeout + + def __repr__(self): + return "YdbRetryOperationSleepOpt(%s)" % self.timeout + class YdbRetryOperationFinalResult(object): def __init__(self, result): self.result = result self.exc = None + def __eq__(self, other): + return ( + type(self) == type(other) + and self.result == other.result + and self.exc == other.exc + ) + + def __repr__(self): + return "YdbRetryOperationFinalResult(%s, exc=%s)" % (self.result, self.exc) + def set_exception(self, exc): self.exc = exc @@ -922,56 +957,28 @@ def retry_operation_impl(callee, retry_settings=None, *args, **kwargs): if result.exc is not None: raise result.exc - except ( - issues.Aborted, - issues.BadSession, - issues.NotFound, - issues.InternalError, - ) as e: - status = e - retry_settings.on_ydb_error_callback(e) - - if isinstance(e, issues.NotFound) and not retry_settings.retry_not_found: - raise e - - if ( - isinstance(e, issues.InternalError) - and not retry_settings.retry_internal_error - ): - raise e - - except ( - issues.Overloaded, - issues.SessionPoolEmpty, - issues.ConnectionError, - ) as e: - status = e - retry_settings.on_ydb_error_callback(e) - yield YdbRetryOperationSleepOpt( - retry_settings.slow_backoff.calc_timeout(attempt) - ) - - except issues.Unavailable as e: + except issues.Error as e: status = e retry_settings.on_ydb_error_callback(e) - yield YdbRetryOperationSleepOpt( - retry_settings.fast_backoff.calc_timeout(attempt) - ) - except issues.Undetermined as e: - status = e - retry_settings.on_ydb_error_callback(e) - if not retry_settings.idempotent: - # operation is not idempotent, so we cannot retry. + retriable_info = check_retriable_error(e, retry_settings, attempt) + if not retriable_info.is_retriable: raise - yield YdbRetryOperationSleepOpt( - retry_settings.fast_backoff.calc_timeout(attempt) - ) + skip_yield_error_types = [ + issues.Aborted, + issues.BadSession, + issues.NotFound, + issues.InternalError, + ] - except issues.Error as e: - retry_settings.on_ydb_error_callback(e) - raise + yield_sleep = True + for t in skip_yield_error_types: + if isinstance(e, t): + yield_sleep = False + + if yield_sleep: + yield YdbRetryOperationSleepOpt(retriable_info.sleep_timeout_seconds) except Exception as e: # you should provide your own handler you want @@ -999,6 +1006,7 @@ class TableClientSettings(object): self._native_json_in_result_sets = False self._native_interval_in_result_sets = False self._native_timestamp_in_result_sets = False + self._allow_truncated_result = convert._default_allow_truncated_result def with_native_timestamp_in_result_sets(self, enabled): # type:(bool) -> ydb.TableClientSettings @@ -1035,6 +1043,11 @@ class TableClientSettings(object): self._make_result_sets_lazy = enabled return self + def with_allow_truncated_result(self, enabled): + # type:(bool) -> ydb.TableClientSettings + self._allow_truncated_result = enabled + return self + class ScanQueryResult(object): def __init__(self, result, table_client_settings): @@ -1171,7 +1184,9 @@ class ISession: pass @abstractmethod - def transaction(self, tx_mode=None): + def transaction( + self, tx_mode=None, allow_split_transactions=_allow_split_transaction + ): pass @abstractmethod @@ -1676,8 +1691,16 @@ class BaseSession(ISession): self._state.endpoint, ) - def transaction(self, tx_mode=None): - return TxContext(self._driver, self._state, self, tx_mode) + def transaction( + self, tx_mode=None, allow_split_transactions=_allow_split_transaction + ): + return TxContext( + self._driver, + self._state, + self, + tx_mode, + allow_split_transactions=allow_split_transactions, + ) def has_prepared(self, query): return query in self._state @@ -2189,9 +2212,26 @@ class ITxContext: class BaseTxContext(ITxContext): - __slots__ = ("_tx_state", "_session_state", "_driver", "session") + __slots__ = ( + "_tx_state", + "_session_state", + "_driver", + "session", + "_finished", + "_allow_split_transactions", + ) - def __init__(self, driver, session_state, session, tx_mode=None): + _COMMIT = "commit" + _ROLLBACK = "rollback" + + def __init__( + self, + driver, + session_state, + session, + tx_mode=None, + allow_split_transactions=_allow_split_transaction, + ): """ An object that provides a simple transaction context manager that allows statements execution in a transaction. You don't have to open transaction explicitly, because context manager encapsulates @@ -2214,6 +2254,8 @@ class BaseTxContext(ITxContext): self._tx_state = _tx_ctx_impl.TxState(tx_mode) self._session_state = session_state self.session = session + self._finished = "" + self._allow_split_transactions = allow_split_transactions def __enter__(self): """ @@ -2271,6 +2313,11 @@ class BaseTxContext(ITxContext): :return: A result sets or exception in case of execution errors """ + + self._check_split() + if commit_tx: + self._set_finish(self._COMMIT) + return self._driver( _tx_ctx_impl.execute_request_factory( self._session_state, @@ -2297,8 +2344,12 @@ class BaseTxContext(ITxContext): :return: A committed transaction or exception if commit is failed """ + + self._set_finish(self._COMMIT) + if self._tx_state.tx_id is None and not self._tx_state.dead: return self + return self._driver( _tx_ctx_impl.commit_request_factory(self._session_state, self._tx_state), _apis.TableService.Stub, @@ -2318,8 +2369,12 @@ class BaseTxContext(ITxContext): :return: A rolled back transaction or exception if rollback is failed """ + + self._set_finish(self._ROLLBACK) + if self._tx_state.tx_id is None and not self._tx_state.dead: return self + return self._driver( _tx_ctx_impl.rollback_request_factory(self._session_state, self._tx_state), _apis.TableService.Stub, @@ -2340,6 +2395,9 @@ class BaseTxContext(ITxContext): """ if self._tx_state.tx_id is not None: return self + + self._check_split() + return self._driver( _tx_ctx_impl.begin_request_factory(self._session_state, self._tx_state), _apis.TableService.Stub, @@ -2350,6 +2408,21 @@ class BaseTxContext(ITxContext): self._session_state.endpoint, ) + def _set_finish(self, val): + self._check_split(val) + self._finished = val + + def _check_split(self, allow=""): + """ + Deny all operaions with transaction after commit/rollback. + Exception: double commit and double rollbacks, because it is safe + """ + if self._allow_split_transactions: + return + + if self._finished != "" and self._finished != allow: + raise RuntimeError("Any operation with finished transaction is denied") + class TxContext(BaseTxContext): @_utilities.wrap_async_call_exceptions @@ -2365,6 +2438,9 @@ class TxContext(BaseTxContext): :return: A future of query execution """ + + self._check_split() + return self._driver.future( _tx_ctx_impl.execute_request_factory( self._session_state, @@ -2396,8 +2472,11 @@ class TxContext(BaseTxContext): :return: A future of commit call """ + self._set_finish(self._COMMIT) + if self._tx_state.tx_id is None and not self._tx_state.dead: return _utilities.wrap_result_in_future(self) + return self._driver.future( _tx_ctx_impl.commit_request_factory(self._session_state, self._tx_state), _apis.TableService.Stub, @@ -2418,8 +2497,11 @@ class TxContext(BaseTxContext): :return: A future of rollback call """ + self._set_finish(self._ROLLBACK) + if self._tx_state.tx_id is None and not self._tx_state.dead: return _utilities.wrap_result_in_future(self) + return self._driver.future( _tx_ctx_impl.rollback_request_factory(self._session_state, self._tx_state), _apis.TableService.Stub, @@ -2441,6 +2523,9 @@ class TxContext(BaseTxContext): """ if self._tx_state.tx_id is not None: return _utilities.wrap_result_in_future(self) + + self._check_split() + return self._driver.future( _tx_ctx_impl.begin_request_factory(self._session_state, self._tx_state), _apis.TableService.Stub, diff --git a/ydb/public/sdk/python2/ydb/tracing.py b/contrib/python/ydb/py2/ydb/tracing.py index 798bab02a3..798bab02a3 100644 --- a/ydb/public/sdk/python2/ydb/tracing.py +++ b/contrib/python/ydb/py2/ydb/tracing.py diff --git a/ydb/public/sdk/python2/ydb/types.py b/contrib/python/ydb/py2/ydb/types.py index 598a9013aa..a62c8a74a0 100644 --- a/ydb/public/sdk/python2/ydb/types.py +++ b/contrib/python/ydb/py2/ydb/types.py @@ -12,6 +12,7 @@ from google.protobuf import struct_pb2 _SECONDS_IN_DAY = 60 * 60 * 24 _EPOCH = datetime(1970, 1, 1) + if six.PY3: _from_bytes = None else: @@ -20,13 +21,20 @@ else: return _utilities.from_bytes(x) -def _from_date_number(x, table_client_settings): +def _from_date(x, table_client_settings): if ( table_client_settings is not None and table_client_settings._native_date_in_result_sets ): - return date.fromordinal(x + date(1970, 1, 1).toordinal()) - return x + return _EPOCH.date() + timedelta(days=x.uint32_value) + return x.uint32_value + + +def _to_date(pb, value): + if isinstance(value, date): + pb.uint32_value = (value - _EPOCH.date()).days + else: + pb.uint32_value = value def _from_datetime_number(x, table_client_settings): @@ -122,8 +130,9 @@ class PrimitiveType(enum.Enum): UUID = (_apis.primitive_types.UUID, None, _to_uuid, _from_uuid) Date = ( _apis.primitive_types.DATE, - "uint32_value", - _from_date_number, + None, + _from_date, + _to_date, ) Datetime = ( _apis.primitive_types.DATETIME, diff --git a/contrib/python/ydb/py2/ydb/ydb_version.py b/contrib/python/ydb/py2/ydb/ydb_version.py new file mode 100644 index 0000000000..c85cca1e48 --- /dev/null +++ b/contrib/python/ydb/py2/ydb/ydb_version.py @@ -0,0 +1 @@ +VERSION = "2.15.1" diff --git a/contrib/python/ydb/py3/.dist-info/METADATA b/contrib/python/ydb/py3/.dist-info/METADATA new file mode 100644 index 0000000000..058eb80b41 --- /dev/null +++ b/contrib/python/ydb/py3/.dist-info/METADATA @@ -0,0 +1,60 @@ +Metadata-Version: 2.1 +Name: ydb +Version: 3.7.0 +Summary: YDB Python SDK +Home-page: http://github.com/ydb-platform/ydb-python-sdk +Author: Yandex LLC +Author-email: ydb@yandex-team.ru +License: Apache 2.0 +Classifier: Programming Language :: Python +Classifier: Programming Language :: Python :: 2 +Classifier: Programming Language :: Python :: 2.7 +Classifier: Programming Language :: Python :: 3 +Classifier: Programming Language :: Python :: 3.6 +Description-Content-Type: text/markdown +License-File: LICENSE +License-File: AUTHORS +Requires-Dist: grpcio >=1.42.0 +Requires-Dist: packaging +Requires-Dist: protobuf <5.0.0,>=3.13.0 +Requires-Dist: aiohttp <4 +Provides-Extra: yc +Requires-Dist: yandexcloud ; extra == 'yc' + +YDB Python SDK +--- +[](https://github.com/ydb-platform/ydb/blob/main/LICENSE) +[](https://badge.fury.io/py/ydb) +[](https://github.com/ydb-platform/ydb-python-sdk/actions/workflows/tests.yaml) +[](https://github.com/ydb-platform/ydb-python-sdk/actions/workflows/style.yaml) + +Officially supported Python client for YDB. + +## Quickstart + +### Prerequisites + +- Python 3.8 or higher +- `pip` version 9.0.1 or higher + +If necessary, upgrade your version of `pip`: + +```sh +$ python -m pip install --upgrade pip +``` + +If you cannot upgrade `pip` due to a system-owned installation, you can +run the example in a virtualenv: + +```sh +$ python -m pip install virtualenv +$ virtualenv venv +$ source venv/bin/activate +$ python -m pip install --upgrade pip +``` + +Install YDB python sdk: + +```sh +$ python -m pip install ydb +``` diff --git a/contrib/python/ydb/py3/.dist-info/top_level.txt b/contrib/python/ydb/py3/.dist-info/top_level.txt new file mode 100644 index 0000000000..2578e07d93 --- /dev/null +++ b/contrib/python/ydb/py3/.dist-info/top_level.txt @@ -0,0 +1,2 @@ +tests +ydb diff --git a/contrib/python/ydb/py3/AUTHORS b/contrib/python/ydb/py3/AUTHORS new file mode 100644 index 0000000000..69fee17e94 --- /dev/null +++ b/contrib/python/ydb/py3/AUTHORS @@ -0,0 +1,5 @@ +The following authors have created the source code of "Yandex Database Python SDK" +published and distributed by YANDEX LLC as the owner: + +Vitalii Gridnev <gridnevvvit@gmail.com> +Timofey Koolin <timofey.koolin@gmail.com> diff --git a/contrib/python/ydb/py3/LICENSE b/contrib/python/ydb/py3/LICENSE new file mode 100644 index 0000000000..cabac2dec9 --- /dev/null +++ b/contrib/python/ydb/py3/LICENSE @@ -0,0 +1,202 @@ +Copyright 2022 YANDEX LLC + Apache License + Version 2.0, January 2004 + http://www.apache.org/licenses/ + + TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION + + 1. Definitions. + + "License" shall mean the terms and conditions for use, reproduction, + and distribution as defined by Sections 1 through 9 of this document. + + "Licensor" shall mean the copyright owner or entity authorized by + the copyright owner that is granting the License. + + "Legal Entity" shall mean the union of the acting entity and all + other entities that control, are controlled by, or are under common + control with that entity. For the purposes of this definition, + "control" means (i) the power, direct or indirect, to cause the + direction or management of such entity, whether by contract or + otherwise, or (ii) ownership of fifty percent (50%) or more of the + outstanding shares, or (iii) beneficial ownership of such entity. + + "You" (or "Your") shall mean an individual or Legal Entity + exercising permissions granted by this License. + + "Source" form shall mean the preferred form for making modifications, + including but not limited to software source code, documentation + source, and configuration files. + + "Object" form shall mean any form resulting from mechanical + transformation or translation of a Source form, including but + not limited to compiled object code, generated documentation, + and conversions to other media types. + + "Work" shall mean the work of authorship, whether in Source or + Object form, made available under the License, as indicated by a + copyright notice that is included in or attached to the work + (an example is provided in the Appendix below). + + "Derivative Works" shall mean any work, whether in Source or Object + form, that is based on (or derived from) the Work and for which the + editorial revisions, annotations, elaborations, or other modifications + represent, as a whole, an original work of authorship. For the purposes + of this License, Derivative Works shall not include works that remain + separable from, or merely link (or bind by name) to the interfaces of, + the Work and Derivative Works thereof. + + "Contribution" shall mean any work of authorship, including + the original version of the Work and any modifications or additions + to that Work or Derivative Works thereof, that is intentionally + submitted to Licensor for inclusion in the Work by the copyright owner + or by an individual or Legal Entity authorized to submit on behalf of + the copyright owner. For the purposes of this definition, "submitted" + means any form of electronic, verbal, or written communication sent + to the Licensor or its representatives, including but not limited to + communication on electronic mailing lists, source code control systems, + and issue tracking systems that are managed by, or on behalf of, the + Licensor for the purpose of discussing and improving the Work, but + excluding communication that is conspicuously marked or otherwise + designated in writing by the copyright owner as "Not a Contribution." + + "Contributor" shall mean Licensor and any individual or Legal Entity + on behalf of whom a Contribution has been received by Licensor and + subsequently incorporated within the Work. + + 2. Grant of Copyright License. Subject to the terms and conditions of + this License, each Contributor hereby grants to You a perpetual, + worldwide, non-exclusive, no-charge, royalty-free, irrevocable + copyright license to reproduce, prepare Derivative Works of, + publicly display, publicly perform, sublicense, and distribute the + Work and such Derivative Works in Source or Object form. + + 3. Grant of Patent License. Subject to the terms and conditions of + this License, each Contributor hereby grants to You a perpetual, + worldwide, non-exclusive, no-charge, royalty-free, irrevocable + (except as stated in this section) patent license to make, have made, + use, offer to sell, sell, import, and otherwise transfer the Work, + where such license applies only to those patent claims licensable + by such Contributor that are necessarily infringed by their + Contribution(s) alone or by combination of their Contribution(s) + with the Work to which such Contribution(s) was submitted. If You + institute patent litigation against any entity (including a + cross-claim or counterclaim in a lawsuit) alleging that the Work + or a Contribution incorporated within the Work constitutes direct + or contributory patent infringement, then any patent licenses + granted to You under this License for that Work shall terminate + as of the date such litigation is filed. + + 4. Redistribution. You may reproduce and distribute copies of the + Work or Derivative Works thereof in any medium, with or without + modifications, and in Source or Object form, provided that You + meet the following conditions: + + (a) You must give any other recipients of the Work or + Derivative Works a copy of this License; and + + (b) You must cause any modified files to carry prominent notices + stating that You changed the files; and + + (c) You must retain, in the Source form of any Derivative Works + that You distribute, all copyright, patent, trademark, and + attribution notices from the Source form of the Work, + excluding those notices that do not pertain to any part of + the Derivative Works; and + + (d) If the Work includes a "NOTICE" text file as part of its + distribution, then any Derivative Works that You distribute must + include a readable copy of the attribution notices contained + within such NOTICE file, excluding those notices that do not + pertain to any part of the Derivative Works, in at least one + of the following places: within a NOTICE text file distributed + as part of the Derivative Works; within the Source form or + documentation, if provided along with the Derivative Works; or, + within a display generated by the Derivative Works, if and + wherever such third-party notices normally appear. The contents + of the NOTICE file are for informational purposes only and + do not modify the License. You may add Your own attribution + notices within Derivative Works that You distribute, alongside + or as an addendum to the NOTICE text from the Work, provided + that such additional attribution notices cannot be construed + as modifying the License. + + You may add Your own copyright statement to Your modifications and + may provide additional or different license terms and conditions + for use, reproduction, or distribution of Your modifications, or + for any such Derivative Works as a whole, provided Your use, + reproduction, and distribution of the Work otherwise complies with + the conditions stated in this License. + + 5. Submission of Contributions. Unless You explicitly state otherwise, + any Contribution intentionally submitted for inclusion in the Work + by You to the Licensor shall be under the terms and conditions of + this License, without any additional terms or conditions. + Notwithstanding the above, nothing herein shall supersede or modify + the terms of any separate license agreement you may have executed + with Licensor regarding such Contributions. + + 6. Trademarks. This License does not grant permission to use the trade + names, trademarks, service marks, or product names of the Licensor, + except as required for reasonable and customary use in describing the + origin of the Work and reproducing the content of the NOTICE file. + + 7. Disclaimer of Warranty. Unless required by applicable law or + agreed to in writing, Licensor provides the Work (and each + Contributor provides its Contributions) on an "AS IS" BASIS, + WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or + implied, including, without limitation, any warranties or conditions + of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A + PARTICULAR PURPOSE. You are solely responsible for determining the + appropriateness of using or redistributing the Work and assume any + risks associated with Your exercise of permissions under this License. + + 8. Limitation of Liability. In no event and under no legal theory, + whether in tort (including negligence), contract, or otherwise, + unless required by applicable law (such as deliberate and grossly + negligent acts) or agreed to in writing, shall any Contributor be + liable to You for damages, including any direct, indirect, special, + incidental, or consequential damages of any character arising as a + result of this License or out of the use or inability to use the + Work (including but not limited to damages for loss of goodwill, + work stoppage, computer failure or malfunction, or any and all + other commercial damages or losses), even if such Contributor + has been advised of the possibility of such damages. + + 9. Accepting Warranty or Additional Liability. While redistributing + the Work or Derivative Works thereof, You may choose to offer, + and charge a fee for, acceptance of support, warranty, indemnity, + or other liability obligations and/or rights consistent with this + License. However, in accepting such obligations, You may act only + on Your own behalf and on Your sole responsibility, not on behalf + of any other Contributor, and only if You agree to indemnify, + defend, and hold each Contributor harmless for any liability + incurred by, or claims asserted against, such Contributor by reason + of your accepting any such warranty or additional liability. + + END OF TERMS AND CONDITIONS + + APPENDIX: How to apply the Apache License to your work. + + To apply the Apache License to your work, attach the following + boilerplate notice, with the fields enclosed by brackets "[]" + replaced with your own identifying information. (Don't include + the brackets!) The text should be enclosed in the appropriate + comment syntax for the file format. We also recommend that a + file or class name and description of purpose be included on the + same "printed page" as the copyright notice for easier + identification within third-party archives. + +Copyright 2022 YANDEX LLC + + Licensed under the Apache License, Version 2.0 (the "License"); + you may not use this file except in compliance with the License. + You may obtain a copy of the License at + + http://www.apache.org/licenses/LICENSE-2.0 + + Unless required by applicable law or agreed to in writing, software + distributed under the License is distributed on an "AS IS" BASIS, + WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + See the License for the specific language governing permissions and + limitations under the License. diff --git a/contrib/python/ydb/py3/README.md b/contrib/python/ydb/py3/README.md new file mode 100644 index 0000000000..cfc57eb276 --- /dev/null +++ b/contrib/python/ydb/py3/README.md @@ -0,0 +1,37 @@ +YDB Python SDK +--- +[](https://github.com/ydb-platform/ydb/blob/main/LICENSE) +[](https://badge.fury.io/py/ydb) +[](https://github.com/ydb-platform/ydb-python-sdk/actions/workflows/tests.yaml) +[](https://github.com/ydb-platform/ydb-python-sdk/actions/workflows/style.yaml) + +Officially supported Python client for YDB. + +## Quickstart + +### Prerequisites + +- Python 3.8 or higher +- `pip` version 9.0.1 or higher + +If necessary, upgrade your version of `pip`: + +```sh +$ python -m pip install --upgrade pip +``` + +If you cannot upgrade `pip` due to a system-owned installation, you can +run the example in a virtualenv: + +```sh +$ python -m pip install virtualenv +$ virtualenv venv +$ source venv/bin/activate +$ python -m pip install --upgrade pip +``` + +Install YDB python sdk: + +```sh +$ python -m pip install ydb +``` diff --git a/ydb/public/sdk/python3/ya.make b/contrib/python/ydb/py3/ya.make index 3e4fbeb109..f92572fa82 100644 --- a/ydb/public/sdk/python3/ya.make +++ b/contrib/python/ydb/py3/ya.make @@ -1,5 +1,25 @@ +# Generated by devtools/yamaker (pypi). + PY3_LIBRARY() +VERSION(3.7.0) + +LICENSE(Apache-2.0) + +PEERDIR( + contrib/python/aiohttp + contrib/python/grpcio + contrib/python/packaging + contrib/python/protobuf +) + +NO_LINT() + +NO_CHECK_IMPORTS( + ydb.public.api.grpc + ydb.public.api.grpc.* +) + PY_SRCS( TOP_LEVEL ydb/__init__.py @@ -16,7 +36,6 @@ PY_SRCS( ydb/_sp_impl.py ydb/_topic_common/__init__.py ydb/_topic_common/common.py - ydb/_topic_common/test_helpers.py ydb/_topic_reader/__init__.py ydb/_topic_reader/datatypes.py ydb/_topic_reader/topic_reader.py @@ -70,12 +89,10 @@ PY_SRCS( ydb/ydb_version.py ) -PEERDIR( - contrib/python/grpcio - contrib/python/protobuf - ydb/public/api/grpc - ydb/public/api/grpc/draft - ydb/public/api/protos +RESOURCE_FILES( + PREFIX contrib/python/ydb/py3/ + .dist-info/METADATA + .dist-info/top_level.txt ) END() diff --git a/ydb/public/sdk/python3/ydb/__init__.py b/contrib/python/ydb/py3/ydb/__init__.py index 902b1e0850..902b1e0850 100644 --- a/ydb/public/sdk/python3/ydb/__init__.py +++ b/contrib/python/ydb/py3/ydb/__init__.py diff --git a/ydb/public/sdk/python3/ydb/_apis.py b/contrib/python/ydb/py3/ydb/_apis.py index 27bc1bbec8..27bc1bbec8 100644 --- a/ydb/public/sdk/python3/ydb/_apis.py +++ b/contrib/python/ydb/py3/ydb/_apis.py diff --git a/ydb/public/sdk/python3/ydb/_errors.py b/contrib/python/ydb/py3/ydb/_errors.py index 17002d2574..17002d2574 100644 --- a/ydb/public/sdk/python3/ydb/_errors.py +++ b/contrib/python/ydb/py3/ydb/_errors.py diff --git a/ydb/public/sdk/python3/ydb/_topic_writer/__init__.py b/contrib/python/ydb/py3/ydb/_grpc/__init__.py index e69de29bb2..e69de29bb2 100644 --- a/ydb/public/sdk/python3/ydb/_topic_writer/__init__.py +++ b/contrib/python/ydb/py3/ydb/_grpc/__init__.py diff --git a/contrib/python/ydb/py3/ydb/_grpc/common/__init__.py b/contrib/python/ydb/py3/ydb/_grpc/common/__init__.py new file mode 100644 index 0000000000..9dc3de8379 --- /dev/null +++ b/contrib/python/ydb/py3/ydb/_grpc/common/__init__.py @@ -0,0 +1,13 @@ +import sys +try: + from ydb.public.api.grpc import * # noqa + sys.modules["ydb._grpc.common"] = sys.modules["ydb.public.api.grpc"] + + from ydb.public.api import protos + sys.modules["ydb._grpc.common.protos"] = sys.modules["ydb.public.api.protos"] +except ImportError: + from contrib.ydb.public.api.grpc import * # noqa + sys.modules["ydb._grpc.common"] = sys.modules["contrib.ydb.public.api.grpc"] + + from contrib.ydb.public.api import protos + sys.modules["ydb._grpc.common.protos"] = sys.modules["contrib.ydb.public.api.protos"] diff --git a/contrib/python/ydb/py3/ydb/_grpc/grpcwrapper/__init__.py b/contrib/python/ydb/py3/ydb/_grpc/grpcwrapper/__init__.py new file mode 100644 index 0000000000..e69de29bb2 --- /dev/null +++ b/contrib/python/ydb/py3/ydb/_grpc/grpcwrapper/__init__.py diff --git a/ydb/public/sdk/python3/ydb/_grpc/grpcwrapper/common_utils.py b/contrib/python/ydb/py3/ydb/_grpc/grpcwrapper/common_utils.py index bc294025f3..bc294025f3 100644 --- a/ydb/public/sdk/python3/ydb/_grpc/grpcwrapper/common_utils.py +++ b/contrib/python/ydb/py3/ydb/_grpc/grpcwrapper/common_utils.py diff --git a/ydb/public/sdk/python3/ydb/_grpc/grpcwrapper/ydb_scheme.py b/contrib/python/ydb/py3/ydb/_grpc/grpcwrapper/ydb_scheme.py index b992203570..b992203570 100644 --- a/ydb/public/sdk/python3/ydb/_grpc/grpcwrapper/ydb_scheme.py +++ b/contrib/python/ydb/py3/ydb/_grpc/grpcwrapper/ydb_scheme.py diff --git a/ydb/public/sdk/python3/ydb/_grpc/grpcwrapper/ydb_topic.py b/contrib/python/ydb/py3/ydb/_grpc/grpcwrapper/ydb_topic.py index 5b5e294a21..5b5e294a21 100644 --- a/ydb/public/sdk/python3/ydb/_grpc/grpcwrapper/ydb_topic.py +++ b/contrib/python/ydb/py3/ydb/_grpc/grpcwrapper/ydb_topic.py diff --git a/ydb/public/sdk/python3/ydb/_grpc/grpcwrapper/ydb_topic_public_types.py b/contrib/python/ydb/py3/ydb/_grpc/grpcwrapper/ydb_topic_public_types.py index df280a8bb5..df280a8bb5 100644 --- a/ydb/public/sdk/python3/ydb/_grpc/grpcwrapper/ydb_topic_public_types.py +++ b/contrib/python/ydb/py3/ydb/_grpc/grpcwrapper/ydb_topic_public_types.py diff --git a/ydb/public/sdk/python3/ydb/_session_impl.py b/contrib/python/ydb/py3/ydb/_session_impl.py index a61612bcd7..a61612bcd7 100644 --- a/ydb/public/sdk/python3/ydb/_session_impl.py +++ b/contrib/python/ydb/py3/ydb/_session_impl.py diff --git a/ydb/public/sdk/python3/ydb/_sp_impl.py b/contrib/python/ydb/py3/ydb/_sp_impl.py index dfea89f9c4..dfea89f9c4 100644 --- a/ydb/public/sdk/python3/ydb/_sp_impl.py +++ b/contrib/python/ydb/py3/ydb/_sp_impl.py diff --git a/contrib/python/ydb/py3/ydb/_topic_common/__init__.py b/contrib/python/ydb/py3/ydb/_topic_common/__init__.py new file mode 100644 index 0000000000..e69de29bb2 --- /dev/null +++ b/contrib/python/ydb/py3/ydb/_topic_common/__init__.py diff --git a/ydb/public/sdk/python3/ydb/_topic_common/common.py b/contrib/python/ydb/py3/ydb/_topic_common/common.py index 7a97336edf..7a97336edf 100644 --- a/ydb/public/sdk/python3/ydb/_topic_common/common.py +++ b/contrib/python/ydb/py3/ydb/_topic_common/common.py diff --git a/contrib/python/ydb/py3/ydb/_topic_reader/__init__.py b/contrib/python/ydb/py3/ydb/_topic_reader/__init__.py new file mode 100644 index 0000000000..e69de29bb2 --- /dev/null +++ b/contrib/python/ydb/py3/ydb/_topic_reader/__init__.py diff --git a/ydb/public/sdk/python3/ydb/_topic_reader/datatypes.py b/contrib/python/ydb/py3/ydb/_topic_reader/datatypes.py index 28155ea7a3..28155ea7a3 100644 --- a/ydb/public/sdk/python3/ydb/_topic_reader/datatypes.py +++ b/contrib/python/ydb/py3/ydb/_topic_reader/datatypes.py diff --git a/ydb/public/sdk/python3/ydb/_topic_reader/topic_reader.py b/contrib/python/ydb/py3/ydb/_topic_reader/topic_reader.py index 17fb288555..17fb288555 100644 --- a/ydb/public/sdk/python3/ydb/_topic_reader/topic_reader.py +++ b/contrib/python/ydb/py3/ydb/_topic_reader/topic_reader.py diff --git a/ydb/public/sdk/python3/ydb/_topic_reader/topic_reader_asyncio.py b/contrib/python/ydb/py3/ydb/_topic_reader/topic_reader_asyncio.py index 50684f7cf9..50684f7cf9 100644 --- a/ydb/public/sdk/python3/ydb/_topic_reader/topic_reader_asyncio.py +++ b/contrib/python/ydb/py3/ydb/_topic_reader/topic_reader_asyncio.py diff --git a/ydb/public/sdk/python3/ydb/_topic_reader/topic_reader_sync.py b/contrib/python/ydb/py3/ydb/_topic_reader/topic_reader_sync.py index e5b4e1a2b4..e5b4e1a2b4 100644 --- a/ydb/public/sdk/python3/ydb/_topic_reader/topic_reader_sync.py +++ b/contrib/python/ydb/py3/ydb/_topic_reader/topic_reader_sync.py diff --git a/contrib/python/ydb/py3/ydb/_topic_writer/__init__.py b/contrib/python/ydb/py3/ydb/_topic_writer/__init__.py new file mode 100644 index 0000000000..e69de29bb2 --- /dev/null +++ b/contrib/python/ydb/py3/ydb/_topic_writer/__init__.py diff --git a/ydb/public/sdk/python3/ydb/_topic_writer/topic_writer.py b/contrib/python/ydb/py3/ydb/_topic_writer/topic_writer.py index 527bf03eac..527bf03eac 100644 --- a/ydb/public/sdk/python3/ydb/_topic_writer/topic_writer.py +++ b/contrib/python/ydb/py3/ydb/_topic_writer/topic_writer.py diff --git a/ydb/public/sdk/python3/ydb/_topic_writer/topic_writer_asyncio.py b/contrib/python/ydb/py3/ydb/_topic_writer/topic_writer_asyncio.py index dd969c7e8e..dd969c7e8e 100644 --- a/ydb/public/sdk/python3/ydb/_topic_writer/topic_writer_asyncio.py +++ b/contrib/python/ydb/py3/ydb/_topic_writer/topic_writer_asyncio.py diff --git a/ydb/public/sdk/python3/ydb/_topic_writer/topic_writer_sync.py b/contrib/python/ydb/py3/ydb/_topic_writer/topic_writer_sync.py index a5193caf7c..a5193caf7c 100644 --- a/ydb/public/sdk/python3/ydb/_topic_writer/topic_writer_sync.py +++ b/contrib/python/ydb/py3/ydb/_topic_writer/topic_writer_sync.py diff --git a/ydb/public/sdk/python3/ydb/_tx_ctx_impl.py b/contrib/python/ydb/py3/ydb/_tx_ctx_impl.py index 3c7eec39e6..3c7eec39e6 100644 --- a/ydb/public/sdk/python3/ydb/_tx_ctx_impl.py +++ b/contrib/python/ydb/py3/ydb/_tx_ctx_impl.py diff --git a/ydb/public/sdk/python3/ydb/_utilities.py b/contrib/python/ydb/py3/ydb/_utilities.py index e89b0af315..e89b0af315 100644 --- a/ydb/public/sdk/python3/ydb/_utilities.py +++ b/contrib/python/ydb/py3/ydb/_utilities.py diff --git a/ydb/public/sdk/python3/ydb/aio/__init__.py b/contrib/python/ydb/py3/ydb/aio/__init__.py index acc44db57a..acc44db57a 100644 --- a/ydb/public/sdk/python3/ydb/aio/__init__.py +++ b/contrib/python/ydb/py3/ydb/aio/__init__.py diff --git a/ydb/public/sdk/python3/ydb/aio/_utilities.py b/contrib/python/ydb/py3/ydb/aio/_utilities.py index 10cbead667..10cbead667 100644 --- a/ydb/public/sdk/python3/ydb/aio/_utilities.py +++ b/contrib/python/ydb/py3/ydb/aio/_utilities.py diff --git a/ydb/public/sdk/python3/ydb/aio/connection.py b/contrib/python/ydb/py3/ydb/aio/connection.py index 9c66149002..9c66149002 100644 --- a/ydb/public/sdk/python3/ydb/aio/connection.py +++ b/contrib/python/ydb/py3/ydb/aio/connection.py diff --git a/ydb/public/sdk/python3/ydb/aio/credentials.py b/contrib/python/ydb/py3/ydb/aio/credentials.py index 48db925eba..48db925eba 100644 --- a/ydb/public/sdk/python3/ydb/aio/credentials.py +++ b/contrib/python/ydb/py3/ydb/aio/credentials.py diff --git a/ydb/public/sdk/python3/ydb/aio/driver.py b/contrib/python/ydb/py3/ydb/aio/driver.py index 0f4f3630f9..0f4f3630f9 100644 --- a/ydb/public/sdk/python3/ydb/aio/driver.py +++ b/contrib/python/ydb/py3/ydb/aio/driver.py diff --git a/ydb/public/sdk/python3/ydb/aio/iam.py b/contrib/python/ydb/py3/ydb/aio/iam.py index eab8faffe0..eab8faffe0 100644 --- a/ydb/public/sdk/python3/ydb/aio/iam.py +++ b/contrib/python/ydb/py3/ydb/aio/iam.py diff --git a/ydb/public/sdk/python3/ydb/aio/pool.py b/contrib/python/ydb/py3/ydb/aio/pool.py index c637a7ca96..c637a7ca96 100644 --- a/ydb/public/sdk/python3/ydb/aio/pool.py +++ b/contrib/python/ydb/py3/ydb/aio/pool.py diff --git a/ydb/public/sdk/python3/ydb/aio/resolver.py b/contrib/python/ydb/py3/ydb/aio/resolver.py index e8d27bac58..e8d27bac58 100644 --- a/ydb/public/sdk/python3/ydb/aio/resolver.py +++ b/contrib/python/ydb/py3/ydb/aio/resolver.py diff --git a/ydb/public/sdk/python3/ydb/aio/scheme.py b/contrib/python/ydb/py3/ydb/aio/scheme.py index 8b1233ec74..8b1233ec74 100644 --- a/ydb/public/sdk/python3/ydb/aio/scheme.py +++ b/contrib/python/ydb/py3/ydb/aio/scheme.py diff --git a/ydb/public/sdk/python3/ydb/aio/table.py b/contrib/python/ydb/py3/ydb/aio/table.py index 2a33cf786e..2a33cf786e 100644 --- a/ydb/public/sdk/python3/ydb/aio/table.py +++ b/contrib/python/ydb/py3/ydb/aio/table.py diff --git a/ydb/public/sdk/python3/ydb/auth_helpers.py b/contrib/python/ydb/py3/ydb/auth_helpers.py index 6399c3cfdf..6399c3cfdf 100644 --- a/ydb/public/sdk/python3/ydb/auth_helpers.py +++ b/contrib/python/ydb/py3/ydb/auth_helpers.py diff --git a/ydb/public/sdk/python3/ydb/connection.py b/contrib/python/ydb/py3/ydb/connection.py index 1c4bd9c78d..1c4bd9c78d 100644 --- a/ydb/public/sdk/python3/ydb/connection.py +++ b/contrib/python/ydb/py3/ydb/connection.py diff --git a/ydb/public/sdk/python3/ydb/convert.py b/contrib/python/ydb/py3/ydb/convert.py index 6c4164bc44..6c4164bc44 100644 --- a/ydb/public/sdk/python3/ydb/convert.py +++ b/contrib/python/ydb/py3/ydb/convert.py diff --git a/ydb/public/sdk/python3/ydb/credentials.py b/contrib/python/ydb/py3/ydb/credentials.py index ab50279846..ab50279846 100644 --- a/ydb/public/sdk/python3/ydb/credentials.py +++ b/contrib/python/ydb/py3/ydb/credentials.py diff --git a/ydb/public/sdk/python3/ydb/dbapi/__init__.py b/contrib/python/ydb/py3/ydb/dbapi/__init__.py index 7363921192..7363921192 100644 --- a/ydb/public/sdk/python3/ydb/dbapi/__init__.py +++ b/contrib/python/ydb/py3/ydb/dbapi/__init__.py diff --git a/ydb/public/sdk/python3/ydb/dbapi/connection.py b/contrib/python/ydb/py3/ydb/dbapi/connection.py index 4c22acd132..4c22acd132 100644 --- a/ydb/public/sdk/python3/ydb/dbapi/connection.py +++ b/contrib/python/ydb/py3/ydb/dbapi/connection.py diff --git a/ydb/public/sdk/python3/ydb/dbapi/cursor.py b/contrib/python/ydb/py3/ydb/dbapi/cursor.py index 92c471d562..92c471d562 100644 --- a/ydb/public/sdk/python3/ydb/dbapi/cursor.py +++ b/contrib/python/ydb/py3/ydb/dbapi/cursor.py diff --git a/ydb/public/sdk/python3/ydb/dbapi/errors.py b/contrib/python/ydb/py3/ydb/dbapi/errors.py index 65e453856a..65e453856a 100644 --- a/ydb/public/sdk/python3/ydb/dbapi/errors.py +++ b/contrib/python/ydb/py3/ydb/dbapi/errors.py diff --git a/ydb/public/sdk/python3/ydb/default_pem.py b/contrib/python/ydb/py3/ydb/default_pem.py index b8272efd29..b8272efd29 100644 --- a/ydb/public/sdk/python3/ydb/default_pem.py +++ b/contrib/python/ydb/py3/ydb/default_pem.py diff --git a/ydb/public/sdk/python3/ydb/driver.py b/contrib/python/ydb/py3/ydb/driver.py index 89109b9b57..89109b9b57 100644 --- a/ydb/public/sdk/python3/ydb/driver.py +++ b/contrib/python/ydb/py3/ydb/driver.py diff --git a/ydb/public/sdk/python3/ydb/export.py b/contrib/python/ydb/py3/ydb/export.py index 827925c7bc..827925c7bc 100644 --- a/ydb/public/sdk/python3/ydb/export.py +++ b/contrib/python/ydb/py3/ydb/export.py diff --git a/ydb/public/sdk/python3/ydb/global_settings.py b/contrib/python/ydb/py3/ydb/global_settings.py index 8edac3f4b4..8edac3f4b4 100644 --- a/ydb/public/sdk/python3/ydb/global_settings.py +++ b/contrib/python/ydb/py3/ydb/global_settings.py diff --git a/ydb/public/sdk/python3/ydb/iam/__init__.py b/contrib/python/ydb/py3/ydb/iam/__init__.py index 7167efe13e..7167efe13e 100644 --- a/ydb/public/sdk/python3/ydb/iam/__init__.py +++ b/contrib/python/ydb/py3/ydb/iam/__init__.py diff --git a/ydb/public/sdk/python3/ydb/iam/auth.py b/contrib/python/ydb/py3/ydb/iam/auth.py index 82e7c9f6c8..82e7c9f6c8 100644 --- a/ydb/public/sdk/python3/ydb/iam/auth.py +++ b/contrib/python/ydb/py3/ydb/iam/auth.py diff --git a/ydb/public/sdk/python3/ydb/import_client.py b/contrib/python/ydb/py3/ydb/import_client.py index d94294ca7c..d94294ca7c 100644 --- a/ydb/public/sdk/python3/ydb/import_client.py +++ b/contrib/python/ydb/py3/ydb/import_client.py diff --git a/ydb/public/sdk/python3/ydb/interceptor.py b/contrib/python/ydb/py3/ydb/interceptor.py index 5405177a8e..5405177a8e 100644 --- a/ydb/public/sdk/python3/ydb/interceptor.py +++ b/contrib/python/ydb/py3/ydb/interceptor.py diff --git a/ydb/public/sdk/python3/ydb/issues.py b/contrib/python/ydb/py3/ydb/issues.py index 065dcbc80c..065dcbc80c 100644 --- a/ydb/public/sdk/python3/ydb/issues.py +++ b/contrib/python/ydb/py3/ydb/issues.py diff --git a/ydb/public/sdk/python3/ydb/operation.py b/contrib/python/ydb/py3/ydb/operation.py index 6084ef0f18..6084ef0f18 100644 --- a/ydb/public/sdk/python3/ydb/operation.py +++ b/contrib/python/ydb/py3/ydb/operation.py diff --git a/ydb/public/sdk/python3/ydb/pool.py b/contrib/python/ydb/py3/ydb/pool.py index 1e75950ea8..1e75950ea8 100644 --- a/ydb/public/sdk/python3/ydb/pool.py +++ b/contrib/python/ydb/py3/ydb/pool.py diff --git a/ydb/public/sdk/python3/ydb/resolver.py b/contrib/python/ydb/py3/ydb/resolver.py index b795af928c..b795af928c 100644 --- a/ydb/public/sdk/python3/ydb/resolver.py +++ b/contrib/python/ydb/py3/ydb/resolver.py diff --git a/ydb/public/sdk/python3/ydb/scheme.py b/contrib/python/ydb/py3/ydb/scheme.py index 6019c763e5..6019c763e5 100644 --- a/ydb/public/sdk/python3/ydb/scheme.py +++ b/contrib/python/ydb/py3/ydb/scheme.py diff --git a/ydb/public/sdk/python3/ydb/scripting.py b/contrib/python/ydb/py3/ydb/scripting.py index 595f4e1695..595f4e1695 100644 --- a/ydb/public/sdk/python3/ydb/scripting.py +++ b/contrib/python/ydb/py3/ydb/scripting.py diff --git a/ydb/public/sdk/python3/ydb/settings.py b/contrib/python/ydb/py3/ydb/settings.py index 6739a46fab..6739a46fab 100644 --- a/ydb/public/sdk/python3/ydb/settings.py +++ b/contrib/python/ydb/py3/ydb/settings.py diff --git a/ydb/public/sdk/python3/ydb/sqlalchemy/__init__.py b/contrib/python/ydb/py3/ydb/sqlalchemy/__init__.py index d1fc4c90de..d1fc4c90de 100644 --- a/ydb/public/sdk/python3/ydb/sqlalchemy/__init__.py +++ b/contrib/python/ydb/py3/ydb/sqlalchemy/__init__.py diff --git a/ydb/public/sdk/python3/ydb/sqlalchemy/types.py b/contrib/python/ydb/py3/ydb/sqlalchemy/types.py index f6e10ccd6f..f6e10ccd6f 100644 --- a/ydb/public/sdk/python3/ydb/sqlalchemy/types.py +++ b/contrib/python/ydb/py3/ydb/sqlalchemy/types.py diff --git a/ydb/public/sdk/python3/ydb/table.py b/contrib/python/ydb/py3/ydb/table.py index c21392bb4c..c21392bb4c 100644 --- a/ydb/public/sdk/python3/ydb/table.py +++ b/contrib/python/ydb/py3/ydb/table.py diff --git a/ydb/public/sdk/python3/ydb/topic.py b/contrib/python/ydb/py3/ydb/topic.py index 2175af47f7..2175af47f7 100644 --- a/ydb/public/sdk/python3/ydb/topic.py +++ b/contrib/python/ydb/py3/ydb/topic.py diff --git a/ydb/public/sdk/python3/ydb/tracing.py b/contrib/python/ydb/py3/ydb/tracing.py index fcc780d6b5..fcc780d6b5 100644 --- a/ydb/public/sdk/python3/ydb/tracing.py +++ b/contrib/python/ydb/py3/ydb/tracing.py diff --git a/ydb/public/sdk/python3/ydb/types.py b/contrib/python/ydb/py3/ydb/types.py index cf13aac0a2..cf13aac0a2 100644 --- a/ydb/public/sdk/python3/ydb/types.py +++ b/contrib/python/ydb/py3/ydb/types.py diff --git a/ydb/public/sdk/python3/ydb/ydb_version.py b/contrib/python/ydb/py3/ydb/ydb_version.py index 709cb0a9e5..709cb0a9e5 100644 --- a/ydb/public/sdk/python3/ydb/ydb_version.py +++ b/contrib/python/ydb/py3/ydb/ydb_version.py diff --git a/contrib/restricted/llhttp/LICENSE-MIT b/contrib/restricted/llhttp/LICENSE-MIT new file mode 100644 index 0000000000..6c1512dd6b --- /dev/null +++ b/contrib/restricted/llhttp/LICENSE-MIT @@ -0,0 +1,22 @@ +This software is licensed under the MIT License. + +Copyright Fedor Indutny, 2018. + +Permission is hereby granted, free of charge, to any person obtaining a +copy of this software and associated documentation files (the +"Software"), to deal in the Software without restriction, including +without limitation the rights to use, copy, modify, merge, publish, +distribute, sublicense, and/or sell copies of the Software, and to permit +persons to whom the Software is furnished to do so, subject to the +following conditions: + +The above copyright notice and this permission notice shall be included +in all copies or substantial portions of the Software. + +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS +OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF +MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN +NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, +DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR +OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE +USE OR OTHER DEALINGS IN THE SOFTWARE. diff --git a/contrib/restricted/llhttp/README.md b/contrib/restricted/llhttp/README.md new file mode 100644 index 0000000000..e982ed0ae7 --- /dev/null +++ b/contrib/restricted/llhttp/README.md @@ -0,0 +1,482 @@ +# llhttp +[](https://github.com/nodejs/llhttp/actions?query=workflow%3ACI) + +Port of [http_parser][0] to [llparse][1]. + +## Why? + +Let's face it, [http_parser][0] is practically unmaintainable. Even +introduction of a single new method results in a significant code churn. + +This project aims to: + +* Make it maintainable +* Verifiable +* Improving benchmarks where possible + +More details in [Fedor Indutny's talk at JSConf EU 2019](https://youtu.be/x3k_5Mi66sY) + +## How? + +Over time, different approaches for improving [http_parser][0]'s code base +were tried. However, all of them failed due to resulting significant performance +degradation. + +This project is a port of [http_parser][0] to TypeScript. [llparse][1] is used +to generate the output C source file, which could be compiled and +linked with the embedder's program (like [Node.js][7]). + +## Performance + +So far llhttp outperforms http_parser: + +| | input size | bandwidth | reqs/sec | time | +|:----------------|-----------:|-------------:|-----------:|--------:| +| **llhttp** | 8192.00 mb | 1777.24 mb/s | 3583799.39 req/sec | 4.61 s | +| **http_parser** | 8192.00 mb | 694.66 mb/s | 1406180.33 req/sec | 11.79 s | + +llhttp is faster by approximately **156%**. + +## Maintenance + +llhttp project has about 1400 lines of TypeScript code describing the parser +itself and around 450 lines of C code and headers providing the helper methods. +The whole [http_parser][0] is implemented in approximately 2500 lines of C, and +436 lines of headers. + +All optimizations and multi-character matching in llhttp are generated +automatically, and thus doesn't add any extra maintenance cost. On the contrary, +most of http_parser's code is hand-optimized and unrolled. Instead describing +"how" it should parse the HTTP requests/responses, a maintainer should +implement the new features in [http_parser][0] cautiously, considering +possible performance degradation and manually optimizing the new code. + +## Verification + +The state machine graph is encoded explicitly in llhttp. The [llparse][1] +automatically checks the graph for absence of loops and correct reporting of the +input ranges (spans) like header names and values. In the future, additional +checks could be performed to get even stricter verification of the llhttp. + +## Usage + +```C +#include "stdio.h" +#include "llhttp.h" +#include "string.h" + +int handle_on_message_complete(llhttp_t* parser) { + fprintf(stdout, "Message completed!\n"); + return 0; +} + +int main() { + llhttp_t parser; + llhttp_settings_t settings; + + /*Initialize user callbacks and settings */ + llhttp_settings_init(&settings); + + /*Set user callback */ + settings.on_message_complete = handle_on_message_complete; + + /*Initialize the parser in HTTP_BOTH mode, meaning that it will select between + *HTTP_REQUEST and HTTP_RESPONSE parsing automatically while reading the first + *input. + */ + llhttp_init(&parser, HTTP_BOTH, &settings); + + /*Parse request! */ + const char* request = "GET / HTTP/1.1\r\n\r\n"; + int request_len = strlen(request); + + enum llhttp_errno err = llhttp_execute(&parser, request, request_len); + if (err == HPE_OK) { + fprintf(stdout, "Successfully parsed!\n"); + } else { + fprintf(stderr, "Parse error: %s %s\n", llhttp_errno_name(err), parser.reason); + } +} +``` +For more information on API usage, please refer to [src/native/api.h](https://github.com/nodejs/llhttp/blob/main/src/native/api.h). + +## API + +### llhttp_settings_t + +The settings object contains a list of callbacks that the parser will invoke. + +The following callbacks can return `0` (proceed normally), `-1` (error) or `HPE_PAUSED` (pause the parser): + +* `on_message_begin`: Invoked when a new request/response starts. +* `on_message_complete`: Invoked when a request/response has been completedly parsed. +* `on_url_complete`: Invoked after the URL has been parsed. +* `on_method_complete`: Invoked after the HTTP method has been parsed. +* `on_version_complete`: Invoked after the HTTP version has been parsed. +* `on_status_complete`: Invoked after the status code has been parsed. +* `on_header_field_complete`: Invoked after a header name has been parsed. +* `on_header_value_complete`: Invoked after a header value has been parsed. +* `on_chunk_header`: Invoked after a new chunk is started. The current chunk length is stored in `parser->content_length`. +* `on_chunk_extension_name_complete`: Invoked after a chunk extension name is started. +* `on_chunk_extension_value_complete`: Invoked after a chunk extension value is started. +* `on_chunk_complete`: Invoked after a new chunk is received. +* `on_reset`: Invoked after `on_message_complete` and before `on_message_begin` when a new message + is received on the same parser. This is not invoked for the first message of the parser. + +The following callbacks can return `0` (proceed normally), `-1` (error) or `HPE_USER` (error from the callback): + +* `on_url`: Invoked when another character of the URL is received. +* `on_status`: Invoked when another character of the status is received. +* `on_method`: Invoked when another character of the method is received. + When parser is created with `HTTP_BOTH` and the input is a response, this also invoked for the sequence `HTTP/` + of the first message. +* `on_version`: Invoked when another character of the version is received. +* `on_header_field`: Invoked when another character of a header name is received. +* `on_header_value`: Invoked when another character of a header value is received. +* `on_chunk_extension_name`: Invoked when another character of a chunk extension name is received. +* `on_chunk_extension_value`: Invoked when another character of a extension value is received. + +The callback `on_headers_complete`, invoked when headers are completed, can return: + +* `0`: Proceed normally. +* `1`: Assume that request/response has no body, and proceed to parsing the next message. +* `2`: Assume absence of body (as above) and make `llhttp_execute()` return `HPE_PAUSED_UPGRADE`. +* `-1`: Error +* `HPE_PAUSED`: Pause the parser. + +### `void llhttp_init(llhttp_t* parser, llhttp_type_t type, const llhttp_settings_t* settings)` + +Initialize the parser with specific type and user settings. + +### `uint8_t llhttp_get_type(llhttp_t* parser)` + +Returns the type of the parser. + +### `uint8_t llhttp_get_http_major(llhttp_t* parser)` + +Returns the major version of the HTTP protocol of the current request/response. + +### `uint8_t llhttp_get_http_minor(llhttp_t* parser)` + +Returns the minor version of the HTTP protocol of the current request/response. + +### `uint8_t llhttp_get_method(llhttp_t* parser)` + +Returns the method of the current request. + +### `int llhttp_get_status_code(llhttp_t* parser)` + +Returns the method of the current response. + +### `uint8_t llhttp_get_upgrade(llhttp_t* parser)` + +Returns `1` if request includes the `Connection: upgrade` header. + +### `void llhttp_reset(llhttp_t* parser)` + +Reset an already initialized parser back to the start state, preserving the +existing parser type, callback settings, user data, and lenient flags. + +### `void llhttp_settings_init(llhttp_settings_t* settings)` + +Initialize the settings object. + +### `llhttp_errno_t llhttp_execute(llhttp_t* parser, const char* data, size_t len)` + +Parse full or partial request/response, invoking user callbacks along the way. + +If any of `llhttp_data_cb` returns errno not equal to `HPE_OK` - the parsing interrupts, +and such errno is returned from `llhttp_execute()`. If `HPE_PAUSED` was used as a errno, +the execution can be resumed with `llhttp_resume()` call. + +In a special case of CONNECT/Upgrade request/response `HPE_PAUSED_UPGRADE` is returned +after fully parsing the request/response. If the user wishes to continue parsing, +they need to invoke `llhttp_resume_after_upgrade()`. + +**if this function ever returns a non-pause type error, it will continue to return +the same error upon each successive call up until `llhttp_init()` is called.** + +### `llhttp_errno_t llhttp_finish(llhttp_t* parser)` + +This method should be called when the other side has no further bytes to +send (e.g. shutdown of readable side of the TCP connection.) + +Requests without `Content-Length` and other messages might require treating +all incoming bytes as the part of the body, up to the last byte of the +connection. + +This method will invoke `on_message_complete()` callback if the +request was terminated safely. Otherwise a error code would be returned. + + +### `int llhttp_message_needs_eof(const llhttp_t* parser)` + +Returns `1` if the incoming message is parsed until the last byte, and has to be completed by calling `llhttp_finish()` on EOF. + +### `int llhttp_should_keep_alive(const llhttp_t* parser)` + +Returns `1` if there might be any other messages following the last that was +successfully parsed. + +### `void llhttp_pause(llhttp_t* parser)` + +Make further calls of `llhttp_execute()` return `HPE_PAUSED` and set +appropriate error reason. + +**Do not call this from user callbacks! User callbacks must return +`HPE_PAUSED` if pausing is required.** + +### `void llhttp_resume(llhttp_t* parser)` + +Might be called to resume the execution after the pause in user's callback. + +See `llhttp_execute()` above for details. + +**Call this only if `llhttp_execute()` returns `HPE_PAUSED`.** + +### `void llhttp_resume_after_upgrade(llhttp_t* parser)` + +Might be called to resume the execution after the pause in user's callback. +See `llhttp_execute()` above for details. + +**Call this only if `llhttp_execute()` returns `HPE_PAUSED_UPGRADE`** + +### `llhttp_errno_t llhttp_get_errno(const llhttp_t* parser)` + +Returns the latest error. + +### `const char* llhttp_get_error_reason(const llhttp_t* parser)` + +Returns the verbal explanation of the latest returned error. + +**User callback should set error reason when returning the error. See +`llhttp_set_error_reason()` for details.** + +### `void llhttp_set_error_reason(llhttp_t* parser, const char* reason)` + +Assign verbal description to the returned error. Must be called in user +callbacks right before returning the errno. + +**`HPE_USER` error code might be useful in user callbacks.** + +### `const char* llhttp_get_error_pos(const llhttp_t* parser)` + +Returns the pointer to the last parsed byte before the returned error. The +pointer is relative to the `data` argument of `llhttp_execute()`. + +**This method might be useful for counting the number of parsed bytes.** + +### `const char* llhttp_errno_name(llhttp_errno_t err)` + +Returns textual name of error code. + +### `const char* llhttp_method_name(llhttp_method_t method)` + +Returns textual name of HTTP method. + +### `const char* llhttp_status_name(llhttp_status_t status)` + +Returns textual name of HTTP status. + +### `void llhttp_set_lenient_headers(llhttp_t* parser, int enabled)` + +Enables/disables lenient header value parsing (disabled by default). +Lenient parsing disables header value token checks, extending llhttp's +protocol support to highly non-compliant clients/server. + +No `HPE_INVALID_HEADER_TOKEN` will be raised for incorrect header values when +lenient parsing is "on". + +**Enabling this flag can pose a security issue since you will be exposed to request smuggling attacks. USE WITH CAUTION!** + +### `void llhttp_set_lenient_chunked_length(llhttp_t* parser, int enabled)` + +Enables/disables lenient handling of conflicting `Transfer-Encoding` and +`Content-Length` headers (disabled by default). + +Normally `llhttp` would error when `Transfer-Encoding` is present in +conjunction with `Content-Length`. + +This error is important to prevent HTTP request smuggling, but may be less desirable +for small number of cases involving legacy servers. + +**Enabling this flag can pose a security issue since you will be exposed to request smuggling attacks. USE WITH CAUTION!** + +### `void llhttp_set_lenient_keep_alive(llhttp_t* parser, int enabled)` + +Enables/disables lenient handling of `Connection: close` and HTTP/1.0 +requests responses. + +Normally `llhttp` would error the HTTP request/response +after the request/response with `Connection: close` and `Content-Length`. + +This is important to prevent cache poisoning attacks, +but might interact badly with outdated and insecure clients. + +With this flag the extra request/response will be parsed normally. + +**Enabling this flag can pose a security issue since you will be exposed to poisoning attacks. USE WITH CAUTION!** + +### `void llhttp_set_lenient_transfer_encoding(llhttp_t* parser, int enabled)` + +Enables/disables lenient handling of `Transfer-Encoding` header. + +Normally `llhttp` would error when a `Transfer-Encoding` has `chunked` value +and another value after it (either in a single header or in multiple +headers whose value are internally joined using `, `). + +This is mandated by the spec to reliably determine request body size and thus +avoid request smuggling. + +With this flag the extra value will be parsed normally. + +**Enabling this flag can pose a security issue since you will be exposed to request smuggling attacks. USE WITH CAUTION!** + +### `void llhttp_set_lenient_version(llhttp_t* parser, int enabled)` + +Enables/disables lenient handling of HTTP version. + +Normally `llhttp` would error when the HTTP version in the request or status line +is not `0.9`, `1.0`, `1.1` or `2.0`. +With this flag the extra value will be parsed normally. + +**Enabling this flag can pose a security issue since you will allow unsupported HTTP versions. USE WITH CAUTION!** + +### `void llhttp_set_lenient_data_after_close(llhttp_t* parser, int enabled)` + +Enables/disables lenient handling of additional data received after a message ends +and keep-alive is disabled. + +Normally `llhttp` would error when additional unexpected data is received if the message +contains the `Connection` header with `close` value. +With this flag the extra data will discarded without throwing an error. + +**Enabling this flag can pose a security issue since you will be exposed to poisoning attacks. USE WITH CAUTION!** + +### `void llhttp_set_lenient_optional_lf_after_cr(llhttp_t* parser, int enabled)` + +Enables/disables lenient handling of incomplete CRLF sequences. + +Normally `llhttp` would error when a CR is not followed by LF when terminating the +request line, the status line, the headers or a chunk header. +With this flag only a CR is required to terminate such sections. + +**Enabling this flag can pose a security issue since you will be exposed to request smuggling attacks. USE WITH CAUTION!** + +### `void llhttp_set_lenient_optional_crlf_after_chunk(llhttp_t* parser, int enabled)` + +Enables/disables lenient handling of chunks not separated via CRLF. + +Normally `llhttp` would error when after a chunk data a CRLF is missing before +starting a new chunk. +With this flag the new chunk can start immediately after the previous one. + +**Enabling this flag can pose a security issue since you will be exposed to request smuggling attacks. USE WITH CAUTION!** + +## Build Instructions + +Make sure you have [Node.js](https://nodejs.org/), npm and npx installed. Then under project directory run: + +```sh +npm install +make +``` + +--- + +### Bindings to other languages + +* Lua: [MunifTanjim/llhttp.lua][11] +* Python: [pallas/pyllhttp][8] +* Ruby: [metabahn/llhttp][9] +* Rust: [JackLiar/rust-llhttp][10] + +### Using with CMake + +If you want to use this library in a CMake project as a shared library, you can use the snippet below. + +``` +FetchContent_Declare(llhttp + URL "https://github.com/nodejs/llhttp/archive/refs/tags/release/v8.1.0.tar.gz") + +FetchContent_MakeAvailable(llhttp) + +# Link with the llhttp_shared target +target_link_libraries(${EXAMPLE_PROJECT_NAME} ${PROJECT_LIBRARIES} llhttp_shared ${PROJECT_NAME}) +``` + +If you want to use this library in a CMake project as a static library, you can set some cache variables first. + +``` +FetchContent_Declare(llhttp + URL "https://github.com/nodejs/llhttp/archive/refs/tags/release/v8.1.0.tar.gz") + +set(BUILD_SHARED_LIBS OFF CACHE INTERNAL "") +set(BUILD_STATIC_LIBS ON CACHE INTERNAL "") +FetchContent_MakeAvailable(llhttp) + +# Link with the llhttp_static target +target_link_libraries(${EXAMPLE_PROJECT_NAME} ${PROJECT_LIBRARIES} llhttp_static ${PROJECT_NAME}) +``` + +_Note that using the git repo directly (e.g., via a git repo url and tag) will not work with FetchContent_Declare because [CMakeLists.txt](./CMakeLists.txt) requires string replacements (e.g., `_RELEASE_`) before it will build._ + +## Building on Windows + +### Installation + +* `choco install git` +* `choco install node` +* `choco install llvm` (or install the `C++ Clang tools for Windows` optional package from the Visual Studio 2019 installer) +* `choco install make` (or if you have MinGW, it comes bundled) + +1. Ensure that `Clang` and `make` are in your system path. +2. Using Git Bash, clone the repo to your preferred location. +3. Cd into the cloned directory and run `npm install` +5. Run `make` +6. Your `repo/build` directory should now have `libllhttp.a` and `libllhttp.so` static and dynamic libraries. +7. When building your executable, you can link to these libraries. Make sure to set the build folder as an include path when building so you can reference the declarations in `repo/build/llhttp.h`. + +### A simple example on linking with the library: + +Assuming you have an executable `main.cpp` in your current working directory, you would run: `clang++ -Os -g3 -Wall -Wextra -Wno-unused-parameter -I/path/to/llhttp/build main.cpp /path/to/llhttp/build/libllhttp.a -o main.exe`. + +If you are getting `unresolved external symbol` linker errors you are likely attempting to build `llhttp.c` without linking it with object files from `api.c` and `http.c`. + +#### LICENSE + +This software is licensed under the MIT License. + +Copyright Fedor Indutny, 2018. + +Permission is hereby granted, free of charge, to any person obtaining a +copy of this software and associated documentation files (the +"Software"), to deal in the Software without restriction, including +without limitation the rights to use, copy, modify, merge, publish, +distribute, sublicense, and/or sell copies of the Software, and to permit +persons to whom the Software is furnished to do so, subject to the +following conditions: + +The above copyright notice and this permission notice shall be included +in all copies or substantial portions of the Software. + +THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS +OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF +MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN +NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, +DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR +OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE +USE OR OTHER DEALINGS IN THE SOFTWARE. + +[0]: https://github.com/nodejs/http-parser +[1]: https://github.com/nodejs/llparse +[2]: https://en.wikipedia.org/wiki/Register_allocation#Spilling +[3]: https://en.wikipedia.org/wiki/Tail_call +[4]: https://llvm.org/docs/LangRef.html +[5]: https://llvm.org/docs/LangRef.html#call-instruction +[6]: https://clang.llvm.org/ +[7]: https://github.com/nodejs/node +[8]: https://github.com/pallas/pyllhttp +[9]: https://github.com/metabahn/llhttp +[10]: https://github.com/JackLiar/rust-llhttp +[11]: https://github.com/MunifTanjim/llhttp.lua diff --git a/contrib/restricted/llhttp/include/llhttp.h b/contrib/restricted/llhttp/include/llhttp.h new file mode 100644 index 0000000000..6588ae5e36 --- /dev/null +++ b/contrib/restricted/llhttp/include/llhttp.h @@ -0,0 +1,871 @@ + +#ifndef INCLUDE_LLHTTP_H_ +#define INCLUDE_LLHTTP_H_ + +#define LLHTTP_VERSION_MAJOR 9 +#define LLHTTP_VERSION_MINOR 0 +#define LLHTTP_VERSION_PATCH 1 + +#ifndef INCLUDE_LLHTTP_ITSELF_H_ +#define INCLUDE_LLHTTP_ITSELF_H_ +#ifdef __cplusplus +extern "C" { +#endif + +#include <stdint.h> + +typedef struct llhttp__internal_s llhttp__internal_t; +struct llhttp__internal_s { + int32_t _index; + void* _span_pos0; + void* _span_cb0; + int32_t error; + const char* reason; + const char* error_pos; + void* data; + void* _current; + uint64_t content_length; + uint8_t type; + uint8_t method; + uint8_t http_major; + uint8_t http_minor; + uint8_t header_state; + uint8_t lenient_flags; + uint8_t upgrade; + uint8_t finish; + uint16_t flags; + uint16_t status_code; + uint8_t initial_message_completed; + void* settings; +}; + +int llhttp__internal_init(llhttp__internal_t* s); +int llhttp__internal_execute(llhttp__internal_t* s, const char* p, const char* endp); + +#ifdef __cplusplus +} /* extern "C" */ +#endif +#endif /* INCLUDE_LLHTTP_ITSELF_H_ */ + + +#ifndef LLLLHTTP_C_HEADERS_ +#define LLLLHTTP_C_HEADERS_ +#ifdef __cplusplus +extern "C" { +#endif + +enum llhttp_errno { + HPE_OK = 0, + HPE_INTERNAL = 1, + HPE_STRICT = 2, + HPE_CR_EXPECTED = 25, + HPE_LF_EXPECTED = 3, + HPE_UNEXPECTED_CONTENT_LENGTH = 4, + HPE_UNEXPECTED_SPACE = 30, + HPE_CLOSED_CONNECTION = 5, + HPE_INVALID_METHOD = 6, + HPE_INVALID_URL = 7, + HPE_INVALID_CONSTANT = 8, + HPE_INVALID_VERSION = 9, + HPE_INVALID_HEADER_TOKEN = 10, + HPE_INVALID_CONTENT_LENGTH = 11, + HPE_INVALID_CHUNK_SIZE = 12, + HPE_INVALID_STATUS = 13, + HPE_INVALID_EOF_STATE = 14, + HPE_INVALID_TRANSFER_ENCODING = 15, + HPE_CB_MESSAGE_BEGIN = 16, + HPE_CB_HEADERS_COMPLETE = 17, + HPE_CB_MESSAGE_COMPLETE = 18, + HPE_CB_CHUNK_HEADER = 19, + HPE_CB_CHUNK_COMPLETE = 20, + HPE_PAUSED = 21, + HPE_PAUSED_UPGRADE = 22, + HPE_PAUSED_H2_UPGRADE = 23, + HPE_USER = 24, + HPE_CB_URL_COMPLETE = 26, + HPE_CB_STATUS_COMPLETE = 27, + HPE_CB_METHOD_COMPLETE = 32, + HPE_CB_VERSION_COMPLETE = 33, + HPE_CB_HEADER_FIELD_COMPLETE = 28, + HPE_CB_HEADER_VALUE_COMPLETE = 29, + HPE_CB_CHUNK_EXTENSION_NAME_COMPLETE = 34, + HPE_CB_CHUNK_EXTENSION_VALUE_COMPLETE = 35, + HPE_CB_RESET = 31 +}; +typedef enum llhttp_errno llhttp_errno_t; + +enum llhttp_flags { + F_CONNECTION_KEEP_ALIVE = 0x1, + F_CONNECTION_CLOSE = 0x2, + F_CONNECTION_UPGRADE = 0x4, + F_CHUNKED = 0x8, + F_UPGRADE = 0x10, + F_CONTENT_LENGTH = 0x20, + F_SKIPBODY = 0x40, + F_TRAILING = 0x80, + F_TRANSFER_ENCODING = 0x200 +}; +typedef enum llhttp_flags llhttp_flags_t; + +enum llhttp_lenient_flags { + LENIENT_HEADERS = 0x1, + LENIENT_CHUNKED_LENGTH = 0x2, + LENIENT_KEEP_ALIVE = 0x4, + LENIENT_TRANSFER_ENCODING = 0x8, + LENIENT_VERSION = 0x10, + LENIENT_DATA_AFTER_CLOSE = 0x20, + LENIENT_OPTIONAL_LF_AFTER_CR = 0x40, + LENIENT_OPTIONAL_CRLF_AFTER_CHUNK = 0x80 +}; +typedef enum llhttp_lenient_flags llhttp_lenient_flags_t; + +enum llhttp_type { + HTTP_BOTH = 0, + HTTP_REQUEST = 1, + HTTP_RESPONSE = 2 +}; +typedef enum llhttp_type llhttp_type_t; + +enum llhttp_finish { + HTTP_FINISH_SAFE = 0, + HTTP_FINISH_SAFE_WITH_CB = 1, + HTTP_FINISH_UNSAFE = 2 +}; +typedef enum llhttp_finish llhttp_finish_t; + +enum llhttp_method { + HTTP_DELETE = 0, + HTTP_GET = 1, + HTTP_HEAD = 2, + HTTP_POST = 3, + HTTP_PUT = 4, + HTTP_CONNECT = 5, + HTTP_OPTIONS = 6, + HTTP_TRACE = 7, + HTTP_COPY = 8, + HTTP_LOCK = 9, + HTTP_MKCOL = 10, + HTTP_MOVE = 11, + HTTP_PROPFIND = 12, + HTTP_PROPPATCH = 13, + HTTP_SEARCH = 14, + HTTP_UNLOCK = 15, + HTTP_BIND = 16, + HTTP_REBIND = 17, + HTTP_UNBIND = 18, + HTTP_ACL = 19, + HTTP_REPORT = 20, + HTTP_MKACTIVITY = 21, + HTTP_CHECKOUT = 22, + HTTP_MERGE = 23, + HTTP_MSEARCH = 24, + HTTP_NOTIFY = 25, + HTTP_SUBSCRIBE = 26, + HTTP_UNSUBSCRIBE = 27, + HTTP_PATCH = 28, + HTTP_PURGE = 29, + HTTP_MKCALENDAR = 30, + HTTP_LINK = 31, + HTTP_UNLINK = 32, + HTTP_SOURCE = 33, + HTTP_PRI = 34, + HTTP_DESCRIBE = 35, + HTTP_ANNOUNCE = 36, + HTTP_SETUP = 37, + HTTP_PLAY = 38, + HTTP_PAUSE = 39, + HTTP_TEARDOWN = 40, + HTTP_GET_PARAMETER = 41, + HTTP_SET_PARAMETER = 42, + HTTP_REDIRECT = 43, + HTTP_RECORD = 44, + HTTP_FLUSH = 45 +}; +typedef enum llhttp_method llhttp_method_t; + +enum llhttp_status { + HTTP_STATUS_CONTINUE = 100, + HTTP_STATUS_SWITCHING_PROTOCOLS = 101, + HTTP_STATUS_PROCESSING = 102, + HTTP_STATUS_EARLY_HINTS = 103, + HTTP_STATUS_RESPONSE_IS_STALE = 110, + HTTP_STATUS_REVALIDATION_FAILED = 111, + HTTP_STATUS_DISCONNECTED_OPERATION = 112, + HTTP_STATUS_HEURISTIC_EXPIRATION = 113, + HTTP_STATUS_MISCELLANEOUS_WARNING = 199, + HTTP_STATUS_OK = 200, + HTTP_STATUS_CREATED = 201, + HTTP_STATUS_ACCEPTED = 202, + HTTP_STATUS_NON_AUTHORITATIVE_INFORMATION = 203, + HTTP_STATUS_NO_CONTENT = 204, + HTTP_STATUS_RESET_CONTENT = 205, + HTTP_STATUS_PARTIAL_CONTENT = 206, + HTTP_STATUS_MULTI_STATUS = 207, + HTTP_STATUS_ALREADY_REPORTED = 208, + HTTP_STATUS_TRANSFORMATION_APPLIED = 214, + HTTP_STATUS_IM_USED = 226, + HTTP_STATUS_MISCELLANEOUS_PERSISTENT_WARNING = 299, + HTTP_STATUS_MULTIPLE_CHOICES = 300, + HTTP_STATUS_MOVED_PERMANENTLY = 301, + HTTP_STATUS_FOUND = 302, + HTTP_STATUS_SEE_OTHER = 303, + HTTP_STATUS_NOT_MODIFIED = 304, + HTTP_STATUS_USE_PROXY = 305, + HTTP_STATUS_SWITCH_PROXY = 306, + HTTP_STATUS_TEMPORARY_REDIRECT = 307, + HTTP_STATUS_PERMANENT_REDIRECT = 308, + HTTP_STATUS_BAD_REQUEST = 400, + HTTP_STATUS_UNAUTHORIZED = 401, + HTTP_STATUS_PAYMENT_REQUIRED = 402, + HTTP_STATUS_FORBIDDEN = 403, + HTTP_STATUS_NOT_FOUND = 404, + HTTP_STATUS_METHOD_NOT_ALLOWED = 405, + HTTP_STATUS_NOT_ACCEPTABLE = 406, + HTTP_STATUS_PROXY_AUTHENTICATION_REQUIRED = 407, + HTTP_STATUS_REQUEST_TIMEOUT = 408, + HTTP_STATUS_CONFLICT = 409, + HTTP_STATUS_GONE = 410, + HTTP_STATUS_LENGTH_REQUIRED = 411, + HTTP_STATUS_PRECONDITION_FAILED = 412, + HTTP_STATUS_PAYLOAD_TOO_LARGE = 413, + HTTP_STATUS_URI_TOO_LONG = 414, + HTTP_STATUS_UNSUPPORTED_MEDIA_TYPE = 415, + HTTP_STATUS_RANGE_NOT_SATISFIABLE = 416, + HTTP_STATUS_EXPECTATION_FAILED = 417, + HTTP_STATUS_IM_A_TEAPOT = 418, + HTTP_STATUS_PAGE_EXPIRED = 419, + HTTP_STATUS_ENHANCE_YOUR_CALM = 420, + HTTP_STATUS_MISDIRECTED_REQUEST = 421, + HTTP_STATUS_UNPROCESSABLE_ENTITY = 422, + HTTP_STATUS_LOCKED = 423, + HTTP_STATUS_FAILED_DEPENDENCY = 424, + HTTP_STATUS_TOO_EARLY = 425, + HTTP_STATUS_UPGRADE_REQUIRED = 426, + HTTP_STATUS_PRECONDITION_REQUIRED = 428, + HTTP_STATUS_TOO_MANY_REQUESTS = 429, + HTTP_STATUS_REQUEST_HEADER_FIELDS_TOO_LARGE_UNOFFICIAL = 430, + HTTP_STATUS_REQUEST_HEADER_FIELDS_TOO_LARGE = 431, + HTTP_STATUS_LOGIN_TIMEOUT = 440, + HTTP_STATUS_NO_RESPONSE = 444, + HTTP_STATUS_RETRY_WITH = 449, + HTTP_STATUS_BLOCKED_BY_PARENTAL_CONTROL = 450, + HTTP_STATUS_UNAVAILABLE_FOR_LEGAL_REASONS = 451, + HTTP_STATUS_CLIENT_CLOSED_LOAD_BALANCED_REQUEST = 460, + HTTP_STATUS_INVALID_X_FORWARDED_FOR = 463, + HTTP_STATUS_REQUEST_HEADER_TOO_LARGE = 494, + HTTP_STATUS_SSL_CERTIFICATE_ERROR = 495, + HTTP_STATUS_SSL_CERTIFICATE_REQUIRED = 496, + HTTP_STATUS_HTTP_REQUEST_SENT_TO_HTTPS_PORT = 497, + HTTP_STATUS_INVALID_TOKEN = 498, + HTTP_STATUS_CLIENT_CLOSED_REQUEST = 499, + HTTP_STATUS_INTERNAL_SERVER_ERROR = 500, + HTTP_STATUS_NOT_IMPLEMENTED = 501, + HTTP_STATUS_BAD_GATEWAY = 502, + HTTP_STATUS_SERVICE_UNAVAILABLE = 503, + HTTP_STATUS_GATEWAY_TIMEOUT = 504, + HTTP_STATUS_HTTP_VERSION_NOT_SUPPORTED = 505, + HTTP_STATUS_VARIANT_ALSO_NEGOTIATES = 506, + HTTP_STATUS_INSUFFICIENT_STORAGE = 507, + HTTP_STATUS_LOOP_DETECTED = 508, + HTTP_STATUS_BANDWIDTH_LIMIT_EXCEEDED = 509, + HTTP_STATUS_NOT_EXTENDED = 510, + HTTP_STATUS_NETWORK_AUTHENTICATION_REQUIRED = 511, + HTTP_STATUS_WEB_SERVER_UNKNOWN_ERROR = 520, + HTTP_STATUS_WEB_SERVER_IS_DOWN = 521, + HTTP_STATUS_CONNECTION_TIMEOUT = 522, + HTTP_STATUS_ORIGIN_IS_UNREACHABLE = 523, + HTTP_STATUS_TIMEOUT_OCCURED = 524, + HTTP_STATUS_SSL_HANDSHAKE_FAILED = 525, + HTTP_STATUS_INVALID_SSL_CERTIFICATE = 526, + HTTP_STATUS_RAILGUN_ERROR = 527, + HTTP_STATUS_SITE_IS_OVERLOADED = 529, + HTTP_STATUS_SITE_IS_FROZEN = 530, + HTTP_STATUS_IDENTITY_PROVIDER_AUTHENTICATION_ERROR = 561, + HTTP_STATUS_NETWORK_READ_TIMEOUT = 598, + HTTP_STATUS_NETWORK_CONNECT_TIMEOUT = 599 +}; +typedef enum llhttp_status llhttp_status_t; + +#define HTTP_ERRNO_MAP(XX) \ + XX(0, OK, OK) \ + XX(1, INTERNAL, INTERNAL) \ + XX(2, STRICT, STRICT) \ + XX(25, CR_EXPECTED, CR_EXPECTED) \ + XX(3, LF_EXPECTED, LF_EXPECTED) \ + XX(4, UNEXPECTED_CONTENT_LENGTH, UNEXPECTED_CONTENT_LENGTH) \ + XX(30, UNEXPECTED_SPACE, UNEXPECTED_SPACE) \ + XX(5, CLOSED_CONNECTION, CLOSED_CONNECTION) \ + XX(6, INVALID_METHOD, INVALID_METHOD) \ + XX(7, INVALID_URL, INVALID_URL) \ + XX(8, INVALID_CONSTANT, INVALID_CONSTANT) \ + XX(9, INVALID_VERSION, INVALID_VERSION) \ + XX(10, INVALID_HEADER_TOKEN, INVALID_HEADER_TOKEN) \ + XX(11, INVALID_CONTENT_LENGTH, INVALID_CONTENT_LENGTH) \ + XX(12, INVALID_CHUNK_SIZE, INVALID_CHUNK_SIZE) \ + XX(13, INVALID_STATUS, INVALID_STATUS) \ + XX(14, INVALID_EOF_STATE, INVALID_EOF_STATE) \ + XX(15, INVALID_TRANSFER_ENCODING, INVALID_TRANSFER_ENCODING) \ + XX(16, CB_MESSAGE_BEGIN, CB_MESSAGE_BEGIN) \ + XX(17, CB_HEADERS_COMPLETE, CB_HEADERS_COMPLETE) \ + XX(18, CB_MESSAGE_COMPLETE, CB_MESSAGE_COMPLETE) \ + XX(19, CB_CHUNK_HEADER, CB_CHUNK_HEADER) \ + XX(20, CB_CHUNK_COMPLETE, CB_CHUNK_COMPLETE) \ + XX(21, PAUSED, PAUSED) \ + XX(22, PAUSED_UPGRADE, PAUSED_UPGRADE) \ + XX(23, PAUSED_H2_UPGRADE, PAUSED_H2_UPGRADE) \ + XX(24, USER, USER) \ + XX(26, CB_URL_COMPLETE, CB_URL_COMPLETE) \ + XX(27, CB_STATUS_COMPLETE, CB_STATUS_COMPLETE) \ + XX(32, CB_METHOD_COMPLETE, CB_METHOD_COMPLETE) \ + XX(33, CB_VERSION_COMPLETE, CB_VERSION_COMPLETE) \ + XX(28, CB_HEADER_FIELD_COMPLETE, CB_HEADER_FIELD_COMPLETE) \ + XX(29, CB_HEADER_VALUE_COMPLETE, CB_HEADER_VALUE_COMPLETE) \ + XX(34, CB_CHUNK_EXTENSION_NAME_COMPLETE, CB_CHUNK_EXTENSION_NAME_COMPLETE) \ + XX(35, CB_CHUNK_EXTENSION_VALUE_COMPLETE, CB_CHUNK_EXTENSION_VALUE_COMPLETE) \ + XX(31, CB_RESET, CB_RESET) \ + + +#define HTTP_METHOD_MAP(XX) \ + XX(0, DELETE, DELETE) \ + XX(1, GET, GET) \ + XX(2, HEAD, HEAD) \ + XX(3, POST, POST) \ + XX(4, PUT, PUT) \ + XX(5, CONNECT, CONNECT) \ + XX(6, OPTIONS, OPTIONS) \ + XX(7, TRACE, TRACE) \ + XX(8, COPY, COPY) \ + XX(9, LOCK, LOCK) \ + XX(10, MKCOL, MKCOL) \ + XX(11, MOVE, MOVE) \ + XX(12, PROPFIND, PROPFIND) \ + XX(13, PROPPATCH, PROPPATCH) \ + XX(14, SEARCH, SEARCH) \ + XX(15, UNLOCK, UNLOCK) \ + XX(16, BIND, BIND) \ + XX(17, REBIND, REBIND) \ + XX(18, UNBIND, UNBIND) \ + XX(19, ACL, ACL) \ + XX(20, REPORT, REPORT) \ + XX(21, MKACTIVITY, MKACTIVITY) \ + XX(22, CHECKOUT, CHECKOUT) \ + XX(23, MERGE, MERGE) \ + XX(24, MSEARCH, M-SEARCH) \ + XX(25, NOTIFY, NOTIFY) \ + XX(26, SUBSCRIBE, SUBSCRIBE) \ + XX(27, UNSUBSCRIBE, UNSUBSCRIBE) \ + XX(28, PATCH, PATCH) \ + XX(29, PURGE, PURGE) \ + XX(30, MKCALENDAR, MKCALENDAR) \ + XX(31, LINK, LINK) \ + XX(32, UNLINK, UNLINK) \ + XX(33, SOURCE, SOURCE) \ + + +#define RTSP_METHOD_MAP(XX) \ + XX(1, GET, GET) \ + XX(3, POST, POST) \ + XX(6, OPTIONS, OPTIONS) \ + XX(35, DESCRIBE, DESCRIBE) \ + XX(36, ANNOUNCE, ANNOUNCE) \ + XX(37, SETUP, SETUP) \ + XX(38, PLAY, PLAY) \ + XX(39, PAUSE, PAUSE) \ + XX(40, TEARDOWN, TEARDOWN) \ + XX(41, GET_PARAMETER, GET_PARAMETER) \ + XX(42, SET_PARAMETER, SET_PARAMETER) \ + XX(43, REDIRECT, REDIRECT) \ + XX(44, RECORD, RECORD) \ + XX(45, FLUSH, FLUSH) \ + + +#define HTTP_ALL_METHOD_MAP(XX) \ + XX(0, DELETE, DELETE) \ + XX(1, GET, GET) \ + XX(2, HEAD, HEAD) \ + XX(3, POST, POST) \ + XX(4, PUT, PUT) \ + XX(5, CONNECT, CONNECT) \ + XX(6, OPTIONS, OPTIONS) \ + XX(7, TRACE, TRACE) \ + XX(8, COPY, COPY) \ + XX(9, LOCK, LOCK) \ + XX(10, MKCOL, MKCOL) \ + XX(11, MOVE, MOVE) \ + XX(12, PROPFIND, PROPFIND) \ + XX(13, PROPPATCH, PROPPATCH) \ + XX(14, SEARCH, SEARCH) \ + XX(15, UNLOCK, UNLOCK) \ + XX(16, BIND, BIND) \ + XX(17, REBIND, REBIND) \ + XX(18, UNBIND, UNBIND) \ + XX(19, ACL, ACL) \ + XX(20, REPORT, REPORT) \ + XX(21, MKACTIVITY, MKACTIVITY) \ + XX(22, CHECKOUT, CHECKOUT) \ + XX(23, MERGE, MERGE) \ + XX(24, MSEARCH, M-SEARCH) \ + XX(25, NOTIFY, NOTIFY) \ + XX(26, SUBSCRIBE, SUBSCRIBE) \ + XX(27, UNSUBSCRIBE, UNSUBSCRIBE) \ + XX(28, PATCH, PATCH) \ + XX(29, PURGE, PURGE) \ + XX(30, MKCALENDAR, MKCALENDAR) \ + XX(31, LINK, LINK) \ + XX(32, UNLINK, UNLINK) \ + XX(33, SOURCE, SOURCE) \ + XX(34, PRI, PRI) \ + XX(35, DESCRIBE, DESCRIBE) \ + XX(36, ANNOUNCE, ANNOUNCE) \ + XX(37, SETUP, SETUP) \ + XX(38, PLAY, PLAY) \ + XX(39, PAUSE, PAUSE) \ + XX(40, TEARDOWN, TEARDOWN) \ + XX(41, GET_PARAMETER, GET_PARAMETER) \ + XX(42, SET_PARAMETER, SET_PARAMETER) \ + XX(43, REDIRECT, REDIRECT) \ + XX(44, RECORD, RECORD) \ + XX(45, FLUSH, FLUSH) \ + + +#define HTTP_STATUS_MAP(XX) \ + XX(100, CONTINUE, CONTINUE) \ + XX(101, SWITCHING_PROTOCOLS, SWITCHING_PROTOCOLS) \ + XX(102, PROCESSING, PROCESSING) \ + XX(103, EARLY_HINTS, EARLY_HINTS) \ + XX(110, RESPONSE_IS_STALE, RESPONSE_IS_STALE) \ + XX(111, REVALIDATION_FAILED, REVALIDATION_FAILED) \ + XX(112, DISCONNECTED_OPERATION, DISCONNECTED_OPERATION) \ + XX(113, HEURISTIC_EXPIRATION, HEURISTIC_EXPIRATION) \ + XX(199, MISCELLANEOUS_WARNING, MISCELLANEOUS_WARNING) \ + XX(200, OK, OK) \ + XX(201, CREATED, CREATED) \ + XX(202, ACCEPTED, ACCEPTED) \ + XX(203, NON_AUTHORITATIVE_INFORMATION, NON_AUTHORITATIVE_INFORMATION) \ + XX(204, NO_CONTENT, NO_CONTENT) \ + XX(205, RESET_CONTENT, RESET_CONTENT) \ + XX(206, PARTIAL_CONTENT, PARTIAL_CONTENT) \ + XX(207, MULTI_STATUS, MULTI_STATUS) \ + XX(208, ALREADY_REPORTED, ALREADY_REPORTED) \ + XX(214, TRANSFORMATION_APPLIED, TRANSFORMATION_APPLIED) \ + XX(226, IM_USED, IM_USED) \ + XX(299, MISCELLANEOUS_PERSISTENT_WARNING, MISCELLANEOUS_PERSISTENT_WARNING) \ + XX(300, MULTIPLE_CHOICES, MULTIPLE_CHOICES) \ + XX(301, MOVED_PERMANENTLY, MOVED_PERMANENTLY) \ + XX(302, FOUND, FOUND) \ + XX(303, SEE_OTHER, SEE_OTHER) \ + XX(304, NOT_MODIFIED, NOT_MODIFIED) \ + XX(305, USE_PROXY, USE_PROXY) \ + XX(306, SWITCH_PROXY, SWITCH_PROXY) \ + XX(307, TEMPORARY_REDIRECT, TEMPORARY_REDIRECT) \ + XX(308, PERMANENT_REDIRECT, PERMANENT_REDIRECT) \ + XX(400, BAD_REQUEST, BAD_REQUEST) \ + XX(401, UNAUTHORIZED, UNAUTHORIZED) \ + XX(402, PAYMENT_REQUIRED, PAYMENT_REQUIRED) \ + XX(403, FORBIDDEN, FORBIDDEN) \ + XX(404, NOT_FOUND, NOT_FOUND) \ + XX(405, METHOD_NOT_ALLOWED, METHOD_NOT_ALLOWED) \ + XX(406, NOT_ACCEPTABLE, NOT_ACCEPTABLE) \ + XX(407, PROXY_AUTHENTICATION_REQUIRED, PROXY_AUTHENTICATION_REQUIRED) \ + XX(408, REQUEST_TIMEOUT, REQUEST_TIMEOUT) \ + XX(409, CONFLICT, CONFLICT) \ + XX(410, GONE, GONE) \ + XX(411, LENGTH_REQUIRED, LENGTH_REQUIRED) \ + XX(412, PRECONDITION_FAILED, PRECONDITION_FAILED) \ + XX(413, PAYLOAD_TOO_LARGE, PAYLOAD_TOO_LARGE) \ + XX(414, URI_TOO_LONG, URI_TOO_LONG) \ + XX(415, UNSUPPORTED_MEDIA_TYPE, UNSUPPORTED_MEDIA_TYPE) \ + XX(416, RANGE_NOT_SATISFIABLE, RANGE_NOT_SATISFIABLE) \ + XX(417, EXPECTATION_FAILED, EXPECTATION_FAILED) \ + XX(418, IM_A_TEAPOT, IM_A_TEAPOT) \ + XX(419, PAGE_EXPIRED, PAGE_EXPIRED) \ + XX(420, ENHANCE_YOUR_CALM, ENHANCE_YOUR_CALM) \ + XX(421, MISDIRECTED_REQUEST, MISDIRECTED_REQUEST) \ + XX(422, UNPROCESSABLE_ENTITY, UNPROCESSABLE_ENTITY) \ + XX(423, LOCKED, LOCKED) \ + XX(424, FAILED_DEPENDENCY, FAILED_DEPENDENCY) \ + XX(425, TOO_EARLY, TOO_EARLY) \ + XX(426, UPGRADE_REQUIRED, UPGRADE_REQUIRED) \ + XX(428, PRECONDITION_REQUIRED, PRECONDITION_REQUIRED) \ + XX(429, TOO_MANY_REQUESTS, TOO_MANY_REQUESTS) \ + XX(430, REQUEST_HEADER_FIELDS_TOO_LARGE_UNOFFICIAL, REQUEST_HEADER_FIELDS_TOO_LARGE_UNOFFICIAL) \ + XX(431, REQUEST_HEADER_FIELDS_TOO_LARGE, REQUEST_HEADER_FIELDS_TOO_LARGE) \ + XX(440, LOGIN_TIMEOUT, LOGIN_TIMEOUT) \ + XX(444, NO_RESPONSE, NO_RESPONSE) \ + XX(449, RETRY_WITH, RETRY_WITH) \ + XX(450, BLOCKED_BY_PARENTAL_CONTROL, BLOCKED_BY_PARENTAL_CONTROL) \ + XX(451, UNAVAILABLE_FOR_LEGAL_REASONS, UNAVAILABLE_FOR_LEGAL_REASONS) \ + XX(460, CLIENT_CLOSED_LOAD_BALANCED_REQUEST, CLIENT_CLOSED_LOAD_BALANCED_REQUEST) \ + XX(463, INVALID_X_FORWARDED_FOR, INVALID_X_FORWARDED_FOR) \ + XX(494, REQUEST_HEADER_TOO_LARGE, REQUEST_HEADER_TOO_LARGE) \ + XX(495, SSL_CERTIFICATE_ERROR, SSL_CERTIFICATE_ERROR) \ + XX(496, SSL_CERTIFICATE_REQUIRED, SSL_CERTIFICATE_REQUIRED) \ + XX(497, HTTP_REQUEST_SENT_TO_HTTPS_PORT, HTTP_REQUEST_SENT_TO_HTTPS_PORT) \ + XX(498, INVALID_TOKEN, INVALID_TOKEN) \ + XX(499, CLIENT_CLOSED_REQUEST, CLIENT_CLOSED_REQUEST) \ + XX(500, INTERNAL_SERVER_ERROR, INTERNAL_SERVER_ERROR) \ + XX(501, NOT_IMPLEMENTED, NOT_IMPLEMENTED) \ + XX(502, BAD_GATEWAY, BAD_GATEWAY) \ + XX(503, SERVICE_UNAVAILABLE, SERVICE_UNAVAILABLE) \ + XX(504, GATEWAY_TIMEOUT, GATEWAY_TIMEOUT) \ + XX(505, HTTP_VERSION_NOT_SUPPORTED, HTTP_VERSION_NOT_SUPPORTED) \ + XX(506, VARIANT_ALSO_NEGOTIATES, VARIANT_ALSO_NEGOTIATES) \ + XX(507, INSUFFICIENT_STORAGE, INSUFFICIENT_STORAGE) \ + XX(508, LOOP_DETECTED, LOOP_DETECTED) \ + XX(509, BANDWIDTH_LIMIT_EXCEEDED, BANDWIDTH_LIMIT_EXCEEDED) \ + XX(510, NOT_EXTENDED, NOT_EXTENDED) \ + XX(511, NETWORK_AUTHENTICATION_REQUIRED, NETWORK_AUTHENTICATION_REQUIRED) \ + XX(520, WEB_SERVER_UNKNOWN_ERROR, WEB_SERVER_UNKNOWN_ERROR) \ + XX(521, WEB_SERVER_IS_DOWN, WEB_SERVER_IS_DOWN) \ + XX(522, CONNECTION_TIMEOUT, CONNECTION_TIMEOUT) \ + XX(523, ORIGIN_IS_UNREACHABLE, ORIGIN_IS_UNREACHABLE) \ + XX(524, TIMEOUT_OCCURED, TIMEOUT_OCCURED) \ + XX(525, SSL_HANDSHAKE_FAILED, SSL_HANDSHAKE_FAILED) \ + XX(526, INVALID_SSL_CERTIFICATE, INVALID_SSL_CERTIFICATE) \ + XX(527, RAILGUN_ERROR, RAILGUN_ERROR) \ + XX(529, SITE_IS_OVERLOADED, SITE_IS_OVERLOADED) \ + XX(530, SITE_IS_FROZEN, SITE_IS_FROZEN) \ + XX(561, IDENTITY_PROVIDER_AUTHENTICATION_ERROR, IDENTITY_PROVIDER_AUTHENTICATION_ERROR) \ + XX(598, NETWORK_READ_TIMEOUT, NETWORK_READ_TIMEOUT) \ + XX(599, NETWORK_CONNECT_TIMEOUT, NETWORK_CONNECT_TIMEOUT) \ + + +#ifdef __cplusplus +} /* extern "C" */ +#endif +#endif /* LLLLHTTP_C_HEADERS_ */ + + +#ifndef INCLUDE_LLHTTP_API_H_ +#define INCLUDE_LLHTTP_API_H_ +#ifdef __cplusplus +extern "C" { +#endif +#include <stddef.h> + +#if defined(__wasm__) +#define LLHTTP_EXPORT __attribute__((visibility("default"))) +#else +#define LLHTTP_EXPORT +#endif + +typedef llhttp__internal_t llhttp_t; +typedef struct llhttp_settings_s llhttp_settings_t; + +typedef int (*llhttp_data_cb)(llhttp_t*, const char *at, size_t length); +typedef int (*llhttp_cb)(llhttp_t*); + +struct llhttp_settings_s { + /* Possible return values 0, -1, `HPE_PAUSED` */ + llhttp_cb on_message_begin; + + /* Possible return values 0, -1, HPE_USER */ + llhttp_data_cb on_url; + llhttp_data_cb on_status; + llhttp_data_cb on_method; + llhttp_data_cb on_version; + llhttp_data_cb on_header_field; + llhttp_data_cb on_header_value; + llhttp_data_cb on_chunk_extension_name; + llhttp_data_cb on_chunk_extension_value; + + /* Possible return values: + * 0 - Proceed normally + * 1 - Assume that request/response has no body, and proceed to parsing the + * next message + * 2 - Assume absence of body (as above) and make `llhttp_execute()` return + * `HPE_PAUSED_UPGRADE` + * -1 - Error + * `HPE_PAUSED` + */ + llhttp_cb on_headers_complete; + + /* Possible return values 0, -1, HPE_USER */ + llhttp_data_cb on_body; + + /* Possible return values 0, -1, `HPE_PAUSED` */ + llhttp_cb on_message_complete; + llhttp_cb on_url_complete; + llhttp_cb on_status_complete; + llhttp_cb on_method_complete; + llhttp_cb on_version_complete; + llhttp_cb on_header_field_complete; + llhttp_cb on_header_value_complete; + llhttp_cb on_chunk_extension_name_complete; + llhttp_cb on_chunk_extension_value_complete; + + /* When on_chunk_header is called, the current chunk length is stored + * in parser->content_length. + * Possible return values 0, -1, `HPE_PAUSED` + */ + llhttp_cb on_chunk_header; + llhttp_cb on_chunk_complete; + llhttp_cb on_reset; +}; + +/* Initialize the parser with specific type and user settings. + * + * NOTE: lifetime of `settings` has to be at least the same as the lifetime of + * the `parser` here. In practice, `settings` has to be either a static + * variable or be allocated with `malloc`, `new`, etc. + */ +LLHTTP_EXPORT +void llhttp_init(llhttp_t* parser, llhttp_type_t type, + const llhttp_settings_t* settings); + +LLHTTP_EXPORT +llhttp_t* llhttp_alloc(llhttp_type_t type); + +LLHTTP_EXPORT +void llhttp_free(llhttp_t* parser); + +LLHTTP_EXPORT +uint8_t llhttp_get_type(llhttp_t* parser); + +LLHTTP_EXPORT +uint8_t llhttp_get_http_major(llhttp_t* parser); + +LLHTTP_EXPORT +uint8_t llhttp_get_http_minor(llhttp_t* parser); + +LLHTTP_EXPORT +uint8_t llhttp_get_method(llhttp_t* parser); + +LLHTTP_EXPORT +int llhttp_get_status_code(llhttp_t* parser); + +LLHTTP_EXPORT +uint8_t llhttp_get_upgrade(llhttp_t* parser); + +/* Reset an already initialized parser back to the start state, preserving the + * existing parser type, callback settings, user data, and lenient flags. + */ +LLHTTP_EXPORT +void llhttp_reset(llhttp_t* parser); + +/* Initialize the settings object */ +LLHTTP_EXPORT +void llhttp_settings_init(llhttp_settings_t* settings); + +/* Parse full or partial request/response, invoking user callbacks along the + * way. + * + * If any of `llhttp_data_cb` returns errno not equal to `HPE_OK` - the parsing + * interrupts, and such errno is returned from `llhttp_execute()`. If + * `HPE_PAUSED` was used as a errno, the execution can be resumed with + * `llhttp_resume()` call. + * + * In a special case of CONNECT/Upgrade request/response `HPE_PAUSED_UPGRADE` + * is returned after fully parsing the request/response. If the user wishes to + * continue parsing, they need to invoke `llhttp_resume_after_upgrade()`. + * + * NOTE: if this function ever returns a non-pause type error, it will continue + * to return the same error upon each successive call up until `llhttp_init()` + * is called. + */ +LLHTTP_EXPORT +llhttp_errno_t llhttp_execute(llhttp_t* parser, const char* data, size_t len); + +/* This method should be called when the other side has no further bytes to + * send (e.g. shutdown of readable side of the TCP connection.) + * + * Requests without `Content-Length` and other messages might require treating + * all incoming bytes as the part of the body, up to the last byte of the + * connection. This method will invoke `on_message_complete()` callback if the + * request was terminated safely. Otherwise a error code would be returned. + */ +LLHTTP_EXPORT +llhttp_errno_t llhttp_finish(llhttp_t* parser); + +/* Returns `1` if the incoming message is parsed until the last byte, and has + * to be completed by calling `llhttp_finish()` on EOF + */ +LLHTTP_EXPORT +int llhttp_message_needs_eof(const llhttp_t* parser); + +/* Returns `1` if there might be any other messages following the last that was + * successfully parsed. + */ +LLHTTP_EXPORT +int llhttp_should_keep_alive(const llhttp_t* parser); + +/* Make further calls of `llhttp_execute()` return `HPE_PAUSED` and set + * appropriate error reason. + * + * Important: do not call this from user callbacks! User callbacks must return + * `HPE_PAUSED` if pausing is required. + */ +LLHTTP_EXPORT +void llhttp_pause(llhttp_t* parser); + +/* Might be called to resume the execution after the pause in user's callback. + * See `llhttp_execute()` above for details. + * + * Call this only if `llhttp_execute()` returns `HPE_PAUSED`. + */ +LLHTTP_EXPORT +void llhttp_resume(llhttp_t* parser); + +/* Might be called to resume the execution after the pause in user's callback. + * See `llhttp_execute()` above for details. + * + * Call this only if `llhttp_execute()` returns `HPE_PAUSED_UPGRADE` + */ +LLHTTP_EXPORT +void llhttp_resume_after_upgrade(llhttp_t* parser); + +/* Returns the latest return error */ +LLHTTP_EXPORT +llhttp_errno_t llhttp_get_errno(const llhttp_t* parser); + +/* Returns the verbal explanation of the latest returned error. + * + * Note: User callback should set error reason when returning the error. See + * `llhttp_set_error_reason()` for details. + */ +LLHTTP_EXPORT +const char* llhttp_get_error_reason(const llhttp_t* parser); + +/* Assign verbal description to the returned error. Must be called in user + * callbacks right before returning the errno. + * + * Note: `HPE_USER` error code might be useful in user callbacks. + */ +LLHTTP_EXPORT +void llhttp_set_error_reason(llhttp_t* parser, const char* reason); + +/* Returns the pointer to the last parsed byte before the returned error. The + * pointer is relative to the `data` argument of `llhttp_execute()`. + * + * Note: this method might be useful for counting the number of parsed bytes. + */ +LLHTTP_EXPORT +const char* llhttp_get_error_pos(const llhttp_t* parser); + +/* Returns textual name of error code */ +LLHTTP_EXPORT +const char* llhttp_errno_name(llhttp_errno_t err); + +/* Returns textual name of HTTP method */ +LLHTTP_EXPORT +const char* llhttp_method_name(llhttp_method_t method); + +/* Returns textual name of HTTP status */ +LLHTTP_EXPORT +const char* llhttp_status_name(llhttp_status_t status); + +/* Enables/disables lenient header value parsing (disabled by default). + * + * Lenient parsing disables header value token checks, extending llhttp's + * protocol support to highly non-compliant clients/server. No + * `HPE_INVALID_HEADER_TOKEN` will be raised for incorrect header values when + * lenient parsing is "on". + * + * **Enabling this flag can pose a security issue since you will be exposed to + * request smuggling attacks. USE WITH CAUTION!** + */ +LLHTTP_EXPORT +void llhttp_set_lenient_headers(llhttp_t* parser, int enabled); + + +/* Enables/disables lenient handling of conflicting `Transfer-Encoding` and + * `Content-Length` headers (disabled by default). + * + * Normally `llhttp` would error when `Transfer-Encoding` is present in + * conjunction with `Content-Length`. This error is important to prevent HTTP + * request smuggling, but may be less desirable for small number of cases + * involving legacy servers. + * + * **Enabling this flag can pose a security issue since you will be exposed to + * request smuggling attacks. USE WITH CAUTION!** + */ +LLHTTP_EXPORT +void llhttp_set_lenient_chunked_length(llhttp_t* parser, int enabled); + + +/* Enables/disables lenient handling of `Connection: close` and HTTP/1.0 + * requests responses. + * + * Normally `llhttp` would error on (in strict mode) or discard (in loose mode) + * the HTTP request/response after the request/response with `Connection: close` + * and `Content-Length`. This is important to prevent cache poisoning attacks, + * but might interact badly with outdated and insecure clients. With this flag + * the extra request/response will be parsed normally. + * + * **Enabling this flag can pose a security issue since you will be exposed to + * poisoning attacks. USE WITH CAUTION!** + */ +LLHTTP_EXPORT +void llhttp_set_lenient_keep_alive(llhttp_t* parser, int enabled); + +/* Enables/disables lenient handling of `Transfer-Encoding` header. + * + * Normally `llhttp` would error when a `Transfer-Encoding` has `chunked` value + * and another value after it (either in a single header or in multiple + * headers whose value are internally joined using `, `). + * This is mandated by the spec to reliably determine request body size and thus + * avoid request smuggling. + * With this flag the extra value will be parsed normally. + * + * **Enabling this flag can pose a security issue since you will be exposed to + * request smuggling attacks. USE WITH CAUTION!** + */ +LLHTTP_EXPORT +void llhttp_set_lenient_transfer_encoding(llhttp_t* parser, int enabled); + +/* Enables/disables lenient handling of HTTP version. + * + * Normally `llhttp` would error when the HTTP version in the request or status line + * is not `0.9`, `1.0`, `1.1` or `2.0`. + * With this flag the invalid value will be parsed normally. + * + * **Enabling this flag can pose a security issue since you will allow unsupported + * HTTP versions. USE WITH CAUTION!** + */ +LLHTTP_EXPORT +void llhttp_set_lenient_version(llhttp_t* parser, int enabled); + +/* Enables/disables lenient handling of additional data received after a message ends + * and keep-alive is disabled. + * + * Normally `llhttp` would error when additional unexpected data is received if the message + * contains the `Connection` header with `close` value. + * With this flag the extra data will discarded without throwing an error. + * + * **Enabling this flag can pose a security issue since you will be exposed to + * poisoning attacks. USE WITH CAUTION!** + */ +LLHTTP_EXPORT +void llhttp_set_lenient_data_after_close(llhttp_t* parser, int enabled); + +/* Enables/disables lenient handling of incomplete CRLF sequences. + * + * Normally `llhttp` would error when a CR is not followed by LF when terminating the + * request line, the status line, the headers or a chunk header. + * With this flag only a CR is required to terminate such sections. + * + * **Enabling this flag can pose a security issue since you will be exposed to + * request smuggling attacks. USE WITH CAUTION!** + */ +LLHTTP_EXPORT +void llhttp_set_lenient_optional_lf_after_cr(llhttp_t* parser, int enabled); + +/* Enables/disables lenient handling of chunks not separated via CRLF. + * + * Normally `llhttp` would error when after a chunk data a CRLF is missing before + * starting a new chunk. + * With this flag the new chunk can start immediately after the previous one. + * + * **Enabling this flag can pose a security issue since you will be exposed to + * request smuggling attacks. USE WITH CAUTION!** + */ +LLHTTP_EXPORT +void llhttp_set_lenient_optional_crlf_after_chunk(llhttp_t* parser, int enabled); + +#ifdef __cplusplus +} /* extern "C" */ +#endif +#endif /* INCLUDE_LLHTTP_API_H_ */ + + +#endif /* INCLUDE_LLHTTP_H_ */ diff --git a/contrib/restricted/llhttp/src/api.c b/contrib/restricted/llhttp/src/api.c new file mode 100644 index 0000000000..e0d0385051 --- /dev/null +++ b/contrib/restricted/llhttp/src/api.c @@ -0,0 +1,494 @@ +#include <stdlib.h> +#include <stdio.h> +#include <string.h> + +#include "llhttp.h" + +#define CALLBACK_MAYBE(PARSER, NAME) \ + do { \ + const llhttp_settings_t* settings; \ + settings = (const llhttp_settings_t*) (PARSER)->settings; \ + if (settings == NULL || settings->NAME == NULL) { \ + err = 0; \ + break; \ + } \ + err = settings->NAME((PARSER)); \ + } while (0) + +#define SPAN_CALLBACK_MAYBE(PARSER, NAME, START, LEN) \ + do { \ + const llhttp_settings_t* settings; \ + settings = (const llhttp_settings_t*) (PARSER)->settings; \ + if (settings == NULL || settings->NAME == NULL) { \ + err = 0; \ + break; \ + } \ + err = settings->NAME((PARSER), (START), (LEN)); \ + if (err == -1) { \ + err = HPE_USER; \ + llhttp_set_error_reason((PARSER), "Span callback error in " #NAME); \ + } \ + } while (0) + +void llhttp_init(llhttp_t* parser, llhttp_type_t type, + const llhttp_settings_t* settings) { + llhttp__internal_init(parser); + + parser->type = type; + parser->settings = (void*) settings; +} + + +#if defined(__wasm__) + +extern int wasm_on_message_begin(llhttp_t * p); +extern int wasm_on_url(llhttp_t* p, const char* at, size_t length); +extern int wasm_on_status(llhttp_t* p, const char* at, size_t length); +extern int wasm_on_header_field(llhttp_t* p, const char* at, size_t length); +extern int wasm_on_header_value(llhttp_t* p, const char* at, size_t length); +extern int wasm_on_headers_complete(llhttp_t * p, int status_code, + uint8_t upgrade, int should_keep_alive); +extern int wasm_on_body(llhttp_t* p, const char* at, size_t length); +extern int wasm_on_message_complete(llhttp_t * p); + +static int wasm_on_headers_complete_wrap(llhttp_t* p) { + return wasm_on_headers_complete(p, p->status_code, p->upgrade, + llhttp_should_keep_alive(p)); +} + +const llhttp_settings_t wasm_settings = { + wasm_on_message_begin, + wasm_on_url, + wasm_on_status, + NULL, + NULL, + wasm_on_header_field, + wasm_on_header_value, + NULL, + NULL, + wasm_on_headers_complete_wrap, + wasm_on_body, + wasm_on_message_complete, + NULL, + NULL, + NULL, + NULL, + NULL, + NULL, + NULL, + NULL, + NULL, + NULL, + NULL, +}; + + +llhttp_t* llhttp_alloc(llhttp_type_t type) { + llhttp_t* parser = malloc(sizeof(llhttp_t)); + llhttp_init(parser, type, &wasm_settings); + return parser; +} + +void llhttp_free(llhttp_t* parser) { + free(parser); +} + +#endif // defined(__wasm__) + +/* Some getters required to get stuff from the parser */ + +uint8_t llhttp_get_type(llhttp_t* parser) { + return parser->type; +} + +uint8_t llhttp_get_http_major(llhttp_t* parser) { + return parser->http_major; +} + +uint8_t llhttp_get_http_minor(llhttp_t* parser) { + return parser->http_minor; +} + +uint8_t llhttp_get_method(llhttp_t* parser) { + return parser->method; +} + +int llhttp_get_status_code(llhttp_t* parser) { + return parser->status_code; +} + +uint8_t llhttp_get_upgrade(llhttp_t* parser) { + return parser->upgrade; +} + + +void llhttp_reset(llhttp_t* parser) { + llhttp_type_t type = parser->type; + const llhttp_settings_t* settings = parser->settings; + void* data = parser->data; + uint8_t lenient_flags = parser->lenient_flags; + + llhttp__internal_init(parser); + + parser->type = type; + parser->settings = (void*) settings; + parser->data = data; + parser->lenient_flags = lenient_flags; +} + + +llhttp_errno_t llhttp_execute(llhttp_t* parser, const char* data, size_t len) { + return llhttp__internal_execute(parser, data, data + len); +} + + +void llhttp_settings_init(llhttp_settings_t* settings) { + memset(settings, 0, sizeof(*settings)); +} + + +llhttp_errno_t llhttp_finish(llhttp_t* parser) { + int err; + + /* We're in an error state. Don't bother doing anything. */ + if (parser->error != 0) { + return 0; + } + + switch (parser->finish) { + case HTTP_FINISH_SAFE_WITH_CB: + CALLBACK_MAYBE(parser, on_message_complete); + if (err != HPE_OK) return err; + + /* FALLTHROUGH */ + case HTTP_FINISH_SAFE: + return HPE_OK; + case HTTP_FINISH_UNSAFE: + parser->reason = "Invalid EOF state"; + return HPE_INVALID_EOF_STATE; + default: + abort(); + } +} + + +void llhttp_pause(llhttp_t* parser) { + if (parser->error != HPE_OK) { + return; + } + + parser->error = HPE_PAUSED; + parser->reason = "Paused"; +} + + +void llhttp_resume(llhttp_t* parser) { + if (parser->error != HPE_PAUSED) { + return; + } + + parser->error = 0; +} + + +void llhttp_resume_after_upgrade(llhttp_t* parser) { + if (parser->error != HPE_PAUSED_UPGRADE) { + return; + } + + parser->error = 0; +} + + +llhttp_errno_t llhttp_get_errno(const llhttp_t* parser) { + return parser->error; +} + + +const char* llhttp_get_error_reason(const llhttp_t* parser) { + return parser->reason; +} + + +void llhttp_set_error_reason(llhttp_t* parser, const char* reason) { + parser->reason = reason; +} + + +const char* llhttp_get_error_pos(const llhttp_t* parser) { + return parser->error_pos; +} + + +const char* llhttp_errno_name(llhttp_errno_t err) { +#define HTTP_ERRNO_GEN(CODE, NAME, _) case HPE_##NAME: return "HPE_" #NAME; + switch (err) { + HTTP_ERRNO_MAP(HTTP_ERRNO_GEN) + default: abort(); + } +#undef HTTP_ERRNO_GEN +} + + +const char* llhttp_method_name(llhttp_method_t method) { +#define HTTP_METHOD_GEN(NUM, NAME, STRING) case HTTP_##NAME: return #STRING; + switch (method) { + HTTP_ALL_METHOD_MAP(HTTP_METHOD_GEN) + default: abort(); + } +#undef HTTP_METHOD_GEN +} + +const char* llhttp_status_name(llhttp_status_t status) { +#define HTTP_STATUS_GEN(NUM, NAME, STRING) case HTTP_STATUS_##NAME: return #STRING; + switch (status) { + HTTP_STATUS_MAP(HTTP_STATUS_GEN) + default: abort(); + } +#undef HTTP_STATUS_GEN +} + + +void llhttp_set_lenient_headers(llhttp_t* parser, int enabled) { + if (enabled) { + parser->lenient_flags |= LENIENT_HEADERS; + } else { + parser->lenient_flags &= ~LENIENT_HEADERS; + } +} + + +void llhttp_set_lenient_chunked_length(llhttp_t* parser, int enabled) { + if (enabled) { + parser->lenient_flags |= LENIENT_CHUNKED_LENGTH; + } else { + parser->lenient_flags &= ~LENIENT_CHUNKED_LENGTH; + } +} + + +void llhttp_set_lenient_keep_alive(llhttp_t* parser, int enabled) { + if (enabled) { + parser->lenient_flags |= LENIENT_KEEP_ALIVE; + } else { + parser->lenient_flags &= ~LENIENT_KEEP_ALIVE; + } +} + +void llhttp_set_lenient_transfer_encoding(llhttp_t* parser, int enabled) { + if (enabled) { + parser->lenient_flags |= LENIENT_TRANSFER_ENCODING; + } else { + parser->lenient_flags &= ~LENIENT_TRANSFER_ENCODING; + } +} + +void llhttp_set_lenient_version(llhttp_t* parser, int enabled) { + if (enabled) { + parser->lenient_flags |= LENIENT_VERSION; + } else { + parser->lenient_flags &= ~LENIENT_VERSION; + } +} + +void llhttp_set_lenient_data_after_close(llhttp_t* parser, int enabled) { + if (enabled) { + parser->lenient_flags |= LENIENT_DATA_AFTER_CLOSE; + } else { + parser->lenient_flags &= ~LENIENT_DATA_AFTER_CLOSE; + } +} + +void llhttp_set_lenient_optional_lf_after_cr(llhttp_t* parser, int enabled) { + if (enabled) { + parser->lenient_flags |= LENIENT_OPTIONAL_LF_AFTER_CR; + } else { + parser->lenient_flags &= ~LENIENT_OPTIONAL_LF_AFTER_CR; + } +} + +void llhttp_set_lenient_optional_crlf_after_chunk(llhttp_t* parser, int enabled) { + if (enabled) { + parser->lenient_flags |= LENIENT_OPTIONAL_CRLF_AFTER_CHUNK; + } else { + parser->lenient_flags &= ~LENIENT_OPTIONAL_CRLF_AFTER_CHUNK; + } +} + +/* Callbacks */ + + +int llhttp__on_message_begin(llhttp_t* s, const char* p, const char* endp) { + int err; + CALLBACK_MAYBE(s, on_message_begin); + return err; +} + + +int llhttp__on_url(llhttp_t* s, const char* p, const char* endp) { + int err; + SPAN_CALLBACK_MAYBE(s, on_url, p, endp - p); + return err; +} + + +int llhttp__on_url_complete(llhttp_t* s, const char* p, const char* endp) { + int err; + CALLBACK_MAYBE(s, on_url_complete); + return err; +} + + +int llhttp__on_status(llhttp_t* s, const char* p, const char* endp) { + int err; + SPAN_CALLBACK_MAYBE(s, on_status, p, endp - p); + return err; +} + + +int llhttp__on_status_complete(llhttp_t* s, const char* p, const char* endp) { + int err; + CALLBACK_MAYBE(s, on_status_complete); + return err; +} + + +int llhttp__on_method(llhttp_t* s, const char* p, const char* endp) { + int err; + SPAN_CALLBACK_MAYBE(s, on_method, p, endp - p); + return err; +} + + +int llhttp__on_method_complete(llhttp_t* s, const char* p, const char* endp) { + int err; + CALLBACK_MAYBE(s, on_method_complete); + return err; +} + + +int llhttp__on_version(llhttp_t* s, const char* p, const char* endp) { + int err; + SPAN_CALLBACK_MAYBE(s, on_version, p, endp - p); + return err; +} + + +int llhttp__on_version_complete(llhttp_t* s, const char* p, const char* endp) { + int err; + CALLBACK_MAYBE(s, on_version_complete); + return err; +} + + +int llhttp__on_header_field(llhttp_t* s, const char* p, const char* endp) { + int err; + SPAN_CALLBACK_MAYBE(s, on_header_field, p, endp - p); + return err; +} + + +int llhttp__on_header_field_complete(llhttp_t* s, const char* p, const char* endp) { + int err; + CALLBACK_MAYBE(s, on_header_field_complete); + return err; +} + + +int llhttp__on_header_value(llhttp_t* s, const char* p, const char* endp) { + int err; + SPAN_CALLBACK_MAYBE(s, on_header_value, p, endp - p); + return err; +} + + +int llhttp__on_header_value_complete(llhttp_t* s, const char* p, const char* endp) { + int err; + CALLBACK_MAYBE(s, on_header_value_complete); + return err; +} + + +int llhttp__on_headers_complete(llhttp_t* s, const char* p, const char* endp) { + int err; + CALLBACK_MAYBE(s, on_headers_complete); + return err; +} + + +int llhttp__on_message_complete(llhttp_t* s, const char* p, const char* endp) { + int err; + CALLBACK_MAYBE(s, on_message_complete); + return err; +} + + +int llhttp__on_body(llhttp_t* s, const char* p, const char* endp) { + int err; + SPAN_CALLBACK_MAYBE(s, on_body, p, endp - p); + return err; +} + + +int llhttp__on_chunk_header(llhttp_t* s, const char* p, const char* endp) { + int err; + CALLBACK_MAYBE(s, on_chunk_header); + return err; +} + + +int llhttp__on_chunk_extension_name(llhttp_t* s, const char* p, const char* endp) { + int err; + SPAN_CALLBACK_MAYBE(s, on_chunk_extension_name, p, endp - p); + return err; +} + + +int llhttp__on_chunk_extension_name_complete(llhttp_t* s, const char* p, const char* endp) { + int err; + CALLBACK_MAYBE(s, on_chunk_extension_name_complete); + return err; +} + + +int llhttp__on_chunk_extension_value(llhttp_t* s, const char* p, const char* endp) { + int err; + SPAN_CALLBACK_MAYBE(s, on_chunk_extension_value, p, endp - p); + return err; +} + + +int llhttp__on_chunk_extension_value_complete(llhttp_t* s, const char* p, const char* endp) { + int err; + CALLBACK_MAYBE(s, on_chunk_extension_value_complete); + return err; +} + + +int llhttp__on_chunk_complete(llhttp_t* s, const char* p, const char* endp) { + int err; + CALLBACK_MAYBE(s, on_chunk_complete); + return err; +} + + +int llhttp__on_reset(llhttp_t* s, const char* p, const char* endp) { + int err; + CALLBACK_MAYBE(s, on_reset); + return err; +} + + +/* Private */ + + +void llhttp__debug(llhttp_t* s, const char* p, const char* endp, + const char* msg) { + if (p == endp) { + fprintf(stderr, "p=%p type=%d flags=%02x next=null debug=%s\n", s, s->type, + s->flags, msg); + } else { + fprintf(stderr, "p=%p type=%d flags=%02x next=%02x debug=%s\n", s, + s->type, s->flags, *p, msg); + } +} diff --git a/contrib/restricted/llhttp/src/http.c b/contrib/restricted/llhttp/src/http.c new file mode 100644 index 0000000000..3a66044f5f --- /dev/null +++ b/contrib/restricted/llhttp/src/http.c @@ -0,0 +1,150 @@ +#include <stdio.h> +#ifndef LLHTTP__TEST +# include "llhttp.h" +#else +# define llhttp_t llparse_t +#endif /* */ + +int llhttp_message_needs_eof(const llhttp_t* parser); +int llhttp_should_keep_alive(const llhttp_t* parser); + +int llhttp__before_headers_complete(llhttp_t* parser, const char* p, + const char* endp) { + /* Set this here so that on_headers_complete() callbacks can see it */ + if ((parser->flags & F_UPGRADE) && + (parser->flags & F_CONNECTION_UPGRADE)) { + /* For responses, "Upgrade: foo" and "Connection: upgrade" are + * mandatory only when it is a 101 Switching Protocols response, + * otherwise it is purely informational, to announce support. + */ + parser->upgrade = + (parser->type == HTTP_REQUEST || parser->status_code == 101); + } else { + parser->upgrade = (parser->method == HTTP_CONNECT); + } + return 0; +} + + +/* Return values: + * 0 - No body, `restart`, message_complete + * 1 - CONNECT request, `restart`, message_complete, and pause + * 2 - chunk_size_start + * 3 - body_identity + * 4 - body_identity_eof + * 5 - invalid transfer-encoding for request + */ +int llhttp__after_headers_complete(llhttp_t* parser, const char* p, + const char* endp) { + int hasBody; + + hasBody = parser->flags & F_CHUNKED || parser->content_length > 0; + if (parser->upgrade && (parser->method == HTTP_CONNECT || + (parser->flags & F_SKIPBODY) || !hasBody)) { + /* Exit, the rest of the message is in a different protocol. */ + return 1; + } + + if (parser->flags & F_SKIPBODY) { + return 0; + } else if (parser->flags & F_CHUNKED) { + /* chunked encoding - ignore Content-Length header, prepare for a chunk */ + return 2; + } else if (parser->flags & F_TRANSFER_ENCODING) { + if (parser->type == HTTP_REQUEST && + (parser->lenient_flags & LENIENT_CHUNKED_LENGTH) == 0 && + (parser->lenient_flags & LENIENT_TRANSFER_ENCODING) == 0) { + /* RFC 7230 3.3.3 */ + + /* If a Transfer-Encoding header field + * is present in a request and the chunked transfer coding is not + * the final encoding, the message body length cannot be determined + * reliably; the server MUST respond with the 400 (Bad Request) + * status code and then close the connection. + */ + return 5; + } else { + /* RFC 7230 3.3.3 */ + + /* If a Transfer-Encoding header field is present in a response and + * the chunked transfer coding is not the final encoding, the + * message body length is determined by reading the connection until + * it is closed by the server. + */ + return 4; + } + } else { + if (!(parser->flags & F_CONTENT_LENGTH)) { + if (!llhttp_message_needs_eof(parser)) { + /* Assume content-length 0 - read the next */ + return 0; + } else { + /* Read body until EOF */ + return 4; + } + } else if (parser->content_length == 0) { + /* Content-Length header given but zero: Content-Length: 0\r\n */ + return 0; + } else { + /* Content-Length header given and non-zero */ + return 3; + } + } +} + + +int llhttp__after_message_complete(llhttp_t* parser, const char* p, + const char* endp) { + int should_keep_alive; + + should_keep_alive = llhttp_should_keep_alive(parser); + parser->finish = HTTP_FINISH_SAFE; + parser->flags = 0; + + /* NOTE: this is ignored in loose parsing mode */ + return should_keep_alive; +} + + +int llhttp_message_needs_eof(const llhttp_t* parser) { + if (parser->type == HTTP_REQUEST) { + return 0; + } + + /* See RFC 2616 section 4.4 */ + if (parser->status_code / 100 == 1 || /* 1xx e.g. Continue */ + parser->status_code == 204 || /* No Content */ + parser->status_code == 304 || /* Not Modified */ + (parser->flags & F_SKIPBODY)) { /* response to a HEAD request */ + return 0; + } + + /* RFC 7230 3.3.3, see `llhttp__after_headers_complete` */ + if ((parser->flags & F_TRANSFER_ENCODING) && + (parser->flags & F_CHUNKED) == 0) { + return 1; + } + + if (parser->flags & (F_CHUNKED | F_CONTENT_LENGTH)) { + return 0; + } + + return 1; +} + + +int llhttp_should_keep_alive(const llhttp_t* parser) { + if (parser->http_major > 0 && parser->http_minor > 0) { + /* HTTP/1.1 */ + if (parser->flags & F_CONNECTION_CLOSE) { + return 0; + } + } else { + /* HTTP/1.0 or earlier */ + if (!(parser->flags & F_CONNECTION_KEEP_ALIVE)) { + return 0; + } + } + + return !llhttp_message_needs_eof(parser); +} diff --git a/contrib/restricted/llhttp/src/llhttp.c b/contrib/restricted/llhttp/src/llhttp.c new file mode 100644 index 0000000000..03f6d57c8c --- /dev/null +++ b/contrib/restricted/llhttp/src/llhttp.c @@ -0,0 +1,9537 @@ +#include <stdlib.h> +#include <stdint.h> +#include <string.h> + +#ifdef __SSE4_2__ + #ifdef _MSC_VER + #include <nmmintrin.h> + #else /* !_MSC_VER */ + #include <x86intrin.h> + #endif /* _MSC_VER */ +#endif /* __SSE4_2__ */ + +#ifdef _MSC_VER + #define ALIGN(n) _declspec(align(n)) +#else /* !_MSC_VER */ + #define ALIGN(n) __attribute__((aligned(n))) +#endif /* _MSC_VER */ + +#include "llhttp.h" + +typedef int (*llhttp__internal__span_cb)( + llhttp__internal_t*, const char*, const char*); + +static const unsigned char llparse_blob0[] = { + 0xd, 0xa +}; +static const unsigned char llparse_blob1[] = { + 'o', 'n' +}; +static const unsigned char llparse_blob2[] = { + 'e', 'c', 't', 'i', 'o', 'n' +}; +static const unsigned char llparse_blob3[] = { + 'l', 'o', 's', 'e' +}; +static const unsigned char llparse_blob4[] = { + 'e', 'e', 'p', '-', 'a', 'l', 'i', 'v', 'e' +}; +static const unsigned char llparse_blob5[] = { + 'p', 'g', 'r', 'a', 'd', 'e' +}; +static const unsigned char llparse_blob6[] = { + 'c', 'h', 'u', 'n', 'k', 'e', 'd' +}; +#ifdef __SSE4_2__ +static const unsigned char ALIGN(16) llparse_blob7[] = { + 0x9, 0x9, ' ', '~', 0x80, 0xff, 0x0, 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0, 0x0 +}; +#endif /* __SSE4_2__ */ +#ifdef __SSE4_2__ +static const unsigned char ALIGN(16) llparse_blob8[] = { + '!', '!', '#', '\'', '*', '+', '-', '.', '0', '9', 'A', + 'Z', '^', 'z', '|', '|' +}; +#endif /* __SSE4_2__ */ +#ifdef __SSE4_2__ +static const unsigned char ALIGN(16) llparse_blob9[] = { + '~', '~', 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, + 0x0, 0x0, 0x0, 0x0, 0x0 +}; +#endif /* __SSE4_2__ */ +static const unsigned char llparse_blob10[] = { + 'e', 'n', 't', '-', 'l', 'e', 'n', 'g', 't', 'h' +}; +static const unsigned char llparse_blob11[] = { + 'r', 'o', 'x', 'y', '-', 'c', 'o', 'n', 'n', 'e', 'c', + 't', 'i', 'o', 'n' +}; +static const unsigned char llparse_blob12[] = { + 'r', 'a', 'n', 's', 'f', 'e', 'r', '-', 'e', 'n', 'c', + 'o', 'd', 'i', 'n', 'g' +}; +static const unsigned char llparse_blob13[] = { + 'p', 'g', 'r', 'a', 'd', 'e' +}; +static const unsigned char llparse_blob14[] = { + 'T', 'T', 'P', '/' +}; +static const unsigned char llparse_blob15[] = { + 0xd, 0xa, 0xd, 0xa, 'S', 'M', 0xd, 0xa, 0xd, 0xa +}; +static const unsigned char llparse_blob16[] = { + 'C', 'E', '/' +}; +static const unsigned char llparse_blob17[] = { + 'T', 'S', 'P', '/' +}; +static const unsigned char llparse_blob18[] = { + 'N', 'O', 'U', 'N', 'C', 'E' +}; +static const unsigned char llparse_blob19[] = { + 'I', 'N', 'D' +}; +static const unsigned char llparse_blob20[] = { + 'E', 'C', 'K', 'O', 'U', 'T' +}; +static const unsigned char llparse_blob21[] = { + 'N', 'E', 'C', 'T' +}; +static const unsigned char llparse_blob22[] = { + 'E', 'T', 'E' +}; +static const unsigned char llparse_blob23[] = { + 'C', 'R', 'I', 'B', 'E' +}; +static const unsigned char llparse_blob24[] = { + 'L', 'U', 'S', 'H' +}; +static const unsigned char llparse_blob25[] = { + 'E', 'T' +}; +static const unsigned char llparse_blob26[] = { + 'P', 'A', 'R', 'A', 'M', 'E', 'T', 'E', 'R' +}; +static const unsigned char llparse_blob27[] = { + 'E', 'A', 'D' +}; +static const unsigned char llparse_blob28[] = { + 'N', 'K' +}; +static const unsigned char llparse_blob29[] = { + 'C', 'K' +}; +static const unsigned char llparse_blob30[] = { + 'S', 'E', 'A', 'R', 'C', 'H' +}; +static const unsigned char llparse_blob31[] = { + 'R', 'G', 'E' +}; +static const unsigned char llparse_blob32[] = { + 'C', 'T', 'I', 'V', 'I', 'T', 'Y' +}; +static const unsigned char llparse_blob33[] = { + 'L', 'E', 'N', 'D', 'A', 'R' +}; +static const unsigned char llparse_blob34[] = { + 'V', 'E' +}; +static const unsigned char llparse_blob35[] = { + 'O', 'T', 'I', 'F', 'Y' +}; +static const unsigned char llparse_blob36[] = { + 'P', 'T', 'I', 'O', 'N', 'S' +}; +static const unsigned char llparse_blob37[] = { + 'C', 'H' +}; +static const unsigned char llparse_blob38[] = { + 'S', 'E' +}; +static const unsigned char llparse_blob39[] = { + 'A', 'Y' +}; +static const unsigned char llparse_blob40[] = { + 'S', 'T' +}; +static const unsigned char llparse_blob41[] = { + 'I', 'N', 'D' +}; +static const unsigned char llparse_blob42[] = { + 'A', 'T', 'C', 'H' +}; +static const unsigned char llparse_blob43[] = { + 'G', 'E' +}; +static const unsigned char llparse_blob44[] = { + 'I', 'N', 'D' +}; +static const unsigned char llparse_blob45[] = { + 'O', 'R', 'D' +}; +static const unsigned char llparse_blob46[] = { + 'I', 'R', 'E', 'C', 'T' +}; +static const unsigned char llparse_blob47[] = { + 'O', 'R', 'T' +}; +static const unsigned char llparse_blob48[] = { + 'R', 'C', 'H' +}; +static const unsigned char llparse_blob49[] = { + 'P', 'A', 'R', 'A', 'M', 'E', 'T', 'E', 'R' +}; +static const unsigned char llparse_blob50[] = { + 'U', 'R', 'C', 'E' +}; +static const unsigned char llparse_blob51[] = { + 'B', 'S', 'C', 'R', 'I', 'B', 'E' +}; +static const unsigned char llparse_blob52[] = { + 'A', 'R', 'D', 'O', 'W', 'N' +}; +static const unsigned char llparse_blob53[] = { + 'A', 'C', 'E' +}; +static const unsigned char llparse_blob54[] = { + 'I', 'N', 'D' +}; +static const unsigned char llparse_blob55[] = { + 'N', 'K' +}; +static const unsigned char llparse_blob56[] = { + 'C', 'K' +}; +static const unsigned char llparse_blob57[] = { + 'U', 'B', 'S', 'C', 'R', 'I', 'B', 'E' +}; +static const unsigned char llparse_blob58[] = { + 'H', 'T', 'T', 'P', '/' +}; +static const unsigned char llparse_blob59[] = { + 'A', 'D' +}; +static const unsigned char llparse_blob60[] = { + 'T', 'P', '/' +}; + +enum llparse_match_status_e { + kMatchComplete, + kMatchPause, + kMatchMismatch +}; +typedef enum llparse_match_status_e llparse_match_status_t; + +struct llparse_match_s { + llparse_match_status_t status; + const unsigned char* current; +}; +typedef struct llparse_match_s llparse_match_t; + +static llparse_match_t llparse__match_sequence_id( + llhttp__internal_t* s, const unsigned char* p, + const unsigned char* endp, + const unsigned char* seq, uint32_t seq_len) { + uint32_t index; + llparse_match_t res; + + index = s->_index; + for (; p != endp; p++) { + unsigned char current; + + current = *p; + if (current == seq[index]) { + if (++index == seq_len) { + res.status = kMatchComplete; + goto reset; + } + } else { + res.status = kMatchMismatch; + goto reset; + } + } + s->_index = index; + res.status = kMatchPause; + res.current = p; + return res; +reset: + s->_index = 0; + res.current = p; + return res; +} + +static llparse_match_t llparse__match_sequence_to_lower( + llhttp__internal_t* s, const unsigned char* p, + const unsigned char* endp, + const unsigned char* seq, uint32_t seq_len) { + uint32_t index; + llparse_match_t res; + + index = s->_index; + for (; p != endp; p++) { + unsigned char current; + + current = ((*p) >= 'A' && (*p) <= 'Z' ? (*p | 0x20) : (*p)); + if (current == seq[index]) { + if (++index == seq_len) { + res.status = kMatchComplete; + goto reset; + } + } else { + res.status = kMatchMismatch; + goto reset; + } + } + s->_index = index; + res.status = kMatchPause; + res.current = p; + return res; +reset: + s->_index = 0; + res.current = p; + return res; +} + +static llparse_match_t llparse__match_sequence_to_lower_unsafe( + llhttp__internal_t* s, const unsigned char* p, + const unsigned char* endp, + const unsigned char* seq, uint32_t seq_len) { + uint32_t index; + llparse_match_t res; + + index = s->_index; + for (; p != endp; p++) { + unsigned char current; + + current = ((*p) | 0x20); + if (current == seq[index]) { + if (++index == seq_len) { + res.status = kMatchComplete; + goto reset; + } + } else { + res.status = kMatchMismatch; + goto reset; + } + } + s->_index = index; + res.status = kMatchPause; + res.current = p; + return res; +reset: + s->_index = 0; + res.current = p; + return res; +} + +enum llparse_state_e { + s_error, + s_n_llhttp__internal__n_closed, + s_n_llhttp__internal__n_invoke_llhttp__after_message_complete, + s_n_llhttp__internal__n_pause_1, + s_n_llhttp__internal__n_chunk_data_almost_done, + s_n_llhttp__internal__n_consume_content_length, + s_n_llhttp__internal__n_span_start_llhttp__on_body, + s_n_llhttp__internal__n_invoke_is_equal_content_length, + s_n_llhttp__internal__n_chunk_size_almost_done, + s_n_llhttp__internal__n_invoke_llhttp__on_chunk_extension_name_complete, + s_n_llhttp__internal__n_invoke_llhttp__on_chunk_extension_name_complete_1, + s_n_llhttp__internal__n_invoke_llhttp__on_chunk_extension_value_complete, + s_n_llhttp__internal__n_chunk_extension_quoted_value_done, + s_n_llhttp__internal__n_invoke_llhttp__on_chunk_extension_value_complete_1, + s_n_llhttp__internal__n_error_21, + s_n_llhttp__internal__n_chunk_extension_quoted_value, + s_n_llhttp__internal__n_invoke_llhttp__on_chunk_extension_value_complete_2, + s_n_llhttp__internal__n_error_23, + s_n_llhttp__internal__n_chunk_extension_value, + s_n_llhttp__internal__n_span_start_llhttp__on_chunk_extension_value, + s_n_llhttp__internal__n_error_24, + s_n_llhttp__internal__n_chunk_extension_name, + s_n_llhttp__internal__n_span_start_llhttp__on_chunk_extension_name, + s_n_llhttp__internal__n_chunk_extensions, + s_n_llhttp__internal__n_chunk_size_otherwise, + s_n_llhttp__internal__n_chunk_size, + s_n_llhttp__internal__n_chunk_size_digit, + s_n_llhttp__internal__n_invoke_update_content_length_1, + s_n_llhttp__internal__n_invoke_is_equal_upgrade, + s_n_llhttp__internal__n_invoke_llhttp__on_message_complete_2, + s_n_llhttp__internal__n_consume_content_length_1, + s_n_llhttp__internal__n_span_start_llhttp__on_body_1, + s_n_llhttp__internal__n_eof, + s_n_llhttp__internal__n_span_start_llhttp__on_body_2, + s_n_llhttp__internal__n_invoke_llhttp__after_headers_complete, + s_n_llhttp__internal__n_error_5, + s_n_llhttp__internal__n_headers_almost_done, + s_n_llhttp__internal__n_header_field_colon_discard_ws, + s_n_llhttp__internal__n_invoke_llhttp__on_header_value_complete, + s_n_llhttp__internal__n_span_start_llhttp__on_header_value, + s_n_llhttp__internal__n_header_value_discard_lws, + s_n_llhttp__internal__n_header_value_discard_ws_almost_done, + s_n_llhttp__internal__n_header_value_lws, + s_n_llhttp__internal__n_header_value_almost_done, + s_n_llhttp__internal__n_header_value_lenient, + s_n_llhttp__internal__n_error_41, + s_n_llhttp__internal__n_header_value_otherwise, + s_n_llhttp__internal__n_header_value_connection_token, + s_n_llhttp__internal__n_header_value_connection_ws, + s_n_llhttp__internal__n_header_value_connection_1, + s_n_llhttp__internal__n_header_value_connection_2, + s_n_llhttp__internal__n_header_value_connection_3, + s_n_llhttp__internal__n_header_value_connection, + s_n_llhttp__internal__n_error_43, + s_n_llhttp__internal__n_error_44, + s_n_llhttp__internal__n_header_value_content_length_ws, + s_n_llhttp__internal__n_header_value_content_length, + s_n_llhttp__internal__n_error_46, + s_n_llhttp__internal__n_error_45, + s_n_llhttp__internal__n_header_value_te_token_ows, + s_n_llhttp__internal__n_header_value, + s_n_llhttp__internal__n_header_value_te_token, + s_n_llhttp__internal__n_header_value_te_chunked_last, + s_n_llhttp__internal__n_header_value_te_chunked, + s_n_llhttp__internal__n_span_start_llhttp__on_header_value_1, + s_n_llhttp__internal__n_header_value_discard_ws, + s_n_llhttp__internal__n_invoke_llhttp__on_header_field_complete, + s_n_llhttp__internal__n_header_field_general_otherwise, + s_n_llhttp__internal__n_header_field_general, + s_n_llhttp__internal__n_header_field_colon, + s_n_llhttp__internal__n_header_field_3, + s_n_llhttp__internal__n_header_field_4, + s_n_llhttp__internal__n_header_field_2, + s_n_llhttp__internal__n_header_field_1, + s_n_llhttp__internal__n_header_field_5, + s_n_llhttp__internal__n_header_field_6, + s_n_llhttp__internal__n_header_field_7, + s_n_llhttp__internal__n_header_field, + s_n_llhttp__internal__n_span_start_llhttp__on_header_field, + s_n_llhttp__internal__n_header_field_start, + s_n_llhttp__internal__n_headers_start, + s_n_llhttp__internal__n_url_to_http_09, + s_n_llhttp__internal__n_url_skip_to_http09, + s_n_llhttp__internal__n_url_skip_lf_to_http09_1, + s_n_llhttp__internal__n_url_skip_lf_to_http09, + s_n_llhttp__internal__n_req_pri_upgrade, + s_n_llhttp__internal__n_req_http_complete_crlf, + s_n_llhttp__internal__n_req_http_complete, + s_n_llhttp__internal__n_invoke_load_method_1, + s_n_llhttp__internal__n_invoke_llhttp__on_version_complete, + s_n_llhttp__internal__n_error_51, + s_n_llhttp__internal__n_error_57, + s_n_llhttp__internal__n_req_http_minor, + s_n_llhttp__internal__n_error_58, + s_n_llhttp__internal__n_req_http_dot, + s_n_llhttp__internal__n_error_59, + s_n_llhttp__internal__n_req_http_major, + s_n_llhttp__internal__n_span_start_llhttp__on_version, + s_n_llhttp__internal__n_req_http_start_1, + s_n_llhttp__internal__n_req_http_start_2, + s_n_llhttp__internal__n_req_http_start_3, + s_n_llhttp__internal__n_req_http_start, + s_n_llhttp__internal__n_url_to_http, + s_n_llhttp__internal__n_url_skip_to_http, + s_n_llhttp__internal__n_url_fragment, + s_n_llhttp__internal__n_span_end_stub_query_3, + s_n_llhttp__internal__n_url_query, + s_n_llhttp__internal__n_url_query_or_fragment, + s_n_llhttp__internal__n_url_path, + s_n_llhttp__internal__n_span_start_stub_path_2, + s_n_llhttp__internal__n_span_start_stub_path, + s_n_llhttp__internal__n_span_start_stub_path_1, + s_n_llhttp__internal__n_url_server_with_at, + s_n_llhttp__internal__n_url_server, + s_n_llhttp__internal__n_url_schema_delim_1, + s_n_llhttp__internal__n_url_schema_delim, + s_n_llhttp__internal__n_span_end_stub_schema, + s_n_llhttp__internal__n_url_schema, + s_n_llhttp__internal__n_url_start, + s_n_llhttp__internal__n_span_start_llhttp__on_url_1, + s_n_llhttp__internal__n_url_entry_normal, + s_n_llhttp__internal__n_span_start_llhttp__on_url, + s_n_llhttp__internal__n_url_entry_connect, + s_n_llhttp__internal__n_req_spaces_before_url, + s_n_llhttp__internal__n_req_first_space_before_url, + s_n_llhttp__internal__n_invoke_llhttp__on_method_complete_1, + s_n_llhttp__internal__n_after_start_req_2, + s_n_llhttp__internal__n_after_start_req_3, + s_n_llhttp__internal__n_after_start_req_1, + s_n_llhttp__internal__n_after_start_req_4, + s_n_llhttp__internal__n_after_start_req_6, + s_n_llhttp__internal__n_after_start_req_8, + s_n_llhttp__internal__n_after_start_req_9, + s_n_llhttp__internal__n_after_start_req_7, + s_n_llhttp__internal__n_after_start_req_5, + s_n_llhttp__internal__n_after_start_req_12, + s_n_llhttp__internal__n_after_start_req_13, + s_n_llhttp__internal__n_after_start_req_11, + s_n_llhttp__internal__n_after_start_req_10, + s_n_llhttp__internal__n_after_start_req_14, + s_n_llhttp__internal__n_after_start_req_17, + s_n_llhttp__internal__n_after_start_req_16, + s_n_llhttp__internal__n_after_start_req_15, + s_n_llhttp__internal__n_after_start_req_18, + s_n_llhttp__internal__n_after_start_req_20, + s_n_llhttp__internal__n_after_start_req_21, + s_n_llhttp__internal__n_after_start_req_19, + s_n_llhttp__internal__n_after_start_req_23, + s_n_llhttp__internal__n_after_start_req_24, + s_n_llhttp__internal__n_after_start_req_26, + s_n_llhttp__internal__n_after_start_req_28, + s_n_llhttp__internal__n_after_start_req_29, + s_n_llhttp__internal__n_after_start_req_27, + s_n_llhttp__internal__n_after_start_req_25, + s_n_llhttp__internal__n_after_start_req_30, + s_n_llhttp__internal__n_after_start_req_22, + s_n_llhttp__internal__n_after_start_req_31, + s_n_llhttp__internal__n_after_start_req_32, + s_n_llhttp__internal__n_after_start_req_35, + s_n_llhttp__internal__n_after_start_req_36, + s_n_llhttp__internal__n_after_start_req_34, + s_n_llhttp__internal__n_after_start_req_37, + s_n_llhttp__internal__n_after_start_req_38, + s_n_llhttp__internal__n_after_start_req_42, + s_n_llhttp__internal__n_after_start_req_43, + s_n_llhttp__internal__n_after_start_req_41, + s_n_llhttp__internal__n_after_start_req_40, + s_n_llhttp__internal__n_after_start_req_39, + s_n_llhttp__internal__n_after_start_req_45, + s_n_llhttp__internal__n_after_start_req_44, + s_n_llhttp__internal__n_after_start_req_33, + s_n_llhttp__internal__n_after_start_req_48, + s_n_llhttp__internal__n_after_start_req_49, + s_n_llhttp__internal__n_after_start_req_50, + s_n_llhttp__internal__n_after_start_req_51, + s_n_llhttp__internal__n_after_start_req_47, + s_n_llhttp__internal__n_after_start_req_46, + s_n_llhttp__internal__n_after_start_req_54, + s_n_llhttp__internal__n_after_start_req_56, + s_n_llhttp__internal__n_after_start_req_57, + s_n_llhttp__internal__n_after_start_req_55, + s_n_llhttp__internal__n_after_start_req_53, + s_n_llhttp__internal__n_after_start_req_58, + s_n_llhttp__internal__n_after_start_req_59, + s_n_llhttp__internal__n_after_start_req_52, + s_n_llhttp__internal__n_after_start_req_61, + s_n_llhttp__internal__n_after_start_req_62, + s_n_llhttp__internal__n_after_start_req_60, + s_n_llhttp__internal__n_after_start_req_65, + s_n_llhttp__internal__n_after_start_req_67, + s_n_llhttp__internal__n_after_start_req_68, + s_n_llhttp__internal__n_after_start_req_66, + s_n_llhttp__internal__n_after_start_req_69, + s_n_llhttp__internal__n_after_start_req_64, + s_n_llhttp__internal__n_after_start_req_63, + s_n_llhttp__internal__n_after_start_req, + s_n_llhttp__internal__n_span_start_llhttp__on_method_1, + s_n_llhttp__internal__n_res_line_almost_done, + s_n_llhttp__internal__n_res_status, + s_n_llhttp__internal__n_span_start_llhttp__on_status, + s_n_llhttp__internal__n_res_status_start, + s_n_llhttp__internal__n_res_status_code_otherwise, + s_n_llhttp__internal__n_res_status_code_digit_3, + s_n_llhttp__internal__n_res_status_code_digit_2, + s_n_llhttp__internal__n_res_status_code_digit_1, + s_n_llhttp__internal__n_res_after_version, + s_n_llhttp__internal__n_invoke_llhttp__on_version_complete_1, + s_n_llhttp__internal__n_error_73, + s_n_llhttp__internal__n_error_85, + s_n_llhttp__internal__n_res_http_minor, + s_n_llhttp__internal__n_error_86, + s_n_llhttp__internal__n_res_http_dot, + s_n_llhttp__internal__n_error_87, + s_n_llhttp__internal__n_res_http_major, + s_n_llhttp__internal__n_span_start_llhttp__on_version_1, + s_n_llhttp__internal__n_start_res, + s_n_llhttp__internal__n_invoke_llhttp__on_method_complete, + s_n_llhttp__internal__n_req_or_res_method_2, + s_n_llhttp__internal__n_invoke_update_type_1, + s_n_llhttp__internal__n_req_or_res_method_3, + s_n_llhttp__internal__n_req_or_res_method_1, + s_n_llhttp__internal__n_req_or_res_method, + s_n_llhttp__internal__n_span_start_llhttp__on_method, + s_n_llhttp__internal__n_start_req_or_res, + s_n_llhttp__internal__n_invoke_load_type, + s_n_llhttp__internal__n_invoke_update_finish, + s_n_llhttp__internal__n_start, +}; +typedef enum llparse_state_e llparse_state_t; + +int llhttp__on_method( + llhttp__internal_t* s, const unsigned char* p, + const unsigned char* endp); + +int llhttp__on_url( + llhttp__internal_t* s, const unsigned char* p, + const unsigned char* endp); + +int llhttp__on_version( + llhttp__internal_t* s, const unsigned char* p, + const unsigned char* endp); + +int llhttp__on_header_field( + llhttp__internal_t* s, const unsigned char* p, + const unsigned char* endp); + +int llhttp__on_header_value( + llhttp__internal_t* s, const unsigned char* p, + const unsigned char* endp); + +int llhttp__on_body( + llhttp__internal_t* s, const unsigned char* p, + const unsigned char* endp); + +int llhttp__on_chunk_extension_name( + llhttp__internal_t* s, const unsigned char* p, + const unsigned char* endp); + +int llhttp__on_chunk_extension_value( + llhttp__internal_t* s, const unsigned char* p, + const unsigned char* endp); + +int llhttp__on_status( + llhttp__internal_t* s, const unsigned char* p, + const unsigned char* endp); + +int llhttp__internal__c_load_initial_message_completed( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + return state->initial_message_completed; +} + +int llhttp__on_reset( + llhttp__internal_t* s, const unsigned char* p, + const unsigned char* endp); + +int llhttp__internal__c_update_finish( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + state->finish = 2; + return 0; +} + +int llhttp__on_message_begin( + llhttp__internal_t* s, const unsigned char* p, + const unsigned char* endp); + +int llhttp__internal__c_load_type( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + return state->type; +} + +int llhttp__internal__c_store_method( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp, + int match) { + state->method = match; + return 0; +} + +int llhttp__on_method_complete( + llhttp__internal_t* s, const unsigned char* p, + const unsigned char* endp); + +int llhttp__internal__c_is_equal_method( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + return state->method == 5; +} + +int llhttp__internal__c_update_http_major( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + state->http_major = 0; + return 0; +} + +int llhttp__internal__c_update_http_minor( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + state->http_minor = 9; + return 0; +} + +int llhttp__on_url_complete( + llhttp__internal_t* s, const unsigned char* p, + const unsigned char* endp); + +int llhttp__internal__c_test_lenient_flags( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + return (state->lenient_flags & 1) == 1; +} + +int llhttp__after_headers_complete( + llhttp__internal_t* s, const unsigned char* p, + const unsigned char* endp); + +int llhttp__on_message_complete( + llhttp__internal_t* s, const unsigned char* p, + const unsigned char* endp); + +int llhttp__after_message_complete( + llhttp__internal_t* s, const unsigned char* p, + const unsigned char* endp); + +int llhttp__internal__c_update_content_length( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + state->content_length = 0; + return 0; +} + +int llhttp__internal__c_update_initial_message_completed( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + state->initial_message_completed = 1; + return 0; +} + +int llhttp__internal__c_update_finish_1( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + state->finish = 0; + return 0; +} + +int llhttp__internal__c_test_lenient_flags_2( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + return (state->lenient_flags & 4) == 4; +} + +int llhttp__internal__c_test_lenient_flags_3( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + return (state->lenient_flags & 32) == 32; +} + +int llhttp__internal__c_mul_add_content_length( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp, + int match) { + /* Multiplication overflow */ + if (state->content_length > 0xffffffffffffffffULL / 16) { + return 1; + } + + state->content_length *= 16; + + /* Addition overflow */ + if (match >= 0) { + if (state->content_length > 0xffffffffffffffffULL - match) { + return 1; + } + } else { + if (state->content_length < 0ULL - match) { + return 1; + } + } + state->content_length += match; + return 0; +} + +int llhttp__on_chunk_header( + llhttp__internal_t* s, const unsigned char* p, + const unsigned char* endp); + +int llhttp__internal__c_is_equal_content_length( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + return state->content_length == 0; +} + +int llhttp__on_chunk_complete( + llhttp__internal_t* s, const unsigned char* p, + const unsigned char* endp); + +int llhttp__internal__c_test_lenient_flags_4( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + return (state->lenient_flags & 128) == 128; +} + +int llhttp__internal__c_or_flags( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + state->flags |= 128; + return 0; +} + +int llhttp__internal__c_test_lenient_flags_5( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + return (state->lenient_flags & 64) == 64; +} + +int llhttp__on_chunk_extension_name_complete( + llhttp__internal_t* s, const unsigned char* p, + const unsigned char* endp); + +int llhttp__on_chunk_extension_value_complete( + llhttp__internal_t* s, const unsigned char* p, + const unsigned char* endp); + +int llhttp__internal__c_is_equal_upgrade( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + return state->upgrade == 1; +} + +int llhttp__internal__c_update_finish_3( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + state->finish = 1; + return 0; +} + +int llhttp__internal__c_test_flags( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + return (state->flags & 128) == 128; +} + +int llhttp__internal__c_test_flags_1( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + return (state->flags & 544) == 544; +} + +int llhttp__internal__c_test_lenient_flags_6( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + return (state->lenient_flags & 2) == 2; +} + +int llhttp__before_headers_complete( + llhttp__internal_t* s, const unsigned char* p, + const unsigned char* endp); + +int llhttp__on_headers_complete( + llhttp__internal_t* s, const unsigned char* p, + const unsigned char* endp); + +int llhttp__internal__c_or_flags_1( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + state->flags |= 64; + return 0; +} + +int llhttp__internal__c_update_upgrade( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + state->upgrade = 1; + return 0; +} + +int llhttp__internal__c_store_header_state( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp, + int match) { + state->header_state = match; + return 0; +} + +int llhttp__on_header_field_complete( + llhttp__internal_t* s, const unsigned char* p, + const unsigned char* endp); + +int llhttp__internal__c_load_header_state( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + return state->header_state; +} + +int llhttp__internal__c_or_flags_3( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + state->flags |= 1; + return 0; +} + +int llhttp__internal__c_update_header_state( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + state->header_state = 1; + return 0; +} + +int llhttp__on_header_value_complete( + llhttp__internal_t* s, const unsigned char* p, + const unsigned char* endp); + +int llhttp__internal__c_or_flags_4( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + state->flags |= 2; + return 0; +} + +int llhttp__internal__c_or_flags_5( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + state->flags |= 4; + return 0; +} + +int llhttp__internal__c_or_flags_6( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + state->flags |= 8; + return 0; +} + +int llhttp__internal__c_update_header_state_3( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + state->header_state = 6; + return 0; +} + +int llhttp__internal__c_update_header_state_1( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + state->header_state = 0; + return 0; +} + +int llhttp__internal__c_update_header_state_6( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + state->header_state = 5; + return 0; +} + +int llhttp__internal__c_update_header_state_7( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + state->header_state = 7; + return 0; +} + +int llhttp__internal__c_test_flags_2( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + return (state->flags & 32) == 32; +} + +int llhttp__internal__c_mul_add_content_length_1( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp, + int match) { + /* Multiplication overflow */ + if (state->content_length > 0xffffffffffffffffULL / 10) { + return 1; + } + + state->content_length *= 10; + + /* Addition overflow */ + if (match >= 0) { + if (state->content_length > 0xffffffffffffffffULL - match) { + return 1; + } + } else { + if (state->content_length < 0ULL - match) { + return 1; + } + } + state->content_length += match; + return 0; +} + +int llhttp__internal__c_or_flags_15( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + state->flags |= 32; + return 0; +} + +int llhttp__internal__c_test_flags_3( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + return (state->flags & 8) == 8; +} + +int llhttp__internal__c_test_lenient_flags_13( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + return (state->lenient_flags & 8) == 8; +} + +int llhttp__internal__c_or_flags_16( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + state->flags |= 512; + return 0; +} + +int llhttp__internal__c_and_flags( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + state->flags &= -9; + return 0; +} + +int llhttp__internal__c_update_header_state_8( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + state->header_state = 8; + return 0; +} + +int llhttp__internal__c_or_flags_18( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + state->flags |= 16; + return 0; +} + +int llhttp__internal__c_load_method( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + return state->method; +} + +int llhttp__internal__c_store_http_major( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp, + int match) { + state->http_major = match; + return 0; +} + +int llhttp__internal__c_store_http_minor( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp, + int match) { + state->http_minor = match; + return 0; +} + +int llhttp__internal__c_test_lenient_flags_15( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + return (state->lenient_flags & 16) == 16; +} + +int llhttp__on_version_complete( + llhttp__internal_t* s, const unsigned char* p, + const unsigned char* endp); + +int llhttp__internal__c_load_http_major( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + return state->http_major; +} + +int llhttp__internal__c_load_http_minor( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + return state->http_minor; +} + +int llhttp__internal__c_update_status_code( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + state->status_code = 0; + return 0; +} + +int llhttp__internal__c_mul_add_status_code( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp, + int match) { + /* Multiplication overflow */ + if (state->status_code > 0xffff / 10) { + return 1; + } + + state->status_code *= 10; + + /* Addition overflow */ + if (match >= 0) { + if (state->status_code > 0xffff - match) { + return 1; + } + } else { + if (state->status_code < 0 - match) { + return 1; + } + } + state->status_code += match; + return 0; +} + +int llhttp__on_status_complete( + llhttp__internal_t* s, const unsigned char* p, + const unsigned char* endp); + +int llhttp__internal__c_update_type( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + state->type = 1; + return 0; +} + +int llhttp__internal__c_update_type_1( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + state->type = 2; + return 0; +} + +int llhttp__internal_init(llhttp__internal_t* state) { + memset(state, 0, sizeof(*state)); + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_start; + return 0; +} + +static llparse_state_t llhttp__internal__run( + llhttp__internal_t* state, + const unsigned char* p, + const unsigned char* endp) { + int match; + switch ((llparse_state_t) (intptr_t) state->_current) { + case s_n_llhttp__internal__n_closed: + s_n_llhttp__internal__n_closed: { + if (p == endp) { + return s_n_llhttp__internal__n_closed; + } + switch (*p) { + case 10: { + p++; + goto s_n_llhttp__internal__n_closed; + } + case 13: { + p++; + goto s_n_llhttp__internal__n_closed; + } + default: { + p++; + goto s_n_llhttp__internal__n_invoke_test_lenient_flags_3; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_invoke_llhttp__after_message_complete: + s_n_llhttp__internal__n_invoke_llhttp__after_message_complete: { + switch (llhttp__after_message_complete(state, p, endp)) { + case 1: + goto s_n_llhttp__internal__n_invoke_update_content_length; + default: + goto s_n_llhttp__internal__n_invoke_update_finish_1; + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_pause_1: + s_n_llhttp__internal__n_pause_1: { + state->error = 0x16; + state->reason = "Pause on CONNECT/Upgrade"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_invoke_llhttp__after_message_complete; + return s_error; + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_chunk_data_almost_done: + s_n_llhttp__internal__n_chunk_data_almost_done: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_chunk_data_almost_done; + } + match_seq = llparse__match_sequence_id(state, p, endp, llparse_blob0, 2); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + goto s_n_llhttp__internal__n_invoke_llhttp__on_chunk_complete; + } + case kMatchPause: { + return s_n_llhttp__internal__n_chunk_data_almost_done; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_invoke_test_lenient_flags_4; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_consume_content_length: + s_n_llhttp__internal__n_consume_content_length: { + size_t avail; + uint64_t need; + + avail = endp - p; + need = state->content_length; + if (avail >= need) { + p += need; + state->content_length = 0; + goto s_n_llhttp__internal__n_span_end_llhttp__on_body; + } + + state->content_length -= avail; + return s_n_llhttp__internal__n_consume_content_length; + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_span_start_llhttp__on_body: + s_n_llhttp__internal__n_span_start_llhttp__on_body: { + if (p == endp) { + return s_n_llhttp__internal__n_span_start_llhttp__on_body; + } + state->_span_pos0 = (void*) p; + state->_span_cb0 = llhttp__on_body; + goto s_n_llhttp__internal__n_consume_content_length; + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_invoke_is_equal_content_length: + s_n_llhttp__internal__n_invoke_is_equal_content_length: { + switch (llhttp__internal__c_is_equal_content_length(state, p, endp)) { + case 0: + goto s_n_llhttp__internal__n_span_start_llhttp__on_body; + default: + goto s_n_llhttp__internal__n_invoke_or_flags; + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_chunk_size_almost_done: + s_n_llhttp__internal__n_chunk_size_almost_done: { + if (p == endp) { + return s_n_llhttp__internal__n_chunk_size_almost_done; + } + switch (*p) { + case 10: { + p++; + goto s_n_llhttp__internal__n_invoke_llhttp__on_chunk_header; + } + default: { + goto s_n_llhttp__internal__n_invoke_test_lenient_flags_5; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_invoke_llhttp__on_chunk_extension_name_complete: + s_n_llhttp__internal__n_invoke_llhttp__on_chunk_extension_name_complete: { + switch (llhttp__on_chunk_extension_name_complete(state, p, endp)) { + case 0: + goto s_n_llhttp__internal__n_chunk_size_almost_done; + case 21: + goto s_n_llhttp__internal__n_pause_5; + default: + goto s_n_llhttp__internal__n_error_15; + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_invoke_llhttp__on_chunk_extension_name_complete_1: + s_n_llhttp__internal__n_invoke_llhttp__on_chunk_extension_name_complete_1: { + switch (llhttp__on_chunk_extension_name_complete(state, p, endp)) { + case 0: + goto s_n_llhttp__internal__n_chunk_extensions; + case 21: + goto s_n_llhttp__internal__n_pause_6; + default: + goto s_n_llhttp__internal__n_error_16; + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_invoke_llhttp__on_chunk_extension_value_complete: + s_n_llhttp__internal__n_invoke_llhttp__on_chunk_extension_value_complete: { + switch (llhttp__on_chunk_extension_value_complete(state, p, endp)) { + case 0: + goto s_n_llhttp__internal__n_chunk_size_almost_done; + case 21: + goto s_n_llhttp__internal__n_pause_7; + default: + goto s_n_llhttp__internal__n_error_18; + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_chunk_extension_quoted_value_done: + s_n_llhttp__internal__n_chunk_extension_quoted_value_done: { + if (p == endp) { + return s_n_llhttp__internal__n_chunk_extension_quoted_value_done; + } + switch (*p) { + case 13: { + p++; + goto s_n_llhttp__internal__n_chunk_size_almost_done; + } + case ';': { + p++; + goto s_n_llhttp__internal__n_chunk_extensions; + } + default: { + goto s_n_llhttp__internal__n_error_20; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_invoke_llhttp__on_chunk_extension_value_complete_1: + s_n_llhttp__internal__n_invoke_llhttp__on_chunk_extension_value_complete_1: { + switch (llhttp__on_chunk_extension_value_complete(state, p, endp)) { + case 0: + goto s_n_llhttp__internal__n_chunk_extension_quoted_value_done; + case 21: + goto s_n_llhttp__internal__n_pause_8; + default: + goto s_n_llhttp__internal__n_error_19; + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_error_21: + s_n_llhttp__internal__n_error_21: { + state->error = 0x2; + state->reason = "Invalid character in chunk extensions quoted value"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_chunk_extension_quoted_value: + s_n_llhttp__internal__n_chunk_extension_quoted_value: { + static uint8_t lookup_table[] = { + 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 1, 1, 2, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1 + }; + if (p == endp) { + return s_n_llhttp__internal__n_chunk_extension_quoted_value; + } + switch (lookup_table[(uint8_t) *p]) { + case 1: { + p++; + goto s_n_llhttp__internal__n_chunk_extension_quoted_value; + } + case 2: { + p++; + goto s_n_llhttp__internal__n_span_end_llhttp__on_chunk_extension_value_1; + } + default: { + goto s_n_llhttp__internal__n_span_end_llhttp__on_chunk_extension_value_2; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_invoke_llhttp__on_chunk_extension_value_complete_2: + s_n_llhttp__internal__n_invoke_llhttp__on_chunk_extension_value_complete_2: { + switch (llhttp__on_chunk_extension_value_complete(state, p, endp)) { + case 0: + goto s_n_llhttp__internal__n_chunk_size_otherwise; + case 21: + goto s_n_llhttp__internal__n_pause_9; + default: + goto s_n_llhttp__internal__n_error_22; + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_error_23: + s_n_llhttp__internal__n_error_23: { + state->error = 0x2; + state->reason = "Invalid character in chunk extensions value"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_chunk_extension_value: + s_n_llhttp__internal__n_chunk_extension_value: { + static uint8_t lookup_table[] = { + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 2, 3, 2, 2, 2, 2, 2, 0, 0, 2, 2, 0, 2, 2, 0, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 0, 4, 0, 0, 0, 0, + 0, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 0, 0, 0, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 0, 2, 0, 2, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 + }; + if (p == endp) { + return s_n_llhttp__internal__n_chunk_extension_value; + } + switch (lookup_table[(uint8_t) *p]) { + case 1: { + goto s_n_llhttp__internal__n_span_end_llhttp__on_chunk_extension_value; + } + case 2: { + p++; + goto s_n_llhttp__internal__n_chunk_extension_value; + } + case 3: { + p++; + goto s_n_llhttp__internal__n_chunk_extension_quoted_value; + } + case 4: { + goto s_n_llhttp__internal__n_span_end_llhttp__on_chunk_extension_value_3; + } + default: { + goto s_n_llhttp__internal__n_span_end_llhttp__on_chunk_extension_value_4; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_span_start_llhttp__on_chunk_extension_value: + s_n_llhttp__internal__n_span_start_llhttp__on_chunk_extension_value: { + if (p == endp) { + return s_n_llhttp__internal__n_span_start_llhttp__on_chunk_extension_value; + } + state->_span_pos0 = (void*) p; + state->_span_cb0 = llhttp__on_chunk_extension_value; + goto s_n_llhttp__internal__n_invoke_llhttp__on_chunk_extension_name_complete_2; + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_error_24: + s_n_llhttp__internal__n_error_24: { + state->error = 0x2; + state->reason = "Invalid character in chunk extensions name"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_chunk_extension_name: + s_n_llhttp__internal__n_chunk_extension_name: { + static uint8_t lookup_table[] = { + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 2, 0, 2, 2, 2, 2, 2, 0, 0, 2, 2, 0, 2, 2, 0, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 0, 3, 0, 4, 0, 0, + 0, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 0, 0, 0, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 0, 2, 0, 2, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 + }; + if (p == endp) { + return s_n_llhttp__internal__n_chunk_extension_name; + } + switch (lookup_table[(uint8_t) *p]) { + case 1: { + goto s_n_llhttp__internal__n_span_end_llhttp__on_chunk_extension_name; + } + case 2: { + p++; + goto s_n_llhttp__internal__n_chunk_extension_name; + } + case 3: { + goto s_n_llhttp__internal__n_span_end_llhttp__on_chunk_extension_name_1; + } + case 4: { + goto s_n_llhttp__internal__n_span_end_llhttp__on_chunk_extension_name_2; + } + default: { + goto s_n_llhttp__internal__n_span_end_llhttp__on_chunk_extension_name_3; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_span_start_llhttp__on_chunk_extension_name: + s_n_llhttp__internal__n_span_start_llhttp__on_chunk_extension_name: { + if (p == endp) { + return s_n_llhttp__internal__n_span_start_llhttp__on_chunk_extension_name; + } + state->_span_pos0 = (void*) p; + state->_span_cb0 = llhttp__on_chunk_extension_name; + goto s_n_llhttp__internal__n_chunk_extension_name; + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_chunk_extensions: + s_n_llhttp__internal__n_chunk_extensions: { + if (p == endp) { + return s_n_llhttp__internal__n_chunk_extensions; + } + switch (*p) { + case 13: { + p++; + goto s_n_llhttp__internal__n_error_13; + } + case ' ': { + p++; + goto s_n_llhttp__internal__n_error_14; + } + default: { + goto s_n_llhttp__internal__n_span_start_llhttp__on_chunk_extension_name; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_chunk_size_otherwise: + s_n_llhttp__internal__n_chunk_size_otherwise: { + if (p == endp) { + return s_n_llhttp__internal__n_chunk_size_otherwise; + } + switch (*p) { + case 13: { + p++; + goto s_n_llhttp__internal__n_chunk_size_almost_done; + } + case ';': { + p++; + goto s_n_llhttp__internal__n_chunk_extensions; + } + default: { + goto s_n_llhttp__internal__n_error_25; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_chunk_size: + s_n_llhttp__internal__n_chunk_size: { + if (p == endp) { + return s_n_llhttp__internal__n_chunk_size; + } + switch (*p) { + case '0': { + p++; + match = 0; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length; + } + case '1': { + p++; + match = 1; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length; + } + case '2': { + p++; + match = 2; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length; + } + case '3': { + p++; + match = 3; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length; + } + case '4': { + p++; + match = 4; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length; + } + case '5': { + p++; + match = 5; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length; + } + case '6': { + p++; + match = 6; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length; + } + case '7': { + p++; + match = 7; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length; + } + case '8': { + p++; + match = 8; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length; + } + case '9': { + p++; + match = 9; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length; + } + case 'A': { + p++; + match = 10; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length; + } + case 'B': { + p++; + match = 11; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length; + } + case 'C': { + p++; + match = 12; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length; + } + case 'D': { + p++; + match = 13; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length; + } + case 'E': { + p++; + match = 14; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length; + } + case 'F': { + p++; + match = 15; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length; + } + case 'a': { + p++; + match = 10; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length; + } + case 'b': { + p++; + match = 11; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length; + } + case 'c': { + p++; + match = 12; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length; + } + case 'd': { + p++; + match = 13; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length; + } + case 'e': { + p++; + match = 14; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length; + } + case 'f': { + p++; + match = 15; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length; + } + default: { + goto s_n_llhttp__internal__n_chunk_size_otherwise; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_chunk_size_digit: + s_n_llhttp__internal__n_chunk_size_digit: { + if (p == endp) { + return s_n_llhttp__internal__n_chunk_size_digit; + } + switch (*p) { + case '0': { + p++; + match = 0; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length; + } + case '1': { + p++; + match = 1; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length; + } + case '2': { + p++; + match = 2; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length; + } + case '3': { + p++; + match = 3; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length; + } + case '4': { + p++; + match = 4; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length; + } + case '5': { + p++; + match = 5; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length; + } + case '6': { + p++; + match = 6; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length; + } + case '7': { + p++; + match = 7; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length; + } + case '8': { + p++; + match = 8; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length; + } + case '9': { + p++; + match = 9; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length; + } + case 'A': { + p++; + match = 10; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length; + } + case 'B': { + p++; + match = 11; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length; + } + case 'C': { + p++; + match = 12; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length; + } + case 'D': { + p++; + match = 13; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length; + } + case 'E': { + p++; + match = 14; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length; + } + case 'F': { + p++; + match = 15; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length; + } + case 'a': { + p++; + match = 10; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length; + } + case 'b': { + p++; + match = 11; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length; + } + case 'c': { + p++; + match = 12; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length; + } + case 'd': { + p++; + match = 13; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length; + } + case 'e': { + p++; + match = 14; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length; + } + case 'f': { + p++; + match = 15; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length; + } + default: { + goto s_n_llhttp__internal__n_error_27; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_invoke_update_content_length_1: + s_n_llhttp__internal__n_invoke_update_content_length_1: { + switch (llhttp__internal__c_update_content_length(state, p, endp)) { + default: + goto s_n_llhttp__internal__n_chunk_size_digit; + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_invoke_is_equal_upgrade: + s_n_llhttp__internal__n_invoke_is_equal_upgrade: { + switch (llhttp__internal__c_is_equal_upgrade(state, p, endp)) { + case 0: + goto s_n_llhttp__internal__n_invoke_llhttp__after_message_complete; + default: + goto s_n_llhttp__internal__n_pause_1; + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_invoke_llhttp__on_message_complete_2: + s_n_llhttp__internal__n_invoke_llhttp__on_message_complete_2: { + switch (llhttp__on_message_complete(state, p, endp)) { + case 0: + goto s_n_llhttp__internal__n_invoke_is_equal_upgrade; + case 21: + goto s_n_llhttp__internal__n_pause_11; + default: + goto s_n_llhttp__internal__n_error_28; + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_consume_content_length_1: + s_n_llhttp__internal__n_consume_content_length_1: { + size_t avail; + uint64_t need; + + avail = endp - p; + need = state->content_length; + if (avail >= need) { + p += need; + state->content_length = 0; + goto s_n_llhttp__internal__n_span_end_llhttp__on_body_1; + } + + state->content_length -= avail; + return s_n_llhttp__internal__n_consume_content_length_1; + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_span_start_llhttp__on_body_1: + s_n_llhttp__internal__n_span_start_llhttp__on_body_1: { + if (p == endp) { + return s_n_llhttp__internal__n_span_start_llhttp__on_body_1; + } + state->_span_pos0 = (void*) p; + state->_span_cb0 = llhttp__on_body; + goto s_n_llhttp__internal__n_consume_content_length_1; + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_eof: + s_n_llhttp__internal__n_eof: { + if (p == endp) { + return s_n_llhttp__internal__n_eof; + } + p++; + goto s_n_llhttp__internal__n_eof; + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_span_start_llhttp__on_body_2: + s_n_llhttp__internal__n_span_start_llhttp__on_body_2: { + if (p == endp) { + return s_n_llhttp__internal__n_span_start_llhttp__on_body_2; + } + state->_span_pos0 = (void*) p; + state->_span_cb0 = llhttp__on_body; + goto s_n_llhttp__internal__n_eof; + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_invoke_llhttp__after_headers_complete: + s_n_llhttp__internal__n_invoke_llhttp__after_headers_complete: { + switch (llhttp__after_headers_complete(state, p, endp)) { + case 1: + goto s_n_llhttp__internal__n_invoke_llhttp__on_message_complete_1; + case 2: + goto s_n_llhttp__internal__n_invoke_update_content_length_1; + case 3: + goto s_n_llhttp__internal__n_span_start_llhttp__on_body_1; + case 4: + goto s_n_llhttp__internal__n_invoke_update_finish_3; + case 5: + goto s_n_llhttp__internal__n_error_29; + default: + goto s_n_llhttp__internal__n_invoke_llhttp__on_message_complete; + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_error_5: + s_n_llhttp__internal__n_error_5: { + state->error = 0xa; + state->reason = "Invalid header field char"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_headers_almost_done: + s_n_llhttp__internal__n_headers_almost_done: { + if (p == endp) { + return s_n_llhttp__internal__n_headers_almost_done; + } + switch (*p) { + case 10: { + p++; + goto s_n_llhttp__internal__n_invoke_test_flags; + } + default: { + goto s_n_llhttp__internal__n_invoke_test_lenient_flags_7; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_header_field_colon_discard_ws: + s_n_llhttp__internal__n_header_field_colon_discard_ws: { + if (p == endp) { + return s_n_llhttp__internal__n_header_field_colon_discard_ws; + } + switch (*p) { + case ' ': { + p++; + goto s_n_llhttp__internal__n_header_field_colon_discard_ws; + } + default: { + goto s_n_llhttp__internal__n_header_field_colon; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_invoke_llhttp__on_header_value_complete: + s_n_llhttp__internal__n_invoke_llhttp__on_header_value_complete: { + switch (llhttp__on_header_value_complete(state, p, endp)) { + case 0: + goto s_n_llhttp__internal__n_header_field_start; + case 21: + goto s_n_llhttp__internal__n_pause_14; + default: + goto s_n_llhttp__internal__n_error_37; + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_span_start_llhttp__on_header_value: + s_n_llhttp__internal__n_span_start_llhttp__on_header_value: { + if (p == endp) { + return s_n_llhttp__internal__n_span_start_llhttp__on_header_value; + } + state->_span_pos0 = (void*) p; + state->_span_cb0 = llhttp__on_header_value; + goto s_n_llhttp__internal__n_span_end_llhttp__on_header_value; + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_header_value_discard_lws: + s_n_llhttp__internal__n_header_value_discard_lws: { + if (p == endp) { + return s_n_llhttp__internal__n_header_value_discard_lws; + } + switch (*p) { + case 9: { + p++; + goto s_n_llhttp__internal__n_invoke_test_lenient_flags_10; + } + case ' ': { + p++; + goto s_n_llhttp__internal__n_invoke_test_lenient_flags_10; + } + default: { + goto s_n_llhttp__internal__n_invoke_load_header_state; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_header_value_discard_ws_almost_done: + s_n_llhttp__internal__n_header_value_discard_ws_almost_done: { + if (p == endp) { + return s_n_llhttp__internal__n_header_value_discard_ws_almost_done; + } + switch (*p) { + case 10: { + p++; + goto s_n_llhttp__internal__n_header_value_discard_lws; + } + default: { + goto s_n_llhttp__internal__n_invoke_test_lenient_flags_11; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_header_value_lws: + s_n_llhttp__internal__n_header_value_lws: { + if (p == endp) { + return s_n_llhttp__internal__n_header_value_lws; + } + switch (*p) { + case 9: { + goto s_n_llhttp__internal__n_invoke_load_header_state_3; + } + case ' ': { + goto s_n_llhttp__internal__n_invoke_load_header_state_3; + } + default: { + goto s_n_llhttp__internal__n_invoke_load_header_state_4; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_header_value_almost_done: + s_n_llhttp__internal__n_header_value_almost_done: { + if (p == endp) { + return s_n_llhttp__internal__n_header_value_almost_done; + } + switch (*p) { + case 10: { + p++; + goto s_n_llhttp__internal__n_header_value_lws; + } + default: { + goto s_n_llhttp__internal__n_error_40; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_header_value_lenient: + s_n_llhttp__internal__n_header_value_lenient: { + if (p == endp) { + return s_n_llhttp__internal__n_header_value_lenient; + } + switch (*p) { + case 10: { + goto s_n_llhttp__internal__n_span_end_llhttp__on_header_value_3; + } + case 13: { + goto s_n_llhttp__internal__n_span_end_llhttp__on_header_value_4; + } + default: { + p++; + goto s_n_llhttp__internal__n_header_value_lenient; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_error_41: + s_n_llhttp__internal__n_error_41: { + state->error = 0xa; + state->reason = "Invalid header value char"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_header_value_otherwise: + s_n_llhttp__internal__n_header_value_otherwise: { + if (p == endp) { + return s_n_llhttp__internal__n_header_value_otherwise; + } + switch (*p) { + case 13: { + goto s_n_llhttp__internal__n_span_end_llhttp__on_header_value_1; + } + default: { + goto s_n_llhttp__internal__n_invoke_test_lenient_flags_12; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_header_value_connection_token: + s_n_llhttp__internal__n_header_value_connection_token: { + static uint8_t lookup_table[] = { + 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1 + }; + if (p == endp) { + return s_n_llhttp__internal__n_header_value_connection_token; + } + switch (lookup_table[(uint8_t) *p]) { + case 1: { + p++; + goto s_n_llhttp__internal__n_header_value_connection_token; + } + case 2: { + p++; + goto s_n_llhttp__internal__n_header_value_connection; + } + default: { + goto s_n_llhttp__internal__n_header_value_otherwise; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_header_value_connection_ws: + s_n_llhttp__internal__n_header_value_connection_ws: { + if (p == endp) { + return s_n_llhttp__internal__n_header_value_connection_ws; + } + switch (*p) { + case 10: { + goto s_n_llhttp__internal__n_header_value_otherwise; + } + case 13: { + goto s_n_llhttp__internal__n_header_value_otherwise; + } + case ' ': { + p++; + goto s_n_llhttp__internal__n_header_value_connection_ws; + } + case ',': { + p++; + goto s_n_llhttp__internal__n_invoke_load_header_state_5; + } + default: { + goto s_n_llhttp__internal__n_invoke_update_header_state_5; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_header_value_connection_1: + s_n_llhttp__internal__n_header_value_connection_1: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_header_value_connection_1; + } + match_seq = llparse__match_sequence_to_lower(state, p, endp, llparse_blob3, 4); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + goto s_n_llhttp__internal__n_invoke_update_header_state_3; + } + case kMatchPause: { + return s_n_llhttp__internal__n_header_value_connection_1; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_header_value_connection_token; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_header_value_connection_2: + s_n_llhttp__internal__n_header_value_connection_2: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_header_value_connection_2; + } + match_seq = llparse__match_sequence_to_lower(state, p, endp, llparse_blob4, 9); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + goto s_n_llhttp__internal__n_invoke_update_header_state_6; + } + case kMatchPause: { + return s_n_llhttp__internal__n_header_value_connection_2; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_header_value_connection_token; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_header_value_connection_3: + s_n_llhttp__internal__n_header_value_connection_3: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_header_value_connection_3; + } + match_seq = llparse__match_sequence_to_lower(state, p, endp, llparse_blob5, 6); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + goto s_n_llhttp__internal__n_invoke_update_header_state_7; + } + case kMatchPause: { + return s_n_llhttp__internal__n_header_value_connection_3; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_header_value_connection_token; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_header_value_connection: + s_n_llhttp__internal__n_header_value_connection: { + if (p == endp) { + return s_n_llhttp__internal__n_header_value_connection; + } + switch (((*p) >= 'A' && (*p) <= 'Z' ? (*p | 0x20) : (*p))) { + case 9: { + p++; + goto s_n_llhttp__internal__n_header_value_connection; + } + case ' ': { + p++; + goto s_n_llhttp__internal__n_header_value_connection; + } + case 'c': { + p++; + goto s_n_llhttp__internal__n_header_value_connection_1; + } + case 'k': { + p++; + goto s_n_llhttp__internal__n_header_value_connection_2; + } + case 'u': { + p++; + goto s_n_llhttp__internal__n_header_value_connection_3; + } + default: { + goto s_n_llhttp__internal__n_header_value_connection_token; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_error_43: + s_n_llhttp__internal__n_error_43: { + state->error = 0xb; + state->reason = "Content-Length overflow"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_error_44: + s_n_llhttp__internal__n_error_44: { + state->error = 0xb; + state->reason = "Invalid character in Content-Length"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_header_value_content_length_ws: + s_n_llhttp__internal__n_header_value_content_length_ws: { + if (p == endp) { + return s_n_llhttp__internal__n_header_value_content_length_ws; + } + switch (*p) { + case 10: { + goto s_n_llhttp__internal__n_invoke_or_flags_15; + } + case 13: { + goto s_n_llhttp__internal__n_invoke_or_flags_15; + } + case ' ': { + p++; + goto s_n_llhttp__internal__n_header_value_content_length_ws; + } + default: { + goto s_n_llhttp__internal__n_span_end_llhttp__on_header_value_6; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_header_value_content_length: + s_n_llhttp__internal__n_header_value_content_length: { + if (p == endp) { + return s_n_llhttp__internal__n_header_value_content_length; + } + switch (*p) { + case '0': { + p++; + match = 0; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length_1; + } + case '1': { + p++; + match = 1; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length_1; + } + case '2': { + p++; + match = 2; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length_1; + } + case '3': { + p++; + match = 3; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length_1; + } + case '4': { + p++; + match = 4; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length_1; + } + case '5': { + p++; + match = 5; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length_1; + } + case '6': { + p++; + match = 6; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length_1; + } + case '7': { + p++; + match = 7; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length_1; + } + case '8': { + p++; + match = 8; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length_1; + } + case '9': { + p++; + match = 9; + goto s_n_llhttp__internal__n_invoke_mul_add_content_length_1; + } + default: { + goto s_n_llhttp__internal__n_header_value_content_length_ws; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_error_46: + s_n_llhttp__internal__n_error_46: { + state->error = 0xf; + state->reason = "Invalid `Transfer-Encoding` header value"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_error_45: + s_n_llhttp__internal__n_error_45: { + state->error = 0xf; + state->reason = "Invalid `Transfer-Encoding` header value"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_header_value_te_token_ows: + s_n_llhttp__internal__n_header_value_te_token_ows: { + if (p == endp) { + return s_n_llhttp__internal__n_header_value_te_token_ows; + } + switch (*p) { + case 9: { + p++; + goto s_n_llhttp__internal__n_header_value_te_token_ows; + } + case ' ': { + p++; + goto s_n_llhttp__internal__n_header_value_te_token_ows; + } + default: { + goto s_n_llhttp__internal__n_header_value_te_chunked; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_header_value: + s_n_llhttp__internal__n_header_value: { + static uint8_t lookup_table[] = { + 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1 + }; + if (p == endp) { + return s_n_llhttp__internal__n_header_value; + } + #ifdef __SSE4_2__ + if (endp - p >= 16) { + __m128i ranges; + __m128i input; + int avail; + int match_len; + + /* Load input */ + input = _mm_loadu_si128((__m128i const*) p); + ranges = _mm_loadu_si128((__m128i const*) llparse_blob7); + + /* Find first character that does not match `ranges` */ + match_len = _mm_cmpestri(ranges, 6, + input, 16, + _SIDD_UBYTE_OPS | _SIDD_CMP_RANGES | + _SIDD_NEGATIVE_POLARITY); + + if (match_len != 0) { + p += match_len; + goto s_n_llhttp__internal__n_header_value; + } + goto s_n_llhttp__internal__n_header_value_otherwise; + } + #endif /* __SSE4_2__ */ + switch (lookup_table[(uint8_t) *p]) { + case 1: { + p++; + goto s_n_llhttp__internal__n_header_value; + } + default: { + goto s_n_llhttp__internal__n_header_value_otherwise; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_header_value_te_token: + s_n_llhttp__internal__n_header_value_te_token: { + static uint8_t lookup_table[] = { + 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1 + }; + if (p == endp) { + return s_n_llhttp__internal__n_header_value_te_token; + } + switch (lookup_table[(uint8_t) *p]) { + case 1: { + p++; + goto s_n_llhttp__internal__n_header_value_te_token; + } + case 2: { + p++; + goto s_n_llhttp__internal__n_header_value_te_token_ows; + } + default: { + goto s_n_llhttp__internal__n_invoke_update_header_state_9; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_header_value_te_chunked_last: + s_n_llhttp__internal__n_header_value_te_chunked_last: { + if (p == endp) { + return s_n_llhttp__internal__n_header_value_te_chunked_last; + } + switch (*p) { + case 10: { + goto s_n_llhttp__internal__n_invoke_update_header_state_8; + } + case 13: { + goto s_n_llhttp__internal__n_invoke_update_header_state_8; + } + case ' ': { + p++; + goto s_n_llhttp__internal__n_header_value_te_chunked_last; + } + case ',': { + goto s_n_llhttp__internal__n_invoke_load_type_1; + } + default: { + goto s_n_llhttp__internal__n_header_value_te_token; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_header_value_te_chunked: + s_n_llhttp__internal__n_header_value_te_chunked: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_header_value_te_chunked; + } + match_seq = llparse__match_sequence_to_lower_unsafe(state, p, endp, llparse_blob6, 7); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + goto s_n_llhttp__internal__n_header_value_te_chunked_last; + } + case kMatchPause: { + return s_n_llhttp__internal__n_header_value_te_chunked; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_header_value_te_token; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_span_start_llhttp__on_header_value_1: + s_n_llhttp__internal__n_span_start_llhttp__on_header_value_1: { + if (p == endp) { + return s_n_llhttp__internal__n_span_start_llhttp__on_header_value_1; + } + state->_span_pos0 = (void*) p; + state->_span_cb0 = llhttp__on_header_value; + goto s_n_llhttp__internal__n_invoke_load_header_state_2; + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_header_value_discard_ws: + s_n_llhttp__internal__n_header_value_discard_ws: { + if (p == endp) { + return s_n_llhttp__internal__n_header_value_discard_ws; + } + switch (*p) { + case 9: { + p++; + goto s_n_llhttp__internal__n_header_value_discard_ws; + } + case 10: { + p++; + goto s_n_llhttp__internal__n_invoke_test_lenient_flags_9; + } + case 13: { + p++; + goto s_n_llhttp__internal__n_header_value_discard_ws_almost_done; + } + case ' ': { + p++; + goto s_n_llhttp__internal__n_header_value_discard_ws; + } + default: { + goto s_n_llhttp__internal__n_span_start_llhttp__on_header_value_1; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_invoke_llhttp__on_header_field_complete: + s_n_llhttp__internal__n_invoke_llhttp__on_header_field_complete: { + switch (llhttp__on_header_field_complete(state, p, endp)) { + case 0: + goto s_n_llhttp__internal__n_header_value_discard_ws; + case 21: + goto s_n_llhttp__internal__n_pause_15; + default: + goto s_n_llhttp__internal__n_error_34; + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_header_field_general_otherwise: + s_n_llhttp__internal__n_header_field_general_otherwise: { + if (p == endp) { + return s_n_llhttp__internal__n_header_field_general_otherwise; + } + switch (*p) { + case ':': { + goto s_n_llhttp__internal__n_span_end_llhttp__on_header_field_2; + } + default: { + goto s_n_llhttp__internal__n_error_47; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_header_field_general: + s_n_llhttp__internal__n_header_field_general: { + static uint8_t lookup_table[] = { + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 1, 0, 1, 1, 1, 1, 1, 0, 0, 1, 1, 0, 1, 1, 0, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, + 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, + 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 0, 1, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 + }; + if (p == endp) { + return s_n_llhttp__internal__n_header_field_general; + } + #ifdef __SSE4_2__ + if (endp - p >= 16) { + __m128i ranges; + __m128i input; + int avail; + int match_len; + + /* Load input */ + input = _mm_loadu_si128((__m128i const*) p); + ranges = _mm_loadu_si128((__m128i const*) llparse_blob8); + + /* Find first character that does not match `ranges` */ + match_len = _mm_cmpestri(ranges, 16, + input, 16, + _SIDD_UBYTE_OPS | _SIDD_CMP_RANGES | + _SIDD_NEGATIVE_POLARITY); + + if (match_len != 0) { + p += match_len; + goto s_n_llhttp__internal__n_header_field_general; + } + ranges = _mm_loadu_si128((__m128i const*) llparse_blob9); + + /* Find first character that does not match `ranges` */ + match_len = _mm_cmpestri(ranges, 2, + input, 16, + _SIDD_UBYTE_OPS | _SIDD_CMP_RANGES | + _SIDD_NEGATIVE_POLARITY); + + if (match_len != 0) { + p += match_len; + goto s_n_llhttp__internal__n_header_field_general; + } + goto s_n_llhttp__internal__n_header_field_general_otherwise; + } + #endif /* __SSE4_2__ */ + switch (lookup_table[(uint8_t) *p]) { + case 1: { + p++; + goto s_n_llhttp__internal__n_header_field_general; + } + default: { + goto s_n_llhttp__internal__n_header_field_general_otherwise; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_header_field_colon: + s_n_llhttp__internal__n_header_field_colon: { + if (p == endp) { + return s_n_llhttp__internal__n_header_field_colon; + } + switch (*p) { + case ' ': { + goto s_n_llhttp__internal__n_invoke_test_lenient_flags_8; + } + case ':': { + goto s_n_llhttp__internal__n_span_end_llhttp__on_header_field_1; + } + default: { + goto s_n_llhttp__internal__n_invoke_update_header_state_10; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_header_field_3: + s_n_llhttp__internal__n_header_field_3: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_header_field_3; + } + match_seq = llparse__match_sequence_to_lower(state, p, endp, llparse_blob2, 6); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + match = 1; + goto s_n_llhttp__internal__n_invoke_store_header_state; + } + case kMatchPause: { + return s_n_llhttp__internal__n_header_field_3; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_invoke_update_header_state_11; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_header_field_4: + s_n_llhttp__internal__n_header_field_4: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_header_field_4; + } + match_seq = llparse__match_sequence_to_lower(state, p, endp, llparse_blob10, 10); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + match = 2; + goto s_n_llhttp__internal__n_invoke_store_header_state; + } + case kMatchPause: { + return s_n_llhttp__internal__n_header_field_4; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_invoke_update_header_state_11; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_header_field_2: + s_n_llhttp__internal__n_header_field_2: { + if (p == endp) { + return s_n_llhttp__internal__n_header_field_2; + } + switch (((*p) >= 'A' && (*p) <= 'Z' ? (*p | 0x20) : (*p))) { + case 'n': { + p++; + goto s_n_llhttp__internal__n_header_field_3; + } + case 't': { + p++; + goto s_n_llhttp__internal__n_header_field_4; + } + default: { + goto s_n_llhttp__internal__n_invoke_update_header_state_11; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_header_field_1: + s_n_llhttp__internal__n_header_field_1: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_header_field_1; + } + match_seq = llparse__match_sequence_to_lower(state, p, endp, llparse_blob1, 2); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + goto s_n_llhttp__internal__n_header_field_2; + } + case kMatchPause: { + return s_n_llhttp__internal__n_header_field_1; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_invoke_update_header_state_11; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_header_field_5: + s_n_llhttp__internal__n_header_field_5: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_header_field_5; + } + match_seq = llparse__match_sequence_to_lower(state, p, endp, llparse_blob11, 15); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + match = 1; + goto s_n_llhttp__internal__n_invoke_store_header_state; + } + case kMatchPause: { + return s_n_llhttp__internal__n_header_field_5; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_invoke_update_header_state_11; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_header_field_6: + s_n_llhttp__internal__n_header_field_6: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_header_field_6; + } + match_seq = llparse__match_sequence_to_lower(state, p, endp, llparse_blob12, 16); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + match = 3; + goto s_n_llhttp__internal__n_invoke_store_header_state; + } + case kMatchPause: { + return s_n_llhttp__internal__n_header_field_6; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_invoke_update_header_state_11; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_header_field_7: + s_n_llhttp__internal__n_header_field_7: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_header_field_7; + } + match_seq = llparse__match_sequence_to_lower(state, p, endp, llparse_blob13, 6); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + match = 4; + goto s_n_llhttp__internal__n_invoke_store_header_state; + } + case kMatchPause: { + return s_n_llhttp__internal__n_header_field_7; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_invoke_update_header_state_11; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_header_field: + s_n_llhttp__internal__n_header_field: { + if (p == endp) { + return s_n_llhttp__internal__n_header_field; + } + switch (((*p) >= 'A' && (*p) <= 'Z' ? (*p | 0x20) : (*p))) { + case 'c': { + p++; + goto s_n_llhttp__internal__n_header_field_1; + } + case 'p': { + p++; + goto s_n_llhttp__internal__n_header_field_5; + } + case 't': { + p++; + goto s_n_llhttp__internal__n_header_field_6; + } + case 'u': { + p++; + goto s_n_llhttp__internal__n_header_field_7; + } + default: { + goto s_n_llhttp__internal__n_invoke_update_header_state_11; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_span_start_llhttp__on_header_field: + s_n_llhttp__internal__n_span_start_llhttp__on_header_field: { + if (p == endp) { + return s_n_llhttp__internal__n_span_start_llhttp__on_header_field; + } + state->_span_pos0 = (void*) p; + state->_span_cb0 = llhttp__on_header_field; + goto s_n_llhttp__internal__n_header_field; + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_header_field_start: + s_n_llhttp__internal__n_header_field_start: { + if (p == endp) { + return s_n_llhttp__internal__n_header_field_start; + } + switch (*p) { + case 10: { + p++; + goto s_n_llhttp__internal__n_invoke_test_lenient_flags_1; + } + case 13: { + p++; + goto s_n_llhttp__internal__n_headers_almost_done; + } + default: { + goto s_n_llhttp__internal__n_span_start_llhttp__on_header_field; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_headers_start: + s_n_llhttp__internal__n_headers_start: { + if (p == endp) { + return s_n_llhttp__internal__n_headers_start; + } + switch (*p) { + case ' ': { + p++; + goto s_n_llhttp__internal__n_invoke_test_lenient_flags; + } + default: { + goto s_n_llhttp__internal__n_header_field_start; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_url_to_http_09: + s_n_llhttp__internal__n_url_to_http_09: { + if (p == endp) { + return s_n_llhttp__internal__n_url_to_http_09; + } + switch (*p) { + case 9: { + p++; + goto s_n_llhttp__internal__n_error_2; + } + case 12: { + p++; + goto s_n_llhttp__internal__n_error_2; + } + default: { + goto s_n_llhttp__internal__n_invoke_update_http_major; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_url_skip_to_http09: + s_n_llhttp__internal__n_url_skip_to_http09: { + if (p == endp) { + return s_n_llhttp__internal__n_url_skip_to_http09; + } + switch (*p) { + case 9: { + p++; + goto s_n_llhttp__internal__n_error_2; + } + case 12: { + p++; + goto s_n_llhttp__internal__n_error_2; + } + default: { + p++; + goto s_n_llhttp__internal__n_url_to_http_09; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_url_skip_lf_to_http09_1: + s_n_llhttp__internal__n_url_skip_lf_to_http09_1: { + if (p == endp) { + return s_n_llhttp__internal__n_url_skip_lf_to_http09_1; + } + switch (*p) { + case 10: { + p++; + goto s_n_llhttp__internal__n_url_to_http_09; + } + default: { + goto s_n_llhttp__internal__n_error_48; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_url_skip_lf_to_http09: + s_n_llhttp__internal__n_url_skip_lf_to_http09: { + if (p == endp) { + return s_n_llhttp__internal__n_url_skip_lf_to_http09; + } + switch (*p) { + case 9: { + p++; + goto s_n_llhttp__internal__n_error_2; + } + case 12: { + p++; + goto s_n_llhttp__internal__n_error_2; + } + case 13: { + p++; + goto s_n_llhttp__internal__n_url_skip_lf_to_http09_1; + } + default: { + goto s_n_llhttp__internal__n_error_48; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_req_pri_upgrade: + s_n_llhttp__internal__n_req_pri_upgrade: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_req_pri_upgrade; + } + match_seq = llparse__match_sequence_id(state, p, endp, llparse_blob15, 10); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + goto s_n_llhttp__internal__n_error_55; + } + case kMatchPause: { + return s_n_llhttp__internal__n_req_pri_upgrade; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_error_56; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_req_http_complete_crlf: + s_n_llhttp__internal__n_req_http_complete_crlf: { + if (p == endp) { + return s_n_llhttp__internal__n_req_http_complete_crlf; + } + switch (*p) { + case 10: { + p++; + goto s_n_llhttp__internal__n_headers_start; + } + default: { + goto s_n_llhttp__internal__n_invoke_test_lenient_flags_16; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_req_http_complete: + s_n_llhttp__internal__n_req_http_complete: { + if (p == endp) { + return s_n_llhttp__internal__n_req_http_complete; + } + switch (*p) { + case 13: { + p++; + goto s_n_llhttp__internal__n_req_http_complete_crlf; + } + default: { + goto s_n_llhttp__internal__n_error_54; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_invoke_load_method_1: + s_n_llhttp__internal__n_invoke_load_method_1: { + switch (llhttp__internal__c_load_method(state, p, endp)) { + case 34: + goto s_n_llhttp__internal__n_req_pri_upgrade; + default: + goto s_n_llhttp__internal__n_req_http_complete; + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_invoke_llhttp__on_version_complete: + s_n_llhttp__internal__n_invoke_llhttp__on_version_complete: { + switch (llhttp__on_version_complete(state, p, endp)) { + case 0: + goto s_n_llhttp__internal__n_invoke_load_method_1; + case 21: + goto s_n_llhttp__internal__n_pause_17; + default: + goto s_n_llhttp__internal__n_error_52; + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_error_51: + s_n_llhttp__internal__n_error_51: { + state->error = 0x9; + state->reason = "Invalid HTTP version"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_error_57: + s_n_llhttp__internal__n_error_57: { + state->error = 0x9; + state->reason = "Invalid minor version"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_req_http_minor: + s_n_llhttp__internal__n_req_http_minor: { + if (p == endp) { + return s_n_llhttp__internal__n_req_http_minor; + } + switch (*p) { + case '0': { + p++; + match = 0; + goto s_n_llhttp__internal__n_invoke_store_http_minor; + } + case '1': { + p++; + match = 1; + goto s_n_llhttp__internal__n_invoke_store_http_minor; + } + case '2': { + p++; + match = 2; + goto s_n_llhttp__internal__n_invoke_store_http_minor; + } + case '3': { + p++; + match = 3; + goto s_n_llhttp__internal__n_invoke_store_http_minor; + } + case '4': { + p++; + match = 4; + goto s_n_llhttp__internal__n_invoke_store_http_minor; + } + case '5': { + p++; + match = 5; + goto s_n_llhttp__internal__n_invoke_store_http_minor; + } + case '6': { + p++; + match = 6; + goto s_n_llhttp__internal__n_invoke_store_http_minor; + } + case '7': { + p++; + match = 7; + goto s_n_llhttp__internal__n_invoke_store_http_minor; + } + case '8': { + p++; + match = 8; + goto s_n_llhttp__internal__n_invoke_store_http_minor; + } + case '9': { + p++; + match = 9; + goto s_n_llhttp__internal__n_invoke_store_http_minor; + } + default: { + goto s_n_llhttp__internal__n_span_end_llhttp__on_version_2; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_error_58: + s_n_llhttp__internal__n_error_58: { + state->error = 0x9; + state->reason = "Expected dot"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_req_http_dot: + s_n_llhttp__internal__n_req_http_dot: { + if (p == endp) { + return s_n_llhttp__internal__n_req_http_dot; + } + switch (*p) { + case '.': { + p++; + goto s_n_llhttp__internal__n_req_http_minor; + } + default: { + goto s_n_llhttp__internal__n_span_end_llhttp__on_version_3; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_error_59: + s_n_llhttp__internal__n_error_59: { + state->error = 0x9; + state->reason = "Invalid major version"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_req_http_major: + s_n_llhttp__internal__n_req_http_major: { + if (p == endp) { + return s_n_llhttp__internal__n_req_http_major; + } + switch (*p) { + case '0': { + p++; + match = 0; + goto s_n_llhttp__internal__n_invoke_store_http_major; + } + case '1': { + p++; + match = 1; + goto s_n_llhttp__internal__n_invoke_store_http_major; + } + case '2': { + p++; + match = 2; + goto s_n_llhttp__internal__n_invoke_store_http_major; + } + case '3': { + p++; + match = 3; + goto s_n_llhttp__internal__n_invoke_store_http_major; + } + case '4': { + p++; + match = 4; + goto s_n_llhttp__internal__n_invoke_store_http_major; + } + case '5': { + p++; + match = 5; + goto s_n_llhttp__internal__n_invoke_store_http_major; + } + case '6': { + p++; + match = 6; + goto s_n_llhttp__internal__n_invoke_store_http_major; + } + case '7': { + p++; + match = 7; + goto s_n_llhttp__internal__n_invoke_store_http_major; + } + case '8': { + p++; + match = 8; + goto s_n_llhttp__internal__n_invoke_store_http_major; + } + case '9': { + p++; + match = 9; + goto s_n_llhttp__internal__n_invoke_store_http_major; + } + default: { + goto s_n_llhttp__internal__n_span_end_llhttp__on_version_4; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_span_start_llhttp__on_version: + s_n_llhttp__internal__n_span_start_llhttp__on_version: { + if (p == endp) { + return s_n_llhttp__internal__n_span_start_llhttp__on_version; + } + state->_span_pos0 = (void*) p; + state->_span_cb0 = llhttp__on_version; + goto s_n_llhttp__internal__n_req_http_major; + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_req_http_start_1: + s_n_llhttp__internal__n_req_http_start_1: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_req_http_start_1; + } + match_seq = llparse__match_sequence_id(state, p, endp, llparse_blob14, 4); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + goto s_n_llhttp__internal__n_invoke_load_method; + } + case kMatchPause: { + return s_n_llhttp__internal__n_req_http_start_1; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_error_62; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_req_http_start_2: + s_n_llhttp__internal__n_req_http_start_2: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_req_http_start_2; + } + match_seq = llparse__match_sequence_id(state, p, endp, llparse_blob16, 3); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + goto s_n_llhttp__internal__n_invoke_load_method_2; + } + case kMatchPause: { + return s_n_llhttp__internal__n_req_http_start_2; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_error_62; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_req_http_start_3: + s_n_llhttp__internal__n_req_http_start_3: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_req_http_start_3; + } + match_seq = llparse__match_sequence_id(state, p, endp, llparse_blob17, 4); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + goto s_n_llhttp__internal__n_invoke_load_method_3; + } + case kMatchPause: { + return s_n_llhttp__internal__n_req_http_start_3; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_error_62; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_req_http_start: + s_n_llhttp__internal__n_req_http_start: { + if (p == endp) { + return s_n_llhttp__internal__n_req_http_start; + } + switch (*p) { + case ' ': { + p++; + goto s_n_llhttp__internal__n_req_http_start; + } + case 'H': { + p++; + goto s_n_llhttp__internal__n_req_http_start_1; + } + case 'I': { + p++; + goto s_n_llhttp__internal__n_req_http_start_2; + } + case 'R': { + p++; + goto s_n_llhttp__internal__n_req_http_start_3; + } + default: { + goto s_n_llhttp__internal__n_error_62; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_url_to_http: + s_n_llhttp__internal__n_url_to_http: { + if (p == endp) { + return s_n_llhttp__internal__n_url_to_http; + } + switch (*p) { + case 9: { + p++; + goto s_n_llhttp__internal__n_error_2; + } + case 12: { + p++; + goto s_n_llhttp__internal__n_error_2; + } + default: { + goto s_n_llhttp__internal__n_invoke_llhttp__on_url_complete_1; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_url_skip_to_http: + s_n_llhttp__internal__n_url_skip_to_http: { + if (p == endp) { + return s_n_llhttp__internal__n_url_skip_to_http; + } + switch (*p) { + case 9: { + p++; + goto s_n_llhttp__internal__n_error_2; + } + case 12: { + p++; + goto s_n_llhttp__internal__n_error_2; + } + default: { + p++; + goto s_n_llhttp__internal__n_url_to_http; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_url_fragment: + s_n_llhttp__internal__n_url_fragment: { + static uint8_t lookup_table[] = { + 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 2, 0, 1, 3, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 4, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, + 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, + 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, + 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, + 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, + 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 + }; + if (p == endp) { + return s_n_llhttp__internal__n_url_fragment; + } + switch (lookup_table[(uint8_t) *p]) { + case 1: { + p++; + goto s_n_llhttp__internal__n_error_2; + } + case 2: { + goto s_n_llhttp__internal__n_span_end_llhttp__on_url_6; + } + case 3: { + goto s_n_llhttp__internal__n_span_end_llhttp__on_url_7; + } + case 4: { + goto s_n_llhttp__internal__n_span_end_llhttp__on_url_8; + } + case 5: { + p++; + goto s_n_llhttp__internal__n_url_fragment; + } + default: { + goto s_n_llhttp__internal__n_error_63; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_span_end_stub_query_3: + s_n_llhttp__internal__n_span_end_stub_query_3: { + if (p == endp) { + return s_n_llhttp__internal__n_span_end_stub_query_3; + } + p++; + goto s_n_llhttp__internal__n_url_fragment; + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_url_query: + s_n_llhttp__internal__n_url_query: { + static uint8_t lookup_table[] = { + 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 2, 0, 1, 3, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 4, 5, 5, 6, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, + 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, + 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, + 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, + 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, + 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 + }; + if (p == endp) { + return s_n_llhttp__internal__n_url_query; + } + switch (lookup_table[(uint8_t) *p]) { + case 1: { + p++; + goto s_n_llhttp__internal__n_error_2; + } + case 2: { + goto s_n_llhttp__internal__n_span_end_llhttp__on_url_9; + } + case 3: { + goto s_n_llhttp__internal__n_span_end_llhttp__on_url_10; + } + case 4: { + goto s_n_llhttp__internal__n_span_end_llhttp__on_url_11; + } + case 5: { + p++; + goto s_n_llhttp__internal__n_url_query; + } + case 6: { + goto s_n_llhttp__internal__n_span_end_stub_query_3; + } + default: { + goto s_n_llhttp__internal__n_error_64; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_url_query_or_fragment: + s_n_llhttp__internal__n_url_query_or_fragment: { + if (p == endp) { + return s_n_llhttp__internal__n_url_query_or_fragment; + } + switch (*p) { + case 9: { + p++; + goto s_n_llhttp__internal__n_error_2; + } + case 10: { + goto s_n_llhttp__internal__n_span_end_llhttp__on_url_3; + } + case 12: { + p++; + goto s_n_llhttp__internal__n_error_2; + } + case 13: { + goto s_n_llhttp__internal__n_span_end_llhttp__on_url_4; + } + case ' ': { + goto s_n_llhttp__internal__n_span_end_llhttp__on_url_5; + } + case '#': { + p++; + goto s_n_llhttp__internal__n_url_fragment; + } + case '?': { + p++; + goto s_n_llhttp__internal__n_url_query; + } + default: { + goto s_n_llhttp__internal__n_error_65; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_url_path: + s_n_llhttp__internal__n_url_path: { + static uint8_t lookup_table[] = { + 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 1, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 2, 2, 0, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 0, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, + 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 + }; + if (p == endp) { + return s_n_llhttp__internal__n_url_path; + } + switch (lookup_table[(uint8_t) *p]) { + case 1: { + p++; + goto s_n_llhttp__internal__n_error_2; + } + case 2: { + p++; + goto s_n_llhttp__internal__n_url_path; + } + default: { + goto s_n_llhttp__internal__n_url_query_or_fragment; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_span_start_stub_path_2: + s_n_llhttp__internal__n_span_start_stub_path_2: { + if (p == endp) { + return s_n_llhttp__internal__n_span_start_stub_path_2; + } + p++; + goto s_n_llhttp__internal__n_url_path; + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_span_start_stub_path: + s_n_llhttp__internal__n_span_start_stub_path: { + if (p == endp) { + return s_n_llhttp__internal__n_span_start_stub_path; + } + p++; + goto s_n_llhttp__internal__n_url_path; + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_span_start_stub_path_1: + s_n_llhttp__internal__n_span_start_stub_path_1: { + if (p == endp) { + return s_n_llhttp__internal__n_span_start_stub_path_1; + } + p++; + goto s_n_llhttp__internal__n_url_path; + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_url_server_with_at: + s_n_llhttp__internal__n_url_server_with_at: { + static uint8_t lookup_table[] = { + 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 2, 0, 1, 3, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 4, 5, 0, 0, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 6, + 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 0, 5, 0, 7, + 8, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, + 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 0, 5, 0, 5, + 0, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, + 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 0, 0, 0, 5, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 + }; + if (p == endp) { + return s_n_llhttp__internal__n_url_server_with_at; + } + switch (lookup_table[(uint8_t) *p]) { + case 1: { + p++; + goto s_n_llhttp__internal__n_error_2; + } + case 2: { + goto s_n_llhttp__internal__n_span_end_llhttp__on_url_12; + } + case 3: { + goto s_n_llhttp__internal__n_span_end_llhttp__on_url_13; + } + case 4: { + goto s_n_llhttp__internal__n_span_end_llhttp__on_url_14; + } + case 5: { + p++; + goto s_n_llhttp__internal__n_url_server; + } + case 6: { + goto s_n_llhttp__internal__n_span_start_stub_path_1; + } + case 7: { + p++; + goto s_n_llhttp__internal__n_url_query; + } + case 8: { + p++; + goto s_n_llhttp__internal__n_error_66; + } + default: { + goto s_n_llhttp__internal__n_error_67; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_url_server: + s_n_llhttp__internal__n_url_server: { + static uint8_t lookup_table[] = { + 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 2, 0, 1, 3, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 4, 5, 0, 0, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 6, + 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 0, 5, 0, 7, + 8, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, + 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 0, 5, 0, 5, + 0, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, + 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 0, 0, 0, 5, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 + }; + if (p == endp) { + return s_n_llhttp__internal__n_url_server; + } + switch (lookup_table[(uint8_t) *p]) { + case 1: { + p++; + goto s_n_llhttp__internal__n_error_2; + } + case 2: { + goto s_n_llhttp__internal__n_span_end_llhttp__on_url; + } + case 3: { + goto s_n_llhttp__internal__n_span_end_llhttp__on_url_1; + } + case 4: { + goto s_n_llhttp__internal__n_span_end_llhttp__on_url_2; + } + case 5: { + p++; + goto s_n_llhttp__internal__n_url_server; + } + case 6: { + goto s_n_llhttp__internal__n_span_start_stub_path; + } + case 7: { + p++; + goto s_n_llhttp__internal__n_url_query; + } + case 8: { + p++; + goto s_n_llhttp__internal__n_url_server_with_at; + } + default: { + goto s_n_llhttp__internal__n_error_68; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_url_schema_delim_1: + s_n_llhttp__internal__n_url_schema_delim_1: { + if (p == endp) { + return s_n_llhttp__internal__n_url_schema_delim_1; + } + switch (*p) { + case '/': { + p++; + goto s_n_llhttp__internal__n_url_server; + } + default: { + goto s_n_llhttp__internal__n_error_69; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_url_schema_delim: + s_n_llhttp__internal__n_url_schema_delim: { + if (p == endp) { + return s_n_llhttp__internal__n_url_schema_delim; + } + switch (*p) { + case 9: { + p++; + goto s_n_llhttp__internal__n_error_2; + } + case 10: { + p++; + goto s_n_llhttp__internal__n_error_2; + } + case 12: { + p++; + goto s_n_llhttp__internal__n_error_2; + } + case 13: { + p++; + goto s_n_llhttp__internal__n_error_2; + } + case ' ': { + p++; + goto s_n_llhttp__internal__n_error_2; + } + case '/': { + p++; + goto s_n_llhttp__internal__n_url_schema_delim_1; + } + default: { + goto s_n_llhttp__internal__n_error_69; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_span_end_stub_schema: + s_n_llhttp__internal__n_span_end_stub_schema: { + if (p == endp) { + return s_n_llhttp__internal__n_span_end_stub_schema; + } + p++; + goto s_n_llhttp__internal__n_url_schema_delim; + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_url_schema: + s_n_llhttp__internal__n_url_schema: { + static uint8_t lookup_table[] = { + 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 1, 1, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 2, 0, 0, 0, 0, 0, + 0, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, + 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 0, 0, 0, 0, 0, + 0, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, + 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 + }; + if (p == endp) { + return s_n_llhttp__internal__n_url_schema; + } + switch (lookup_table[(uint8_t) *p]) { + case 1: { + p++; + goto s_n_llhttp__internal__n_error_2; + } + case 2: { + goto s_n_llhttp__internal__n_span_end_stub_schema; + } + case 3: { + p++; + goto s_n_llhttp__internal__n_url_schema; + } + default: { + goto s_n_llhttp__internal__n_error_70; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_url_start: + s_n_llhttp__internal__n_url_start: { + static uint8_t lookup_table[] = { + 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 1, 1, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 2, 0, 0, 0, 0, 2, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, + 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 0, 0, 0, 0, 0, + 0, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, + 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 + }; + if (p == endp) { + return s_n_llhttp__internal__n_url_start; + } + switch (lookup_table[(uint8_t) *p]) { + case 1: { + p++; + goto s_n_llhttp__internal__n_error_2; + } + case 2: { + goto s_n_llhttp__internal__n_span_start_stub_path_2; + } + case 3: { + goto s_n_llhttp__internal__n_url_schema; + } + default: { + goto s_n_llhttp__internal__n_error_71; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_span_start_llhttp__on_url_1: + s_n_llhttp__internal__n_span_start_llhttp__on_url_1: { + if (p == endp) { + return s_n_llhttp__internal__n_span_start_llhttp__on_url_1; + } + state->_span_pos0 = (void*) p; + state->_span_cb0 = llhttp__on_url; + goto s_n_llhttp__internal__n_url_start; + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_url_entry_normal: + s_n_llhttp__internal__n_url_entry_normal: { + if (p == endp) { + return s_n_llhttp__internal__n_url_entry_normal; + } + switch (*p) { + case 9: { + p++; + goto s_n_llhttp__internal__n_error_2; + } + case 12: { + p++; + goto s_n_llhttp__internal__n_error_2; + } + default: { + goto s_n_llhttp__internal__n_span_start_llhttp__on_url_1; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_span_start_llhttp__on_url: + s_n_llhttp__internal__n_span_start_llhttp__on_url: { + if (p == endp) { + return s_n_llhttp__internal__n_span_start_llhttp__on_url; + } + state->_span_pos0 = (void*) p; + state->_span_cb0 = llhttp__on_url; + goto s_n_llhttp__internal__n_url_server; + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_url_entry_connect: + s_n_llhttp__internal__n_url_entry_connect: { + if (p == endp) { + return s_n_llhttp__internal__n_url_entry_connect; + } + switch (*p) { + case 9: { + p++; + goto s_n_llhttp__internal__n_error_2; + } + case 12: { + p++; + goto s_n_llhttp__internal__n_error_2; + } + default: { + goto s_n_llhttp__internal__n_span_start_llhttp__on_url; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_req_spaces_before_url: + s_n_llhttp__internal__n_req_spaces_before_url: { + if (p == endp) { + return s_n_llhttp__internal__n_req_spaces_before_url; + } + switch (*p) { + case ' ': { + p++; + goto s_n_llhttp__internal__n_req_spaces_before_url; + } + default: { + goto s_n_llhttp__internal__n_invoke_is_equal_method; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_req_first_space_before_url: + s_n_llhttp__internal__n_req_first_space_before_url: { + if (p == endp) { + return s_n_llhttp__internal__n_req_first_space_before_url; + } + switch (*p) { + case ' ': { + p++; + goto s_n_llhttp__internal__n_req_spaces_before_url; + } + default: { + goto s_n_llhttp__internal__n_error_72; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_invoke_llhttp__on_method_complete_1: + s_n_llhttp__internal__n_invoke_llhttp__on_method_complete_1: { + switch (llhttp__on_method_complete(state, p, endp)) { + case 0: + goto s_n_llhttp__internal__n_req_first_space_before_url; + case 21: + goto s_n_llhttp__internal__n_pause_22; + default: + goto s_n_llhttp__internal__n_error_89; + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_2: + s_n_llhttp__internal__n_after_start_req_2: { + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_2; + } + switch (*p) { + case 'L': { + p++; + match = 19; + goto s_n_llhttp__internal__n_invoke_store_method_1; + } + default: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_3: + s_n_llhttp__internal__n_after_start_req_3: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_3; + } + match_seq = llparse__match_sequence_id(state, p, endp, llparse_blob18, 6); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + match = 36; + goto s_n_llhttp__internal__n_invoke_store_method_1; + } + case kMatchPause: { + return s_n_llhttp__internal__n_after_start_req_3; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_1: + s_n_llhttp__internal__n_after_start_req_1: { + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_1; + } + switch (*p) { + case 'C': { + p++; + goto s_n_llhttp__internal__n_after_start_req_2; + } + case 'N': { + p++; + goto s_n_llhttp__internal__n_after_start_req_3; + } + default: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_4: + s_n_llhttp__internal__n_after_start_req_4: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_4; + } + match_seq = llparse__match_sequence_id(state, p, endp, llparse_blob19, 3); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + match = 16; + goto s_n_llhttp__internal__n_invoke_store_method_1; + } + case kMatchPause: { + return s_n_llhttp__internal__n_after_start_req_4; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_6: + s_n_llhttp__internal__n_after_start_req_6: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_6; + } + match_seq = llparse__match_sequence_id(state, p, endp, llparse_blob20, 6); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + match = 22; + goto s_n_llhttp__internal__n_invoke_store_method_1; + } + case kMatchPause: { + return s_n_llhttp__internal__n_after_start_req_6; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_8: + s_n_llhttp__internal__n_after_start_req_8: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_8; + } + match_seq = llparse__match_sequence_id(state, p, endp, llparse_blob21, 4); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + match = 5; + goto s_n_llhttp__internal__n_invoke_store_method_1; + } + case kMatchPause: { + return s_n_llhttp__internal__n_after_start_req_8; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_9: + s_n_llhttp__internal__n_after_start_req_9: { + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_9; + } + switch (*p) { + case 'Y': { + p++; + match = 8; + goto s_n_llhttp__internal__n_invoke_store_method_1; + } + default: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_7: + s_n_llhttp__internal__n_after_start_req_7: { + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_7; + } + switch (*p) { + case 'N': { + p++; + goto s_n_llhttp__internal__n_after_start_req_8; + } + case 'P': { + p++; + goto s_n_llhttp__internal__n_after_start_req_9; + } + default: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_5: + s_n_llhttp__internal__n_after_start_req_5: { + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_5; + } + switch (*p) { + case 'H': { + p++; + goto s_n_llhttp__internal__n_after_start_req_6; + } + case 'O': { + p++; + goto s_n_llhttp__internal__n_after_start_req_7; + } + default: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_12: + s_n_llhttp__internal__n_after_start_req_12: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_12; + } + match_seq = llparse__match_sequence_id(state, p, endp, llparse_blob22, 3); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + match = 0; + goto s_n_llhttp__internal__n_invoke_store_method_1; + } + case kMatchPause: { + return s_n_llhttp__internal__n_after_start_req_12; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_13: + s_n_llhttp__internal__n_after_start_req_13: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_13; + } + match_seq = llparse__match_sequence_id(state, p, endp, llparse_blob23, 5); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + match = 35; + goto s_n_llhttp__internal__n_invoke_store_method_1; + } + case kMatchPause: { + return s_n_llhttp__internal__n_after_start_req_13; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_11: + s_n_llhttp__internal__n_after_start_req_11: { + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_11; + } + switch (*p) { + case 'L': { + p++; + goto s_n_llhttp__internal__n_after_start_req_12; + } + case 'S': { + p++; + goto s_n_llhttp__internal__n_after_start_req_13; + } + default: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_10: + s_n_llhttp__internal__n_after_start_req_10: { + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_10; + } + switch (*p) { + case 'E': { + p++; + goto s_n_llhttp__internal__n_after_start_req_11; + } + default: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_14: + s_n_llhttp__internal__n_after_start_req_14: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_14; + } + match_seq = llparse__match_sequence_id(state, p, endp, llparse_blob24, 4); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + match = 45; + goto s_n_llhttp__internal__n_invoke_store_method_1; + } + case kMatchPause: { + return s_n_llhttp__internal__n_after_start_req_14; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_17: + s_n_llhttp__internal__n_after_start_req_17: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_17; + } + match_seq = llparse__match_sequence_id(state, p, endp, llparse_blob26, 9); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + match = 41; + goto s_n_llhttp__internal__n_invoke_store_method_1; + } + case kMatchPause: { + return s_n_llhttp__internal__n_after_start_req_17; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_16: + s_n_llhttp__internal__n_after_start_req_16: { + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_16; + } + switch (*p) { + case '_': { + p++; + goto s_n_llhttp__internal__n_after_start_req_17; + } + default: { + match = 1; + goto s_n_llhttp__internal__n_invoke_store_method_1; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_15: + s_n_llhttp__internal__n_after_start_req_15: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_15; + } + match_seq = llparse__match_sequence_id(state, p, endp, llparse_blob25, 2); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + goto s_n_llhttp__internal__n_after_start_req_16; + } + case kMatchPause: { + return s_n_llhttp__internal__n_after_start_req_15; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_18: + s_n_llhttp__internal__n_after_start_req_18: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_18; + } + match_seq = llparse__match_sequence_id(state, p, endp, llparse_blob27, 3); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + match = 2; + goto s_n_llhttp__internal__n_invoke_store_method_1; + } + case kMatchPause: { + return s_n_llhttp__internal__n_after_start_req_18; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_20: + s_n_llhttp__internal__n_after_start_req_20: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_20; + } + match_seq = llparse__match_sequence_id(state, p, endp, llparse_blob28, 2); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + match = 31; + goto s_n_llhttp__internal__n_invoke_store_method_1; + } + case kMatchPause: { + return s_n_llhttp__internal__n_after_start_req_20; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_21: + s_n_llhttp__internal__n_after_start_req_21: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_21; + } + match_seq = llparse__match_sequence_id(state, p, endp, llparse_blob29, 2); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + match = 9; + goto s_n_llhttp__internal__n_invoke_store_method_1; + } + case kMatchPause: { + return s_n_llhttp__internal__n_after_start_req_21; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_19: + s_n_llhttp__internal__n_after_start_req_19: { + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_19; + } + switch (*p) { + case 'I': { + p++; + goto s_n_llhttp__internal__n_after_start_req_20; + } + case 'O': { + p++; + goto s_n_llhttp__internal__n_after_start_req_21; + } + default: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_23: + s_n_llhttp__internal__n_after_start_req_23: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_23; + } + match_seq = llparse__match_sequence_id(state, p, endp, llparse_blob30, 6); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + match = 24; + goto s_n_llhttp__internal__n_invoke_store_method_1; + } + case kMatchPause: { + return s_n_llhttp__internal__n_after_start_req_23; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_24: + s_n_llhttp__internal__n_after_start_req_24: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_24; + } + match_seq = llparse__match_sequence_id(state, p, endp, llparse_blob31, 3); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + match = 23; + goto s_n_llhttp__internal__n_invoke_store_method_1; + } + case kMatchPause: { + return s_n_llhttp__internal__n_after_start_req_24; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_26: + s_n_llhttp__internal__n_after_start_req_26: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_26; + } + match_seq = llparse__match_sequence_id(state, p, endp, llparse_blob32, 7); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + match = 21; + goto s_n_llhttp__internal__n_invoke_store_method_1; + } + case kMatchPause: { + return s_n_llhttp__internal__n_after_start_req_26; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_28: + s_n_llhttp__internal__n_after_start_req_28: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_28; + } + match_seq = llparse__match_sequence_id(state, p, endp, llparse_blob33, 6); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + match = 30; + goto s_n_llhttp__internal__n_invoke_store_method_1; + } + case kMatchPause: { + return s_n_llhttp__internal__n_after_start_req_28; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_29: + s_n_llhttp__internal__n_after_start_req_29: { + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_29; + } + switch (*p) { + case 'L': { + p++; + match = 10; + goto s_n_llhttp__internal__n_invoke_store_method_1; + } + default: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_27: + s_n_llhttp__internal__n_after_start_req_27: { + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_27; + } + switch (*p) { + case 'A': { + p++; + goto s_n_llhttp__internal__n_after_start_req_28; + } + case 'O': { + p++; + goto s_n_llhttp__internal__n_after_start_req_29; + } + default: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_25: + s_n_llhttp__internal__n_after_start_req_25: { + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_25; + } + switch (*p) { + case 'A': { + p++; + goto s_n_llhttp__internal__n_after_start_req_26; + } + case 'C': { + p++; + goto s_n_llhttp__internal__n_after_start_req_27; + } + default: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_30: + s_n_llhttp__internal__n_after_start_req_30: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_30; + } + match_seq = llparse__match_sequence_id(state, p, endp, llparse_blob34, 2); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + match = 11; + goto s_n_llhttp__internal__n_invoke_store_method_1; + } + case kMatchPause: { + return s_n_llhttp__internal__n_after_start_req_30; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_22: + s_n_llhttp__internal__n_after_start_req_22: { + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_22; + } + switch (*p) { + case '-': { + p++; + goto s_n_llhttp__internal__n_after_start_req_23; + } + case 'E': { + p++; + goto s_n_llhttp__internal__n_after_start_req_24; + } + case 'K': { + p++; + goto s_n_llhttp__internal__n_after_start_req_25; + } + case 'O': { + p++; + goto s_n_llhttp__internal__n_after_start_req_30; + } + default: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_31: + s_n_llhttp__internal__n_after_start_req_31: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_31; + } + match_seq = llparse__match_sequence_id(state, p, endp, llparse_blob35, 5); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + match = 25; + goto s_n_llhttp__internal__n_invoke_store_method_1; + } + case kMatchPause: { + return s_n_llhttp__internal__n_after_start_req_31; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_32: + s_n_llhttp__internal__n_after_start_req_32: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_32; + } + match_seq = llparse__match_sequence_id(state, p, endp, llparse_blob36, 6); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + match = 6; + goto s_n_llhttp__internal__n_invoke_store_method_1; + } + case kMatchPause: { + return s_n_llhttp__internal__n_after_start_req_32; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_35: + s_n_llhttp__internal__n_after_start_req_35: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_35; + } + match_seq = llparse__match_sequence_id(state, p, endp, llparse_blob37, 2); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + match = 28; + goto s_n_llhttp__internal__n_invoke_store_method_1; + } + case kMatchPause: { + return s_n_llhttp__internal__n_after_start_req_35; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_36: + s_n_llhttp__internal__n_after_start_req_36: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_36; + } + match_seq = llparse__match_sequence_id(state, p, endp, llparse_blob38, 2); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + match = 39; + goto s_n_llhttp__internal__n_invoke_store_method_1; + } + case kMatchPause: { + return s_n_llhttp__internal__n_after_start_req_36; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_34: + s_n_llhttp__internal__n_after_start_req_34: { + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_34; + } + switch (*p) { + case 'T': { + p++; + goto s_n_llhttp__internal__n_after_start_req_35; + } + case 'U': { + p++; + goto s_n_llhttp__internal__n_after_start_req_36; + } + default: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_37: + s_n_llhttp__internal__n_after_start_req_37: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_37; + } + match_seq = llparse__match_sequence_id(state, p, endp, llparse_blob39, 2); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + match = 38; + goto s_n_llhttp__internal__n_invoke_store_method_1; + } + case kMatchPause: { + return s_n_llhttp__internal__n_after_start_req_37; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_38: + s_n_llhttp__internal__n_after_start_req_38: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_38; + } + match_seq = llparse__match_sequence_id(state, p, endp, llparse_blob40, 2); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + match = 3; + goto s_n_llhttp__internal__n_invoke_store_method_1; + } + case kMatchPause: { + return s_n_llhttp__internal__n_after_start_req_38; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_42: + s_n_llhttp__internal__n_after_start_req_42: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_42; + } + match_seq = llparse__match_sequence_id(state, p, endp, llparse_blob41, 3); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + match = 12; + goto s_n_llhttp__internal__n_invoke_store_method_1; + } + case kMatchPause: { + return s_n_llhttp__internal__n_after_start_req_42; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_43: + s_n_llhttp__internal__n_after_start_req_43: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_43; + } + match_seq = llparse__match_sequence_id(state, p, endp, llparse_blob42, 4); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + match = 13; + goto s_n_llhttp__internal__n_invoke_store_method_1; + } + case kMatchPause: { + return s_n_llhttp__internal__n_after_start_req_43; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_41: + s_n_llhttp__internal__n_after_start_req_41: { + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_41; + } + switch (*p) { + case 'F': { + p++; + goto s_n_llhttp__internal__n_after_start_req_42; + } + case 'P': { + p++; + goto s_n_llhttp__internal__n_after_start_req_43; + } + default: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_40: + s_n_llhttp__internal__n_after_start_req_40: { + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_40; + } + switch (*p) { + case 'P': { + p++; + goto s_n_llhttp__internal__n_after_start_req_41; + } + default: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_39: + s_n_llhttp__internal__n_after_start_req_39: { + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_39; + } + switch (*p) { + case 'I': { + p++; + match = 34; + goto s_n_llhttp__internal__n_invoke_store_method_1; + } + case 'O': { + p++; + goto s_n_llhttp__internal__n_after_start_req_40; + } + default: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_45: + s_n_llhttp__internal__n_after_start_req_45: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_45; + } + match_seq = llparse__match_sequence_id(state, p, endp, llparse_blob43, 2); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + match = 29; + goto s_n_llhttp__internal__n_invoke_store_method_1; + } + case kMatchPause: { + return s_n_llhttp__internal__n_after_start_req_45; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_44: + s_n_llhttp__internal__n_after_start_req_44: { + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_44; + } + switch (*p) { + case 'R': { + p++; + goto s_n_llhttp__internal__n_after_start_req_45; + } + case 'T': { + p++; + match = 4; + goto s_n_llhttp__internal__n_invoke_store_method_1; + } + default: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_33: + s_n_llhttp__internal__n_after_start_req_33: { + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_33; + } + switch (*p) { + case 'A': { + p++; + goto s_n_llhttp__internal__n_after_start_req_34; + } + case 'L': { + p++; + goto s_n_llhttp__internal__n_after_start_req_37; + } + case 'O': { + p++; + goto s_n_llhttp__internal__n_after_start_req_38; + } + case 'R': { + p++; + goto s_n_llhttp__internal__n_after_start_req_39; + } + case 'U': { + p++; + goto s_n_llhttp__internal__n_after_start_req_44; + } + default: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_48: + s_n_llhttp__internal__n_after_start_req_48: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_48; + } + match_seq = llparse__match_sequence_id(state, p, endp, llparse_blob44, 3); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + match = 17; + goto s_n_llhttp__internal__n_invoke_store_method_1; + } + case kMatchPause: { + return s_n_llhttp__internal__n_after_start_req_48; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_49: + s_n_llhttp__internal__n_after_start_req_49: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_49; + } + match_seq = llparse__match_sequence_id(state, p, endp, llparse_blob45, 3); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + match = 44; + goto s_n_llhttp__internal__n_invoke_store_method_1; + } + case kMatchPause: { + return s_n_llhttp__internal__n_after_start_req_49; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_50: + s_n_llhttp__internal__n_after_start_req_50: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_50; + } + match_seq = llparse__match_sequence_id(state, p, endp, llparse_blob46, 5); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + match = 43; + goto s_n_llhttp__internal__n_invoke_store_method_1; + } + case kMatchPause: { + return s_n_llhttp__internal__n_after_start_req_50; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_51: + s_n_llhttp__internal__n_after_start_req_51: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_51; + } + match_seq = llparse__match_sequence_id(state, p, endp, llparse_blob47, 3); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + match = 20; + goto s_n_llhttp__internal__n_invoke_store_method_1; + } + case kMatchPause: { + return s_n_llhttp__internal__n_after_start_req_51; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_47: + s_n_llhttp__internal__n_after_start_req_47: { + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_47; + } + switch (*p) { + case 'B': { + p++; + goto s_n_llhttp__internal__n_after_start_req_48; + } + case 'C': { + p++; + goto s_n_llhttp__internal__n_after_start_req_49; + } + case 'D': { + p++; + goto s_n_llhttp__internal__n_after_start_req_50; + } + case 'P': { + p++; + goto s_n_llhttp__internal__n_after_start_req_51; + } + default: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_46: + s_n_llhttp__internal__n_after_start_req_46: { + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_46; + } + switch (*p) { + case 'E': { + p++; + goto s_n_llhttp__internal__n_after_start_req_47; + } + default: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_54: + s_n_llhttp__internal__n_after_start_req_54: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_54; + } + match_seq = llparse__match_sequence_id(state, p, endp, llparse_blob48, 3); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + match = 14; + goto s_n_llhttp__internal__n_invoke_store_method_1; + } + case kMatchPause: { + return s_n_llhttp__internal__n_after_start_req_54; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_56: + s_n_llhttp__internal__n_after_start_req_56: { + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_56; + } + switch (*p) { + case 'P': { + p++; + match = 37; + goto s_n_llhttp__internal__n_invoke_store_method_1; + } + default: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_57: + s_n_llhttp__internal__n_after_start_req_57: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_57; + } + match_seq = llparse__match_sequence_id(state, p, endp, llparse_blob49, 9); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + match = 42; + goto s_n_llhttp__internal__n_invoke_store_method_1; + } + case kMatchPause: { + return s_n_llhttp__internal__n_after_start_req_57; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_55: + s_n_llhttp__internal__n_after_start_req_55: { + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_55; + } + switch (*p) { + case 'U': { + p++; + goto s_n_llhttp__internal__n_after_start_req_56; + } + case '_': { + p++; + goto s_n_llhttp__internal__n_after_start_req_57; + } + default: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_53: + s_n_llhttp__internal__n_after_start_req_53: { + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_53; + } + switch (*p) { + case 'A': { + p++; + goto s_n_llhttp__internal__n_after_start_req_54; + } + case 'T': { + p++; + goto s_n_llhttp__internal__n_after_start_req_55; + } + default: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_58: + s_n_llhttp__internal__n_after_start_req_58: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_58; + } + match_seq = llparse__match_sequence_id(state, p, endp, llparse_blob50, 4); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + match = 33; + goto s_n_llhttp__internal__n_invoke_store_method_1; + } + case kMatchPause: { + return s_n_llhttp__internal__n_after_start_req_58; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_59: + s_n_llhttp__internal__n_after_start_req_59: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_59; + } + match_seq = llparse__match_sequence_id(state, p, endp, llparse_blob51, 7); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + match = 26; + goto s_n_llhttp__internal__n_invoke_store_method_1; + } + case kMatchPause: { + return s_n_llhttp__internal__n_after_start_req_59; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_52: + s_n_llhttp__internal__n_after_start_req_52: { + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_52; + } + switch (*p) { + case 'E': { + p++; + goto s_n_llhttp__internal__n_after_start_req_53; + } + case 'O': { + p++; + goto s_n_llhttp__internal__n_after_start_req_58; + } + case 'U': { + p++; + goto s_n_llhttp__internal__n_after_start_req_59; + } + default: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_61: + s_n_llhttp__internal__n_after_start_req_61: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_61; + } + match_seq = llparse__match_sequence_id(state, p, endp, llparse_blob52, 6); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + match = 40; + goto s_n_llhttp__internal__n_invoke_store_method_1; + } + case kMatchPause: { + return s_n_llhttp__internal__n_after_start_req_61; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_62: + s_n_llhttp__internal__n_after_start_req_62: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_62; + } + match_seq = llparse__match_sequence_id(state, p, endp, llparse_blob53, 3); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + match = 7; + goto s_n_llhttp__internal__n_invoke_store_method_1; + } + case kMatchPause: { + return s_n_llhttp__internal__n_after_start_req_62; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_60: + s_n_llhttp__internal__n_after_start_req_60: { + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_60; + } + switch (*p) { + case 'E': { + p++; + goto s_n_llhttp__internal__n_after_start_req_61; + } + case 'R': { + p++; + goto s_n_llhttp__internal__n_after_start_req_62; + } + default: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_65: + s_n_llhttp__internal__n_after_start_req_65: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_65; + } + match_seq = llparse__match_sequence_id(state, p, endp, llparse_blob54, 3); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + match = 18; + goto s_n_llhttp__internal__n_invoke_store_method_1; + } + case kMatchPause: { + return s_n_llhttp__internal__n_after_start_req_65; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_67: + s_n_llhttp__internal__n_after_start_req_67: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_67; + } + match_seq = llparse__match_sequence_id(state, p, endp, llparse_blob55, 2); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + match = 32; + goto s_n_llhttp__internal__n_invoke_store_method_1; + } + case kMatchPause: { + return s_n_llhttp__internal__n_after_start_req_67; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_68: + s_n_llhttp__internal__n_after_start_req_68: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_68; + } + match_seq = llparse__match_sequence_id(state, p, endp, llparse_blob56, 2); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + match = 15; + goto s_n_llhttp__internal__n_invoke_store_method_1; + } + case kMatchPause: { + return s_n_llhttp__internal__n_after_start_req_68; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_66: + s_n_llhttp__internal__n_after_start_req_66: { + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_66; + } + switch (*p) { + case 'I': { + p++; + goto s_n_llhttp__internal__n_after_start_req_67; + } + case 'O': { + p++; + goto s_n_llhttp__internal__n_after_start_req_68; + } + default: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_69: + s_n_llhttp__internal__n_after_start_req_69: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_69; + } + match_seq = llparse__match_sequence_id(state, p, endp, llparse_blob57, 8); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + match = 27; + goto s_n_llhttp__internal__n_invoke_store_method_1; + } + case kMatchPause: { + return s_n_llhttp__internal__n_after_start_req_69; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_64: + s_n_llhttp__internal__n_after_start_req_64: { + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_64; + } + switch (*p) { + case 'B': { + p++; + goto s_n_llhttp__internal__n_after_start_req_65; + } + case 'L': { + p++; + goto s_n_llhttp__internal__n_after_start_req_66; + } + case 'S': { + p++; + goto s_n_llhttp__internal__n_after_start_req_69; + } + default: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req_63: + s_n_llhttp__internal__n_after_start_req_63: { + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req_63; + } + switch (*p) { + case 'N': { + p++; + goto s_n_llhttp__internal__n_after_start_req_64; + } + default: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_after_start_req: + s_n_llhttp__internal__n_after_start_req: { + if (p == endp) { + return s_n_llhttp__internal__n_after_start_req; + } + switch (*p) { + case 'A': { + p++; + goto s_n_llhttp__internal__n_after_start_req_1; + } + case 'B': { + p++; + goto s_n_llhttp__internal__n_after_start_req_4; + } + case 'C': { + p++; + goto s_n_llhttp__internal__n_after_start_req_5; + } + case 'D': { + p++; + goto s_n_llhttp__internal__n_after_start_req_10; + } + case 'F': { + p++; + goto s_n_llhttp__internal__n_after_start_req_14; + } + case 'G': { + p++; + goto s_n_llhttp__internal__n_after_start_req_15; + } + case 'H': { + p++; + goto s_n_llhttp__internal__n_after_start_req_18; + } + case 'L': { + p++; + goto s_n_llhttp__internal__n_after_start_req_19; + } + case 'M': { + p++; + goto s_n_llhttp__internal__n_after_start_req_22; + } + case 'N': { + p++; + goto s_n_llhttp__internal__n_after_start_req_31; + } + case 'O': { + p++; + goto s_n_llhttp__internal__n_after_start_req_32; + } + case 'P': { + p++; + goto s_n_llhttp__internal__n_after_start_req_33; + } + case 'R': { + p++; + goto s_n_llhttp__internal__n_after_start_req_46; + } + case 'S': { + p++; + goto s_n_llhttp__internal__n_after_start_req_52; + } + case 'T': { + p++; + goto s_n_llhttp__internal__n_after_start_req_60; + } + case 'U': { + p++; + goto s_n_llhttp__internal__n_after_start_req_63; + } + default: { + goto s_n_llhttp__internal__n_error_90; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_span_start_llhttp__on_method_1: + s_n_llhttp__internal__n_span_start_llhttp__on_method_1: { + if (p == endp) { + return s_n_llhttp__internal__n_span_start_llhttp__on_method_1; + } + state->_span_pos0 = (void*) p; + state->_span_cb0 = llhttp__on_method; + goto s_n_llhttp__internal__n_after_start_req; + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_res_line_almost_done: + s_n_llhttp__internal__n_res_line_almost_done: { + if (p == endp) { + return s_n_llhttp__internal__n_res_line_almost_done; + } + switch (*p) { + case 10: { + p++; + goto s_n_llhttp__internal__n_invoke_llhttp__on_status_complete; + } + case 13: { + p++; + goto s_n_llhttp__internal__n_invoke_llhttp__on_status_complete; + } + default: { + goto s_n_llhttp__internal__n_invoke_test_lenient_flags_18; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_res_status: + s_n_llhttp__internal__n_res_status: { + if (p == endp) { + return s_n_llhttp__internal__n_res_status; + } + switch (*p) { + case 10: { + goto s_n_llhttp__internal__n_span_end_llhttp__on_status; + } + case 13: { + goto s_n_llhttp__internal__n_span_end_llhttp__on_status_1; + } + default: { + p++; + goto s_n_llhttp__internal__n_res_status; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_span_start_llhttp__on_status: + s_n_llhttp__internal__n_span_start_llhttp__on_status: { + if (p == endp) { + return s_n_llhttp__internal__n_span_start_llhttp__on_status; + } + state->_span_pos0 = (void*) p; + state->_span_cb0 = llhttp__on_status; + goto s_n_llhttp__internal__n_res_status; + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_res_status_start: + s_n_llhttp__internal__n_res_status_start: { + if (p == endp) { + return s_n_llhttp__internal__n_res_status_start; + } + switch (*p) { + case 10: { + p++; + goto s_n_llhttp__internal__n_invoke_llhttp__on_status_complete; + } + case 13: { + p++; + goto s_n_llhttp__internal__n_res_line_almost_done; + } + default: { + goto s_n_llhttp__internal__n_span_start_llhttp__on_status; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_res_status_code_otherwise: + s_n_llhttp__internal__n_res_status_code_otherwise: { + if (p == endp) { + return s_n_llhttp__internal__n_res_status_code_otherwise; + } + switch (*p) { + case 10: { + goto s_n_llhttp__internal__n_res_status_start; + } + case 13: { + goto s_n_llhttp__internal__n_res_status_start; + } + case ' ': { + p++; + goto s_n_llhttp__internal__n_res_status_start; + } + default: { + goto s_n_llhttp__internal__n_error_77; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_res_status_code_digit_3: + s_n_llhttp__internal__n_res_status_code_digit_3: { + if (p == endp) { + return s_n_llhttp__internal__n_res_status_code_digit_3; + } + switch (*p) { + case '0': { + p++; + match = 0; + goto s_n_llhttp__internal__n_invoke_mul_add_status_code_2; + } + case '1': { + p++; + match = 1; + goto s_n_llhttp__internal__n_invoke_mul_add_status_code_2; + } + case '2': { + p++; + match = 2; + goto s_n_llhttp__internal__n_invoke_mul_add_status_code_2; + } + case '3': { + p++; + match = 3; + goto s_n_llhttp__internal__n_invoke_mul_add_status_code_2; + } + case '4': { + p++; + match = 4; + goto s_n_llhttp__internal__n_invoke_mul_add_status_code_2; + } + case '5': { + p++; + match = 5; + goto s_n_llhttp__internal__n_invoke_mul_add_status_code_2; + } + case '6': { + p++; + match = 6; + goto s_n_llhttp__internal__n_invoke_mul_add_status_code_2; + } + case '7': { + p++; + match = 7; + goto s_n_llhttp__internal__n_invoke_mul_add_status_code_2; + } + case '8': { + p++; + match = 8; + goto s_n_llhttp__internal__n_invoke_mul_add_status_code_2; + } + case '9': { + p++; + match = 9; + goto s_n_llhttp__internal__n_invoke_mul_add_status_code_2; + } + default: { + goto s_n_llhttp__internal__n_error_79; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_res_status_code_digit_2: + s_n_llhttp__internal__n_res_status_code_digit_2: { + if (p == endp) { + return s_n_llhttp__internal__n_res_status_code_digit_2; + } + switch (*p) { + case '0': { + p++; + match = 0; + goto s_n_llhttp__internal__n_invoke_mul_add_status_code_1; + } + case '1': { + p++; + match = 1; + goto s_n_llhttp__internal__n_invoke_mul_add_status_code_1; + } + case '2': { + p++; + match = 2; + goto s_n_llhttp__internal__n_invoke_mul_add_status_code_1; + } + case '3': { + p++; + match = 3; + goto s_n_llhttp__internal__n_invoke_mul_add_status_code_1; + } + case '4': { + p++; + match = 4; + goto s_n_llhttp__internal__n_invoke_mul_add_status_code_1; + } + case '5': { + p++; + match = 5; + goto s_n_llhttp__internal__n_invoke_mul_add_status_code_1; + } + case '6': { + p++; + match = 6; + goto s_n_llhttp__internal__n_invoke_mul_add_status_code_1; + } + case '7': { + p++; + match = 7; + goto s_n_llhttp__internal__n_invoke_mul_add_status_code_1; + } + case '8': { + p++; + match = 8; + goto s_n_llhttp__internal__n_invoke_mul_add_status_code_1; + } + case '9': { + p++; + match = 9; + goto s_n_llhttp__internal__n_invoke_mul_add_status_code_1; + } + default: { + goto s_n_llhttp__internal__n_error_81; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_res_status_code_digit_1: + s_n_llhttp__internal__n_res_status_code_digit_1: { + if (p == endp) { + return s_n_llhttp__internal__n_res_status_code_digit_1; + } + switch (*p) { + case '0': { + p++; + match = 0; + goto s_n_llhttp__internal__n_invoke_mul_add_status_code; + } + case '1': { + p++; + match = 1; + goto s_n_llhttp__internal__n_invoke_mul_add_status_code; + } + case '2': { + p++; + match = 2; + goto s_n_llhttp__internal__n_invoke_mul_add_status_code; + } + case '3': { + p++; + match = 3; + goto s_n_llhttp__internal__n_invoke_mul_add_status_code; + } + case '4': { + p++; + match = 4; + goto s_n_llhttp__internal__n_invoke_mul_add_status_code; + } + case '5': { + p++; + match = 5; + goto s_n_llhttp__internal__n_invoke_mul_add_status_code; + } + case '6': { + p++; + match = 6; + goto s_n_llhttp__internal__n_invoke_mul_add_status_code; + } + case '7': { + p++; + match = 7; + goto s_n_llhttp__internal__n_invoke_mul_add_status_code; + } + case '8': { + p++; + match = 8; + goto s_n_llhttp__internal__n_invoke_mul_add_status_code; + } + case '9': { + p++; + match = 9; + goto s_n_llhttp__internal__n_invoke_mul_add_status_code; + } + default: { + goto s_n_llhttp__internal__n_error_83; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_res_after_version: + s_n_llhttp__internal__n_res_after_version: { + if (p == endp) { + return s_n_llhttp__internal__n_res_after_version; + } + switch (*p) { + case ' ': { + p++; + goto s_n_llhttp__internal__n_invoke_update_status_code; + } + default: { + goto s_n_llhttp__internal__n_error_84; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_invoke_llhttp__on_version_complete_1: + s_n_llhttp__internal__n_invoke_llhttp__on_version_complete_1: { + switch (llhttp__on_version_complete(state, p, endp)) { + case 0: + goto s_n_llhttp__internal__n_res_after_version; + case 21: + goto s_n_llhttp__internal__n_pause_21; + default: + goto s_n_llhttp__internal__n_error_74; + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_error_73: + s_n_llhttp__internal__n_error_73: { + state->error = 0x9; + state->reason = "Invalid HTTP version"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_error_85: + s_n_llhttp__internal__n_error_85: { + state->error = 0x9; + state->reason = "Invalid minor version"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_res_http_minor: + s_n_llhttp__internal__n_res_http_minor: { + if (p == endp) { + return s_n_llhttp__internal__n_res_http_minor; + } + switch (*p) { + case '0': { + p++; + match = 0; + goto s_n_llhttp__internal__n_invoke_store_http_minor_1; + } + case '1': { + p++; + match = 1; + goto s_n_llhttp__internal__n_invoke_store_http_minor_1; + } + case '2': { + p++; + match = 2; + goto s_n_llhttp__internal__n_invoke_store_http_minor_1; + } + case '3': { + p++; + match = 3; + goto s_n_llhttp__internal__n_invoke_store_http_minor_1; + } + case '4': { + p++; + match = 4; + goto s_n_llhttp__internal__n_invoke_store_http_minor_1; + } + case '5': { + p++; + match = 5; + goto s_n_llhttp__internal__n_invoke_store_http_minor_1; + } + case '6': { + p++; + match = 6; + goto s_n_llhttp__internal__n_invoke_store_http_minor_1; + } + case '7': { + p++; + match = 7; + goto s_n_llhttp__internal__n_invoke_store_http_minor_1; + } + case '8': { + p++; + match = 8; + goto s_n_llhttp__internal__n_invoke_store_http_minor_1; + } + case '9': { + p++; + match = 9; + goto s_n_llhttp__internal__n_invoke_store_http_minor_1; + } + default: { + goto s_n_llhttp__internal__n_span_end_llhttp__on_version_7; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_error_86: + s_n_llhttp__internal__n_error_86: { + state->error = 0x9; + state->reason = "Expected dot"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_res_http_dot: + s_n_llhttp__internal__n_res_http_dot: { + if (p == endp) { + return s_n_llhttp__internal__n_res_http_dot; + } + switch (*p) { + case '.': { + p++; + goto s_n_llhttp__internal__n_res_http_minor; + } + default: { + goto s_n_llhttp__internal__n_span_end_llhttp__on_version_8; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_error_87: + s_n_llhttp__internal__n_error_87: { + state->error = 0x9; + state->reason = "Invalid major version"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_res_http_major: + s_n_llhttp__internal__n_res_http_major: { + if (p == endp) { + return s_n_llhttp__internal__n_res_http_major; + } + switch (*p) { + case '0': { + p++; + match = 0; + goto s_n_llhttp__internal__n_invoke_store_http_major_1; + } + case '1': { + p++; + match = 1; + goto s_n_llhttp__internal__n_invoke_store_http_major_1; + } + case '2': { + p++; + match = 2; + goto s_n_llhttp__internal__n_invoke_store_http_major_1; + } + case '3': { + p++; + match = 3; + goto s_n_llhttp__internal__n_invoke_store_http_major_1; + } + case '4': { + p++; + match = 4; + goto s_n_llhttp__internal__n_invoke_store_http_major_1; + } + case '5': { + p++; + match = 5; + goto s_n_llhttp__internal__n_invoke_store_http_major_1; + } + case '6': { + p++; + match = 6; + goto s_n_llhttp__internal__n_invoke_store_http_major_1; + } + case '7': { + p++; + match = 7; + goto s_n_llhttp__internal__n_invoke_store_http_major_1; + } + case '8': { + p++; + match = 8; + goto s_n_llhttp__internal__n_invoke_store_http_major_1; + } + case '9': { + p++; + match = 9; + goto s_n_llhttp__internal__n_invoke_store_http_major_1; + } + default: { + goto s_n_llhttp__internal__n_span_end_llhttp__on_version_9; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_span_start_llhttp__on_version_1: + s_n_llhttp__internal__n_span_start_llhttp__on_version_1: { + if (p == endp) { + return s_n_llhttp__internal__n_span_start_llhttp__on_version_1; + } + state->_span_pos0 = (void*) p; + state->_span_cb0 = llhttp__on_version; + goto s_n_llhttp__internal__n_res_http_major; + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_start_res: + s_n_llhttp__internal__n_start_res: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_start_res; + } + match_seq = llparse__match_sequence_id(state, p, endp, llparse_blob58, 5); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + goto s_n_llhttp__internal__n_span_start_llhttp__on_version_1; + } + case kMatchPause: { + return s_n_llhttp__internal__n_start_res; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_error_91; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_invoke_llhttp__on_method_complete: + s_n_llhttp__internal__n_invoke_llhttp__on_method_complete: { + switch (llhttp__on_method_complete(state, p, endp)) { + case 0: + goto s_n_llhttp__internal__n_req_first_space_before_url; + case 21: + goto s_n_llhttp__internal__n_pause_19; + default: + goto s_n_llhttp__internal__n_error_1; + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_req_or_res_method_2: + s_n_llhttp__internal__n_req_or_res_method_2: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_req_or_res_method_2; + } + match_seq = llparse__match_sequence_id(state, p, endp, llparse_blob59, 2); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + match = 2; + goto s_n_llhttp__internal__n_invoke_store_method; + } + case kMatchPause: { + return s_n_llhttp__internal__n_req_or_res_method_2; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_error_88; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_invoke_update_type_1: + s_n_llhttp__internal__n_invoke_update_type_1: { + switch (llhttp__internal__c_update_type_1(state, p, endp)) { + default: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version_1; + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_req_or_res_method_3: + s_n_llhttp__internal__n_req_or_res_method_3: { + llparse_match_t match_seq; + + if (p == endp) { + return s_n_llhttp__internal__n_req_or_res_method_3; + } + match_seq = llparse__match_sequence_id(state, p, endp, llparse_blob60, 3); + p = match_seq.current; + switch (match_seq.status) { + case kMatchComplete: { + p++; + goto s_n_llhttp__internal__n_span_end_llhttp__on_method_1; + } + case kMatchPause: { + return s_n_llhttp__internal__n_req_or_res_method_3; + } + case kMatchMismatch: { + goto s_n_llhttp__internal__n_error_88; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_req_or_res_method_1: + s_n_llhttp__internal__n_req_or_res_method_1: { + if (p == endp) { + return s_n_llhttp__internal__n_req_or_res_method_1; + } + switch (*p) { + case 'E': { + p++; + goto s_n_llhttp__internal__n_req_or_res_method_2; + } + case 'T': { + p++; + goto s_n_llhttp__internal__n_req_or_res_method_3; + } + default: { + goto s_n_llhttp__internal__n_error_88; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_req_or_res_method: + s_n_llhttp__internal__n_req_or_res_method: { + if (p == endp) { + return s_n_llhttp__internal__n_req_or_res_method; + } + switch (*p) { + case 'H': { + p++; + goto s_n_llhttp__internal__n_req_or_res_method_1; + } + default: { + goto s_n_llhttp__internal__n_error_88; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_span_start_llhttp__on_method: + s_n_llhttp__internal__n_span_start_llhttp__on_method: { + if (p == endp) { + return s_n_llhttp__internal__n_span_start_llhttp__on_method; + } + state->_span_pos0 = (void*) p; + state->_span_cb0 = llhttp__on_method; + goto s_n_llhttp__internal__n_req_or_res_method; + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_start_req_or_res: + s_n_llhttp__internal__n_start_req_or_res: { + if (p == endp) { + return s_n_llhttp__internal__n_start_req_or_res; + } + switch (*p) { + case 'H': { + goto s_n_llhttp__internal__n_span_start_llhttp__on_method; + } + default: { + goto s_n_llhttp__internal__n_invoke_update_type_2; + } + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_invoke_load_type: + s_n_llhttp__internal__n_invoke_load_type: { + switch (llhttp__internal__c_load_type(state, p, endp)) { + case 1: + goto s_n_llhttp__internal__n_span_start_llhttp__on_method_1; + case 2: + goto s_n_llhttp__internal__n_start_res; + default: + goto s_n_llhttp__internal__n_start_req_or_res; + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_invoke_update_finish: + s_n_llhttp__internal__n_invoke_update_finish: { + switch (llhttp__internal__c_update_finish(state, p, endp)) { + default: + goto s_n_llhttp__internal__n_invoke_llhttp__on_message_begin; + } + /* UNREACHABLE */; + abort(); + } + case s_n_llhttp__internal__n_start: + s_n_llhttp__internal__n_start: { + if (p == endp) { + return s_n_llhttp__internal__n_start; + } + switch (*p) { + case 10: { + p++; + goto s_n_llhttp__internal__n_start; + } + case 13: { + p++; + goto s_n_llhttp__internal__n_start; + } + default: { + goto s_n_llhttp__internal__n_invoke_load_initial_message_completed; + } + } + /* UNREACHABLE */; + abort(); + } + default: + /* UNREACHABLE */ + abort(); + } + s_n_llhttp__internal__n_error_2: { + state->error = 0x7; + state->reason = "Invalid characters in url"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_update_finish_2: { + switch (llhttp__internal__c_update_finish_1(state, p, endp)) { + default: + goto s_n_llhttp__internal__n_start; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_update_initial_message_completed: { + switch (llhttp__internal__c_update_initial_message_completed(state, p, endp)) { + default: + goto s_n_llhttp__internal__n_invoke_update_finish_2; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_update_content_length: { + switch (llhttp__internal__c_update_content_length(state, p, endp)) { + default: + goto s_n_llhttp__internal__n_invoke_update_initial_message_completed; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_7: { + state->error = 0x5; + state->reason = "Data after `Connection: close`"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_test_lenient_flags_3: { + switch (llhttp__internal__c_test_lenient_flags_3(state, p, endp)) { + case 1: + goto s_n_llhttp__internal__n_closed; + default: + goto s_n_llhttp__internal__n_error_7; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_test_lenient_flags_2: { + switch (llhttp__internal__c_test_lenient_flags_2(state, p, endp)) { + case 1: + goto s_n_llhttp__internal__n_invoke_update_initial_message_completed; + default: + goto s_n_llhttp__internal__n_closed; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_update_finish_1: { + switch (llhttp__internal__c_update_finish_1(state, p, endp)) { + default: + goto s_n_llhttp__internal__n_invoke_test_lenient_flags_2; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_pause_2: { + state->error = 0x15; + state->reason = "on_message_complete pause"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_pause_1; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_8: { + state->error = 0x12; + state->reason = "`on_message_complete` callback error"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_llhttp__on_message_complete_1: { + switch (llhttp__on_message_complete(state, p, endp)) { + case 0: + goto s_n_llhttp__internal__n_pause_1; + case 21: + goto s_n_llhttp__internal__n_pause_2; + default: + goto s_n_llhttp__internal__n_error_8; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_26: { + state->error = 0xc; + state->reason = "Chunk size overflow"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_pause_3: { + state->error = 0x15; + state->reason = "on_chunk_complete pause"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_invoke_update_content_length_1; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_10: { + state->error = 0x14; + state->reason = "`on_chunk_complete` callback error"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_llhttp__on_chunk_complete: { + switch (llhttp__on_chunk_complete(state, p, endp)) { + case 0: + goto s_n_llhttp__internal__n_invoke_update_content_length_1; + case 21: + goto s_n_llhttp__internal__n_pause_3; + default: + goto s_n_llhttp__internal__n_error_10; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_11: { + state->error = 0x2; + state->reason = "Expected LF after chunk data"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_test_lenient_flags_4: { + switch (llhttp__internal__c_test_lenient_flags_4(state, p, endp)) { + case 1: + goto s_n_llhttp__internal__n_invoke_llhttp__on_chunk_complete; + default: + goto s_n_llhttp__internal__n_error_11; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_body: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_body(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_chunk_data_almost_done; + return s_error; + } + goto s_n_llhttp__internal__n_chunk_data_almost_done; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_or_flags: { + switch (llhttp__internal__c_or_flags(state, p, endp)) { + default: + goto s_n_llhttp__internal__n_header_field_start; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_pause_4: { + state->error = 0x15; + state->reason = "on_chunk_header pause"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_invoke_is_equal_content_length; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_9: { + state->error = 0x13; + state->reason = "`on_chunk_header` callback error"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_llhttp__on_chunk_header: { + switch (llhttp__on_chunk_header(state, p, endp)) { + case 0: + goto s_n_llhttp__internal__n_invoke_is_equal_content_length; + case 21: + goto s_n_llhttp__internal__n_pause_4; + default: + goto s_n_llhttp__internal__n_error_9; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_12: { + state->error = 0x2; + state->reason = "Expected LF after chunk size"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_test_lenient_flags_5: { + switch (llhttp__internal__c_test_lenient_flags_5(state, p, endp)) { + case 1: + goto s_n_llhttp__internal__n_invoke_llhttp__on_chunk_header; + default: + goto s_n_llhttp__internal__n_error_12; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_13: { + state->error = 0x2; + state->reason = "Invalid character in chunk extensions"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_14: { + state->error = 0x2; + state->reason = "Invalid character in chunk extensions"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_pause_5: { + state->error = 0x15; + state->reason = "on_chunk_extension_name pause"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_chunk_size_almost_done; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_15: { + state->error = 0x22; + state->reason = "`on_chunk_extension_name` callback error"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_chunk_extension_name: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_chunk_extension_name(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) (p + 1); + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_invoke_llhttp__on_chunk_extension_name_complete; + return s_error; + } + p++; + goto s_n_llhttp__internal__n_invoke_llhttp__on_chunk_extension_name_complete; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_pause_6: { + state->error = 0x15; + state->reason = "on_chunk_extension_name pause"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_chunk_extensions; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_16: { + state->error = 0x22; + state->reason = "`on_chunk_extension_name` callback error"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_chunk_extension_name_1: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_chunk_extension_name(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) (p + 1); + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_invoke_llhttp__on_chunk_extension_name_complete_1; + return s_error; + } + p++; + goto s_n_llhttp__internal__n_invoke_llhttp__on_chunk_extension_name_complete_1; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_pause_7: { + state->error = 0x15; + state->reason = "on_chunk_extension_value pause"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_chunk_size_almost_done; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_18: { + state->error = 0x23; + state->reason = "`on_chunk_extension_value` callback error"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_chunk_extension_value: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_chunk_extension_value(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) (p + 1); + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_invoke_llhttp__on_chunk_extension_value_complete; + return s_error; + } + p++; + goto s_n_llhttp__internal__n_invoke_llhttp__on_chunk_extension_value_complete; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_20: { + state->error = 0x2; + state->reason = "Invalid character in chunk extensions quote value"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_pause_8: { + state->error = 0x15; + state->reason = "on_chunk_extension_value pause"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_chunk_extension_quoted_value_done; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_19: { + state->error = 0x23; + state->reason = "`on_chunk_extension_value` callback error"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_chunk_extension_value_1: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_chunk_extension_value(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_invoke_llhttp__on_chunk_extension_value_complete_1; + return s_error; + } + goto s_n_llhttp__internal__n_invoke_llhttp__on_chunk_extension_value_complete_1; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_chunk_extension_value_2: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_chunk_extension_value(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) (p + 1); + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_error_21; + return s_error; + } + p++; + goto s_n_llhttp__internal__n_error_21; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_pause_9: { + state->error = 0x15; + state->reason = "on_chunk_extension_value pause"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_chunk_size_otherwise; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_22: { + state->error = 0x23; + state->reason = "`on_chunk_extension_value` callback error"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_chunk_extension_value_3: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_chunk_extension_value(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) (p + 1); + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_invoke_llhttp__on_chunk_extension_value_complete_2; + return s_error; + } + p++; + goto s_n_llhttp__internal__n_invoke_llhttp__on_chunk_extension_value_complete_2; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_chunk_extension_value_4: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_chunk_extension_value(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) (p + 1); + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_error_23; + return s_error; + } + p++; + goto s_n_llhttp__internal__n_error_23; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_pause_10: { + state->error = 0x15; + state->reason = "on_chunk_extension_name pause"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_chunk_extension_value; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_17: { + state->error = 0x22; + state->reason = "`on_chunk_extension_name` callback error"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_llhttp__on_chunk_extension_name_complete_2: { + switch (llhttp__on_chunk_extension_name_complete(state, p, endp)) { + case 0: + goto s_n_llhttp__internal__n_chunk_extension_value; + case 21: + goto s_n_llhttp__internal__n_pause_10; + default: + goto s_n_llhttp__internal__n_error_17; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_chunk_extension_name_2: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_chunk_extension_name(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) (p + 1); + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_span_start_llhttp__on_chunk_extension_value; + return s_error; + } + p++; + goto s_n_llhttp__internal__n_span_start_llhttp__on_chunk_extension_value; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_chunk_extension_name_3: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_chunk_extension_name(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) (p + 1); + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_error_24; + return s_error; + } + p++; + goto s_n_llhttp__internal__n_error_24; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_25: { + state->error = 0xc; + state->reason = "Invalid character in chunk size"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_mul_add_content_length: { + switch (llhttp__internal__c_mul_add_content_length(state, p, endp, match)) { + case 1: + goto s_n_llhttp__internal__n_error_26; + default: + goto s_n_llhttp__internal__n_chunk_size; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_27: { + state->error = 0xc; + state->reason = "Invalid character in chunk size"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_pause_11: { + state->error = 0x15; + state->reason = "on_message_complete pause"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_invoke_is_equal_upgrade; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_28: { + state->error = 0x12; + state->reason = "`on_message_complete` callback error"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_body_1: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_body(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_invoke_llhttp__on_message_complete_2; + return s_error; + } + goto s_n_llhttp__internal__n_invoke_llhttp__on_message_complete_2; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_update_finish_3: { + switch (llhttp__internal__c_update_finish_3(state, p, endp)) { + default: + goto s_n_llhttp__internal__n_span_start_llhttp__on_body_2; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_29: { + state->error = 0xf; + state->reason = "Request has invalid `Transfer-Encoding`"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_pause: { + state->error = 0x15; + state->reason = "on_message_complete pause"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_invoke_llhttp__after_message_complete; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_6: { + state->error = 0x12; + state->reason = "`on_message_complete` callback error"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_llhttp__on_message_complete: { + switch (llhttp__on_message_complete(state, p, endp)) { + case 0: + goto s_n_llhttp__internal__n_invoke_llhttp__after_message_complete; + case 21: + goto s_n_llhttp__internal__n_pause; + default: + goto s_n_llhttp__internal__n_error_6; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_test_lenient_flags_1: { + switch (llhttp__internal__c_test_lenient_flags(state, p, endp)) { + case 1: + goto s_n_llhttp__internal__n_invoke_llhttp__after_headers_complete; + default: + goto s_n_llhttp__internal__n_error_5; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_pause_13: { + state->error = 0x15; + state->reason = "on_chunk_complete pause"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_invoke_llhttp__on_message_complete_2; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_32: { + state->error = 0x14; + state->reason = "`on_chunk_complete` callback error"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_llhttp__on_chunk_complete_1: { + switch (llhttp__on_chunk_complete(state, p, endp)) { + case 0: + goto s_n_llhttp__internal__n_invoke_llhttp__on_message_complete_2; + case 21: + goto s_n_llhttp__internal__n_pause_13; + default: + goto s_n_llhttp__internal__n_error_32; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_31: { + state->error = 0x4; + state->reason = "Content-Length can't be present with Transfer-Encoding"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_or_flags_1: { + switch (llhttp__internal__c_or_flags_1(state, p, endp)) { + default: + goto s_n_llhttp__internal__n_invoke_llhttp__after_headers_complete; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_or_flags_2: { + switch (llhttp__internal__c_or_flags_1(state, p, endp)) { + default: + goto s_n_llhttp__internal__n_invoke_llhttp__after_headers_complete; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_update_upgrade: { + switch (llhttp__internal__c_update_upgrade(state, p, endp)) { + default: + goto s_n_llhttp__internal__n_invoke_or_flags_2; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_pause_12: { + state->error = 0x15; + state->reason = "Paused by on_headers_complete"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_invoke_llhttp__after_headers_complete; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_30: { + state->error = 0x11; + state->reason = "User callback error"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_llhttp__on_headers_complete: { + switch (llhttp__on_headers_complete(state, p, endp)) { + case 0: + goto s_n_llhttp__internal__n_invoke_llhttp__after_headers_complete; + case 1: + goto s_n_llhttp__internal__n_invoke_or_flags_1; + case 2: + goto s_n_llhttp__internal__n_invoke_update_upgrade; + case 21: + goto s_n_llhttp__internal__n_pause_12; + default: + goto s_n_llhttp__internal__n_error_30; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_llhttp__before_headers_complete: { + switch (llhttp__before_headers_complete(state, p, endp)) { + default: + goto s_n_llhttp__internal__n_invoke_llhttp__on_headers_complete; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_test_lenient_flags_6: { + switch (llhttp__internal__c_test_lenient_flags_6(state, p, endp)) { + case 0: + goto s_n_llhttp__internal__n_error_31; + default: + goto s_n_llhttp__internal__n_invoke_llhttp__before_headers_complete; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_test_flags_1: { + switch (llhttp__internal__c_test_flags_1(state, p, endp)) { + case 1: + goto s_n_llhttp__internal__n_invoke_test_lenient_flags_6; + default: + goto s_n_llhttp__internal__n_invoke_llhttp__before_headers_complete; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_test_flags: { + switch (llhttp__internal__c_test_flags(state, p, endp)) { + case 1: + goto s_n_llhttp__internal__n_invoke_llhttp__on_chunk_complete_1; + default: + goto s_n_llhttp__internal__n_invoke_test_flags_1; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_33: { + state->error = 0x2; + state->reason = "Expected LF after headers"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_test_lenient_flags_7: { + switch (llhttp__internal__c_test_lenient_flags_5(state, p, endp)) { + case 1: + goto s_n_llhttp__internal__n_invoke_test_flags; + default: + goto s_n_llhttp__internal__n_error_33; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_header_field: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_header_field(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) (p + 1); + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_error_5; + return s_error; + } + p++; + goto s_n_llhttp__internal__n_error_5; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_test_lenient_flags_8: { + switch (llhttp__internal__c_test_lenient_flags(state, p, endp)) { + case 1: + goto s_n_llhttp__internal__n_header_field_colon_discard_ws; + default: + goto s_n_llhttp__internal__n_span_end_llhttp__on_header_field; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_36: { + state->error = 0xa; + state->reason = "Invalid header value char"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_test_lenient_flags_10: { + switch (llhttp__internal__c_test_lenient_flags(state, p, endp)) { + case 1: + goto s_n_llhttp__internal__n_header_value_discard_ws; + default: + goto s_n_llhttp__internal__n_error_36; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_38: { + state->error = 0xb; + state->reason = "Empty Content-Length"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_pause_14: { + state->error = 0x15; + state->reason = "on_header_value_complete pause"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_header_field_start; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_37: { + state->error = 0x1d; + state->reason = "`on_header_value_complete` callback error"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_header_value: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_header_value(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_invoke_llhttp__on_header_value_complete; + return s_error; + } + goto s_n_llhttp__internal__n_invoke_llhttp__on_header_value_complete; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_update_header_state: { + switch (llhttp__internal__c_update_header_state(state, p, endp)) { + default: + goto s_n_llhttp__internal__n_span_start_llhttp__on_header_value; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_or_flags_3: { + switch (llhttp__internal__c_or_flags_3(state, p, endp)) { + default: + goto s_n_llhttp__internal__n_invoke_update_header_state; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_or_flags_4: { + switch (llhttp__internal__c_or_flags_4(state, p, endp)) { + default: + goto s_n_llhttp__internal__n_invoke_update_header_state; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_or_flags_5: { + switch (llhttp__internal__c_or_flags_5(state, p, endp)) { + default: + goto s_n_llhttp__internal__n_invoke_update_header_state; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_or_flags_6: { + switch (llhttp__internal__c_or_flags_6(state, p, endp)) { + default: + goto s_n_llhttp__internal__n_span_start_llhttp__on_header_value; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_load_header_state_1: { + switch (llhttp__internal__c_load_header_state(state, p, endp)) { + case 5: + goto s_n_llhttp__internal__n_invoke_or_flags_3; + case 6: + goto s_n_llhttp__internal__n_invoke_or_flags_4; + case 7: + goto s_n_llhttp__internal__n_invoke_or_flags_5; + case 8: + goto s_n_llhttp__internal__n_invoke_or_flags_6; + default: + goto s_n_llhttp__internal__n_span_start_llhttp__on_header_value; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_load_header_state: { + switch (llhttp__internal__c_load_header_state(state, p, endp)) { + case 2: + goto s_n_llhttp__internal__n_error_38; + default: + goto s_n_llhttp__internal__n_invoke_load_header_state_1; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_35: { + state->error = 0xa; + state->reason = "Invalid header value char"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_test_lenient_flags_9: { + switch (llhttp__internal__c_test_lenient_flags(state, p, endp)) { + case 1: + goto s_n_llhttp__internal__n_header_value_discard_lws; + default: + goto s_n_llhttp__internal__n_error_35; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_39: { + state->error = 0x2; + state->reason = "Expected LF after CR"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_test_lenient_flags_11: { + switch (llhttp__internal__c_test_lenient_flags(state, p, endp)) { + case 1: + goto s_n_llhttp__internal__n_header_value_discard_lws; + default: + goto s_n_llhttp__internal__n_error_39; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_update_header_state_1: { + switch (llhttp__internal__c_update_header_state_1(state, p, endp)) { + default: + goto s_n_llhttp__internal__n_span_start_llhttp__on_header_value_1; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_load_header_state_3: { + switch (llhttp__internal__c_load_header_state(state, p, endp)) { + case 8: + goto s_n_llhttp__internal__n_invoke_update_header_state_1; + default: + goto s_n_llhttp__internal__n_span_start_llhttp__on_header_value_1; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_update_header_state_2: { + switch (llhttp__internal__c_update_header_state(state, p, endp)) { + default: + goto s_n_llhttp__internal__n_invoke_llhttp__on_header_value_complete; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_or_flags_7: { + switch (llhttp__internal__c_or_flags_3(state, p, endp)) { + default: + goto s_n_llhttp__internal__n_invoke_update_header_state_2; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_or_flags_8: { + switch (llhttp__internal__c_or_flags_4(state, p, endp)) { + default: + goto s_n_llhttp__internal__n_invoke_update_header_state_2; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_or_flags_9: { + switch (llhttp__internal__c_or_flags_5(state, p, endp)) { + default: + goto s_n_llhttp__internal__n_invoke_update_header_state_2; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_or_flags_10: { + switch (llhttp__internal__c_or_flags_6(state, p, endp)) { + default: + goto s_n_llhttp__internal__n_invoke_llhttp__on_header_value_complete; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_load_header_state_4: { + switch (llhttp__internal__c_load_header_state(state, p, endp)) { + case 5: + goto s_n_llhttp__internal__n_invoke_or_flags_7; + case 6: + goto s_n_llhttp__internal__n_invoke_or_flags_8; + case 7: + goto s_n_llhttp__internal__n_invoke_or_flags_9; + case 8: + goto s_n_llhttp__internal__n_invoke_or_flags_10; + default: + goto s_n_llhttp__internal__n_invoke_llhttp__on_header_value_complete; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_40: { + state->error = 0x3; + state->reason = "Missing expected LF after header value"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_header_value_1: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_header_value(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) (p + 1); + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_header_value_almost_done; + return s_error; + } + p++; + goto s_n_llhttp__internal__n_header_value_almost_done; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_header_value_3: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_header_value(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_header_value_almost_done; + return s_error; + } + goto s_n_llhttp__internal__n_header_value_almost_done; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_header_value_4: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_header_value(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) (p + 1); + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_header_value_almost_done; + return s_error; + } + p++; + goto s_n_llhttp__internal__n_header_value_almost_done; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_header_value_2: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_header_value(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_error_41; + return s_error; + } + goto s_n_llhttp__internal__n_error_41; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_test_lenient_flags_12: { + switch (llhttp__internal__c_test_lenient_flags(state, p, endp)) { + case 1: + goto s_n_llhttp__internal__n_header_value_lenient; + default: + goto s_n_llhttp__internal__n_span_end_llhttp__on_header_value_2; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_update_header_state_4: { + switch (llhttp__internal__c_update_header_state(state, p, endp)) { + default: + goto s_n_llhttp__internal__n_header_value_connection; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_or_flags_11: { + switch (llhttp__internal__c_or_flags_3(state, p, endp)) { + default: + goto s_n_llhttp__internal__n_invoke_update_header_state_4; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_or_flags_12: { + switch (llhttp__internal__c_or_flags_4(state, p, endp)) { + default: + goto s_n_llhttp__internal__n_invoke_update_header_state_4; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_or_flags_13: { + switch (llhttp__internal__c_or_flags_5(state, p, endp)) { + default: + goto s_n_llhttp__internal__n_invoke_update_header_state_4; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_or_flags_14: { + switch (llhttp__internal__c_or_flags_6(state, p, endp)) { + default: + goto s_n_llhttp__internal__n_header_value_connection; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_load_header_state_5: { + switch (llhttp__internal__c_load_header_state(state, p, endp)) { + case 5: + goto s_n_llhttp__internal__n_invoke_or_flags_11; + case 6: + goto s_n_llhttp__internal__n_invoke_or_flags_12; + case 7: + goto s_n_llhttp__internal__n_invoke_or_flags_13; + case 8: + goto s_n_llhttp__internal__n_invoke_or_flags_14; + default: + goto s_n_llhttp__internal__n_header_value_connection; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_update_header_state_5: { + switch (llhttp__internal__c_update_header_state_1(state, p, endp)) { + default: + goto s_n_llhttp__internal__n_header_value_connection_token; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_update_header_state_3: { + switch (llhttp__internal__c_update_header_state_3(state, p, endp)) { + default: + goto s_n_llhttp__internal__n_header_value_connection_ws; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_update_header_state_6: { + switch (llhttp__internal__c_update_header_state_6(state, p, endp)) { + default: + goto s_n_llhttp__internal__n_header_value_connection_ws; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_update_header_state_7: { + switch (llhttp__internal__c_update_header_state_7(state, p, endp)) { + default: + goto s_n_llhttp__internal__n_header_value_connection_ws; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_header_value_5: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_header_value(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_error_43; + return s_error; + } + goto s_n_llhttp__internal__n_error_43; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_mul_add_content_length_1: { + switch (llhttp__internal__c_mul_add_content_length_1(state, p, endp, match)) { + case 1: + goto s_n_llhttp__internal__n_span_end_llhttp__on_header_value_5; + default: + goto s_n_llhttp__internal__n_header_value_content_length; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_or_flags_15: { + switch (llhttp__internal__c_or_flags_15(state, p, endp)) { + default: + goto s_n_llhttp__internal__n_header_value_otherwise; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_header_value_6: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_header_value(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_error_44; + return s_error; + } + goto s_n_llhttp__internal__n_error_44; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_42: { + state->error = 0x4; + state->reason = "Duplicate Content-Length"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_test_flags_2: { + switch (llhttp__internal__c_test_flags_2(state, p, endp)) { + case 0: + goto s_n_llhttp__internal__n_header_value_content_length; + default: + goto s_n_llhttp__internal__n_error_42; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_header_value_8: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_header_value(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) (p + 1); + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_error_46; + return s_error; + } + p++; + goto s_n_llhttp__internal__n_error_46; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_update_header_state_8: { + switch (llhttp__internal__c_update_header_state_8(state, p, endp)) { + default: + goto s_n_llhttp__internal__n_header_value_otherwise; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_header_value_7: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_header_value(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) (p + 1); + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_error_45; + return s_error; + } + p++; + goto s_n_llhttp__internal__n_error_45; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_test_lenient_flags_13: { + switch (llhttp__internal__c_test_lenient_flags_13(state, p, endp)) { + case 0: + goto s_n_llhttp__internal__n_span_end_llhttp__on_header_value_7; + default: + goto s_n_llhttp__internal__n_header_value_te_chunked; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_load_type_1: { + switch (llhttp__internal__c_load_type(state, p, endp)) { + case 1: + goto s_n_llhttp__internal__n_invoke_test_lenient_flags_13; + default: + goto s_n_llhttp__internal__n_header_value_te_chunked; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_update_header_state_9: { + switch (llhttp__internal__c_update_header_state_1(state, p, endp)) { + default: + goto s_n_llhttp__internal__n_header_value; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_and_flags: { + switch (llhttp__internal__c_and_flags(state, p, endp)) { + default: + goto s_n_llhttp__internal__n_header_value_te_chunked; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_or_flags_17: { + switch (llhttp__internal__c_or_flags_16(state, p, endp)) { + default: + goto s_n_llhttp__internal__n_invoke_and_flags; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_test_lenient_flags_14: { + switch (llhttp__internal__c_test_lenient_flags_13(state, p, endp)) { + case 0: + goto s_n_llhttp__internal__n_span_end_llhttp__on_header_value_8; + default: + goto s_n_llhttp__internal__n_invoke_or_flags_17; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_load_type_2: { + switch (llhttp__internal__c_load_type(state, p, endp)) { + case 1: + goto s_n_llhttp__internal__n_invoke_test_lenient_flags_14; + default: + goto s_n_llhttp__internal__n_invoke_or_flags_17; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_or_flags_16: { + switch (llhttp__internal__c_or_flags_16(state, p, endp)) { + default: + goto s_n_llhttp__internal__n_invoke_and_flags; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_test_flags_3: { + switch (llhttp__internal__c_test_flags_3(state, p, endp)) { + case 1: + goto s_n_llhttp__internal__n_invoke_load_type_2; + default: + goto s_n_llhttp__internal__n_invoke_or_flags_16; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_or_flags_18: { + switch (llhttp__internal__c_or_flags_18(state, p, endp)) { + default: + goto s_n_llhttp__internal__n_invoke_update_header_state_9; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_load_header_state_2: { + switch (llhttp__internal__c_load_header_state(state, p, endp)) { + case 1: + goto s_n_llhttp__internal__n_header_value_connection; + case 2: + goto s_n_llhttp__internal__n_invoke_test_flags_2; + case 3: + goto s_n_llhttp__internal__n_invoke_test_flags_3; + case 4: + goto s_n_llhttp__internal__n_invoke_or_flags_18; + default: + goto s_n_llhttp__internal__n_header_value; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_pause_15: { + state->error = 0x15; + state->reason = "on_header_field_complete pause"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_header_value_discard_ws; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_34: { + state->error = 0x1c; + state->reason = "`on_header_field_complete` callback error"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_header_field_1: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_header_field(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) (p + 1); + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_invoke_llhttp__on_header_field_complete; + return s_error; + } + p++; + goto s_n_llhttp__internal__n_invoke_llhttp__on_header_field_complete; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_header_field_2: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_header_field(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) (p + 1); + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_invoke_llhttp__on_header_field_complete; + return s_error; + } + p++; + goto s_n_llhttp__internal__n_invoke_llhttp__on_header_field_complete; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_47: { + state->error = 0xa; + state->reason = "Invalid header token"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_update_header_state_10: { + switch (llhttp__internal__c_update_header_state_1(state, p, endp)) { + default: + goto s_n_llhttp__internal__n_header_field_general; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_store_header_state: { + switch (llhttp__internal__c_store_header_state(state, p, endp, match)) { + default: + goto s_n_llhttp__internal__n_header_field_colon; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_update_header_state_11: { + switch (llhttp__internal__c_update_header_state_1(state, p, endp)) { + default: + goto s_n_llhttp__internal__n_header_field_general; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_4: { + state->error = 0x1e; + state->reason = "Unexpected space after start line"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_test_lenient_flags: { + switch (llhttp__internal__c_test_lenient_flags(state, p, endp)) { + case 1: + goto s_n_llhttp__internal__n_header_field_start; + default: + goto s_n_llhttp__internal__n_error_4; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_pause_16: { + state->error = 0x15; + state->reason = "on_url_complete pause"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_headers_start; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_3: { + state->error = 0x1a; + state->reason = "`on_url_complete` callback error"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_llhttp__on_url_complete: { + switch (llhttp__on_url_complete(state, p, endp)) { + case 0: + goto s_n_llhttp__internal__n_headers_start; + case 21: + goto s_n_llhttp__internal__n_pause_16; + default: + goto s_n_llhttp__internal__n_error_3; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_update_http_minor: { + switch (llhttp__internal__c_update_http_minor(state, p, endp)) { + default: + goto s_n_llhttp__internal__n_invoke_llhttp__on_url_complete; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_update_http_major: { + switch (llhttp__internal__c_update_http_major(state, p, endp)) { + default: + goto s_n_llhttp__internal__n_invoke_update_http_minor; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_url_3: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_url(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_url_skip_to_http09; + return s_error; + } + goto s_n_llhttp__internal__n_url_skip_to_http09; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_48: { + state->error = 0x7; + state->reason = "Expected CRLF"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_url_4: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_url(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_url_skip_lf_to_http09; + return s_error; + } + goto s_n_llhttp__internal__n_url_skip_lf_to_http09; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_55: { + state->error = 0x17; + state->reason = "Pause on PRI/Upgrade"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_56: { + state->error = 0x9; + state->reason = "Expected HTTP/2 Connection Preface"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_53: { + state->error = 0x2; + state->reason = "Expected CRLF after version"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_test_lenient_flags_16: { + switch (llhttp__internal__c_test_lenient_flags_5(state, p, endp)) { + case 1: + goto s_n_llhttp__internal__n_headers_start; + default: + goto s_n_llhttp__internal__n_error_53; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_54: { + state->error = 0x9; + state->reason = "Expected CRLF after version"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_pause_17: { + state->error = 0x15; + state->reason = "on_version_complete pause"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_invoke_load_method_1; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_52: { + state->error = 0x21; + state->reason = "`on_version_complete` callback error"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_version_1: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_version(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_invoke_llhttp__on_version_complete; + return s_error; + } + goto s_n_llhttp__internal__n_invoke_llhttp__on_version_complete; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_version: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_version(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_error_51; + return s_error; + } + goto s_n_llhttp__internal__n_error_51; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_load_http_minor: { + switch (llhttp__internal__c_load_http_minor(state, p, endp)) { + case 9: + goto s_n_llhttp__internal__n_span_end_llhttp__on_version_1; + default: + goto s_n_llhttp__internal__n_span_end_llhttp__on_version; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_load_http_minor_1: { + switch (llhttp__internal__c_load_http_minor(state, p, endp)) { + case 0: + goto s_n_llhttp__internal__n_span_end_llhttp__on_version_1; + case 1: + goto s_n_llhttp__internal__n_span_end_llhttp__on_version_1; + default: + goto s_n_llhttp__internal__n_span_end_llhttp__on_version; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_load_http_minor_2: { + switch (llhttp__internal__c_load_http_minor(state, p, endp)) { + case 0: + goto s_n_llhttp__internal__n_span_end_llhttp__on_version_1; + default: + goto s_n_llhttp__internal__n_span_end_llhttp__on_version; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_load_http_major: { + switch (llhttp__internal__c_load_http_major(state, p, endp)) { + case 0: + goto s_n_llhttp__internal__n_invoke_load_http_minor; + case 1: + goto s_n_llhttp__internal__n_invoke_load_http_minor_1; + case 2: + goto s_n_llhttp__internal__n_invoke_load_http_minor_2; + default: + goto s_n_llhttp__internal__n_span_end_llhttp__on_version; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_test_lenient_flags_15: { + switch (llhttp__internal__c_test_lenient_flags_15(state, p, endp)) { + case 1: + goto s_n_llhttp__internal__n_span_end_llhttp__on_version_1; + default: + goto s_n_llhttp__internal__n_invoke_load_http_major; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_store_http_minor: { + switch (llhttp__internal__c_store_http_minor(state, p, endp, match)) { + default: + goto s_n_llhttp__internal__n_invoke_test_lenient_flags_15; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_version_2: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_version(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_error_57; + return s_error; + } + goto s_n_llhttp__internal__n_error_57; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_version_3: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_version(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_error_58; + return s_error; + } + goto s_n_llhttp__internal__n_error_58; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_store_http_major: { + switch (llhttp__internal__c_store_http_major(state, p, endp, match)) { + default: + goto s_n_llhttp__internal__n_req_http_dot; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_version_4: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_version(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_error_59; + return s_error; + } + goto s_n_llhttp__internal__n_error_59; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_50: { + state->error = 0x8; + state->reason = "Invalid method for HTTP/x.x request"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_load_method: { + switch (llhttp__internal__c_load_method(state, p, endp)) { + case 0: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + case 1: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + case 2: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + case 3: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + case 4: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + case 5: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + case 6: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + case 7: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + case 8: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + case 9: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + case 10: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + case 11: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + case 12: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + case 13: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + case 14: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + case 15: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + case 16: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + case 17: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + case 18: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + case 19: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + case 20: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + case 21: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + case 22: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + case 23: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + case 24: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + case 25: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + case 26: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + case 27: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + case 28: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + case 29: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + case 30: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + case 31: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + case 32: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + case 33: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + case 34: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + default: + goto s_n_llhttp__internal__n_error_50; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_62: { + state->error = 0x8; + state->reason = "Expected HTTP/"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_60: { + state->error = 0x8; + state->reason = "Expected SOURCE method for ICE/x.x request"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_load_method_2: { + switch (llhttp__internal__c_load_method(state, p, endp)) { + case 33: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + default: + goto s_n_llhttp__internal__n_error_60; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_61: { + state->error = 0x8; + state->reason = "Invalid method for RTSP/x.x request"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_load_method_3: { + switch (llhttp__internal__c_load_method(state, p, endp)) { + case 1: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + case 3: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + case 6: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + case 35: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + case 36: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + case 37: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + case 38: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + case 39: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + case 40: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + case 41: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + case 42: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + case 43: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + case 44: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + case 45: + goto s_n_llhttp__internal__n_span_start_llhttp__on_version; + default: + goto s_n_llhttp__internal__n_error_61; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_pause_18: { + state->error = 0x15; + state->reason = "on_url_complete pause"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_req_http_start; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_49: { + state->error = 0x1a; + state->reason = "`on_url_complete` callback error"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_llhttp__on_url_complete_1: { + switch (llhttp__on_url_complete(state, p, endp)) { + case 0: + goto s_n_llhttp__internal__n_req_http_start; + case 21: + goto s_n_llhttp__internal__n_pause_18; + default: + goto s_n_llhttp__internal__n_error_49; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_url_5: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_url(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_url_skip_to_http; + return s_error; + } + goto s_n_llhttp__internal__n_url_skip_to_http; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_url_6: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_url(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_url_skip_to_http09; + return s_error; + } + goto s_n_llhttp__internal__n_url_skip_to_http09; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_url_7: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_url(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_url_skip_lf_to_http09; + return s_error; + } + goto s_n_llhttp__internal__n_url_skip_lf_to_http09; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_url_8: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_url(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_url_skip_to_http; + return s_error; + } + goto s_n_llhttp__internal__n_url_skip_to_http; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_63: { + state->error = 0x7; + state->reason = "Invalid char in url fragment start"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_url_9: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_url(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_url_skip_to_http09; + return s_error; + } + goto s_n_llhttp__internal__n_url_skip_to_http09; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_url_10: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_url(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_url_skip_lf_to_http09; + return s_error; + } + goto s_n_llhttp__internal__n_url_skip_lf_to_http09; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_url_11: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_url(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_url_skip_to_http; + return s_error; + } + goto s_n_llhttp__internal__n_url_skip_to_http; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_64: { + state->error = 0x7; + state->reason = "Invalid char in url query"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_65: { + state->error = 0x7; + state->reason = "Invalid char in url path"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_url: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_url(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_url_skip_to_http09; + return s_error; + } + goto s_n_llhttp__internal__n_url_skip_to_http09; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_url_1: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_url(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_url_skip_lf_to_http09; + return s_error; + } + goto s_n_llhttp__internal__n_url_skip_lf_to_http09; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_url_2: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_url(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_url_skip_to_http; + return s_error; + } + goto s_n_llhttp__internal__n_url_skip_to_http; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_url_12: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_url(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_url_skip_to_http09; + return s_error; + } + goto s_n_llhttp__internal__n_url_skip_to_http09; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_url_13: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_url(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_url_skip_lf_to_http09; + return s_error; + } + goto s_n_llhttp__internal__n_url_skip_lf_to_http09; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_url_14: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_url(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_url_skip_to_http; + return s_error; + } + goto s_n_llhttp__internal__n_url_skip_to_http; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_66: { + state->error = 0x7; + state->reason = "Double @ in url"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_67: { + state->error = 0x7; + state->reason = "Unexpected char in url server"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_68: { + state->error = 0x7; + state->reason = "Unexpected char in url server"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_69: { + state->error = 0x7; + state->reason = "Unexpected char in url schema"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_70: { + state->error = 0x7; + state->reason = "Unexpected char in url schema"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_71: { + state->error = 0x7; + state->reason = "Unexpected start char in url"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_is_equal_method: { + switch (llhttp__internal__c_is_equal_method(state, p, endp)) { + case 0: + goto s_n_llhttp__internal__n_url_entry_normal; + default: + goto s_n_llhttp__internal__n_url_entry_connect; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_72: { + state->error = 0x6; + state->reason = "Expected space after method"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_pause_22: { + state->error = 0x15; + state->reason = "on_method_complete pause"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_req_first_space_before_url; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_89: { + state->error = 0x20; + state->reason = "`on_method_complete` callback error"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_method_2: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_method(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_invoke_llhttp__on_method_complete_1; + return s_error; + } + goto s_n_llhttp__internal__n_invoke_llhttp__on_method_complete_1; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_store_method_1: { + switch (llhttp__internal__c_store_method(state, p, endp, match)) { + default: + goto s_n_llhttp__internal__n_span_end_llhttp__on_method_2; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_90: { + state->error = 0x6; + state->reason = "Invalid method encountered"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_82: { + state->error = 0xd; + state->reason = "Invalid status code"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_80: { + state->error = 0xd; + state->reason = "Invalid status code"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_78: { + state->error = 0xd; + state->reason = "Invalid status code"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_pause_20: { + state->error = 0x15; + state->reason = "on_status_complete pause"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_headers_start; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_75: { + state->error = 0x1b; + state->reason = "`on_status_complete` callback error"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_llhttp__on_status_complete: { + switch (llhttp__on_status_complete(state, p, endp)) { + case 0: + goto s_n_llhttp__internal__n_headers_start; + case 21: + goto s_n_llhttp__internal__n_pause_20; + default: + goto s_n_llhttp__internal__n_error_75; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_76: { + state->error = 0x2; + state->reason = "Expected LF after CR"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_test_lenient_flags_18: { + switch (llhttp__internal__c_test_lenient_flags_5(state, p, endp)) { + case 1: + goto s_n_llhttp__internal__n_invoke_llhttp__on_status_complete; + default: + goto s_n_llhttp__internal__n_error_76; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_status: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_status(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) (p + 1); + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_res_line_almost_done; + return s_error; + } + p++; + goto s_n_llhttp__internal__n_res_line_almost_done; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_status_1: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_status(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) (p + 1); + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_res_line_almost_done; + return s_error; + } + p++; + goto s_n_llhttp__internal__n_res_line_almost_done; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_77: { + state->error = 0xd; + state->reason = "Invalid response status"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_mul_add_status_code_2: { + switch (llhttp__internal__c_mul_add_status_code(state, p, endp, match)) { + case 1: + goto s_n_llhttp__internal__n_error_78; + default: + goto s_n_llhttp__internal__n_res_status_code_otherwise; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_79: { + state->error = 0xd; + state->reason = "Invalid status code"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_mul_add_status_code_1: { + switch (llhttp__internal__c_mul_add_status_code(state, p, endp, match)) { + case 1: + goto s_n_llhttp__internal__n_error_80; + default: + goto s_n_llhttp__internal__n_res_status_code_digit_3; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_81: { + state->error = 0xd; + state->reason = "Invalid status code"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_mul_add_status_code: { + switch (llhttp__internal__c_mul_add_status_code(state, p, endp, match)) { + case 1: + goto s_n_llhttp__internal__n_error_82; + default: + goto s_n_llhttp__internal__n_res_status_code_digit_2; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_83: { + state->error = 0xd; + state->reason = "Invalid status code"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_update_status_code: { + switch (llhttp__internal__c_update_status_code(state, p, endp)) { + default: + goto s_n_llhttp__internal__n_res_status_code_digit_1; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_84: { + state->error = 0x9; + state->reason = "Expected space after version"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_pause_21: { + state->error = 0x15; + state->reason = "on_version_complete pause"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_res_after_version; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_74: { + state->error = 0x21; + state->reason = "`on_version_complete` callback error"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_version_6: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_version(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_invoke_llhttp__on_version_complete_1; + return s_error; + } + goto s_n_llhttp__internal__n_invoke_llhttp__on_version_complete_1; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_version_5: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_version(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_error_73; + return s_error; + } + goto s_n_llhttp__internal__n_error_73; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_load_http_minor_3: { + switch (llhttp__internal__c_load_http_minor(state, p, endp)) { + case 9: + goto s_n_llhttp__internal__n_span_end_llhttp__on_version_6; + default: + goto s_n_llhttp__internal__n_span_end_llhttp__on_version_5; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_load_http_minor_4: { + switch (llhttp__internal__c_load_http_minor(state, p, endp)) { + case 0: + goto s_n_llhttp__internal__n_span_end_llhttp__on_version_6; + case 1: + goto s_n_llhttp__internal__n_span_end_llhttp__on_version_6; + default: + goto s_n_llhttp__internal__n_span_end_llhttp__on_version_5; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_load_http_minor_5: { + switch (llhttp__internal__c_load_http_minor(state, p, endp)) { + case 0: + goto s_n_llhttp__internal__n_span_end_llhttp__on_version_6; + default: + goto s_n_llhttp__internal__n_span_end_llhttp__on_version_5; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_load_http_major_1: { + switch (llhttp__internal__c_load_http_major(state, p, endp)) { + case 0: + goto s_n_llhttp__internal__n_invoke_load_http_minor_3; + case 1: + goto s_n_llhttp__internal__n_invoke_load_http_minor_4; + case 2: + goto s_n_llhttp__internal__n_invoke_load_http_minor_5; + default: + goto s_n_llhttp__internal__n_span_end_llhttp__on_version_5; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_test_lenient_flags_17: { + switch (llhttp__internal__c_test_lenient_flags_15(state, p, endp)) { + case 1: + goto s_n_llhttp__internal__n_span_end_llhttp__on_version_6; + default: + goto s_n_llhttp__internal__n_invoke_load_http_major_1; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_store_http_minor_1: { + switch (llhttp__internal__c_store_http_minor(state, p, endp, match)) { + default: + goto s_n_llhttp__internal__n_invoke_test_lenient_flags_17; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_version_7: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_version(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_error_85; + return s_error; + } + goto s_n_llhttp__internal__n_error_85; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_version_8: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_version(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_error_86; + return s_error; + } + goto s_n_llhttp__internal__n_error_86; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_store_http_major_1: { + switch (llhttp__internal__c_store_http_major(state, p, endp, match)) { + default: + goto s_n_llhttp__internal__n_res_http_dot; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_version_9: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_version(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_error_87; + return s_error; + } + goto s_n_llhttp__internal__n_error_87; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_91: { + state->error = 0x8; + state->reason = "Expected HTTP/"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_pause_19: { + state->error = 0x15; + state->reason = "on_method_complete pause"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_req_first_space_before_url; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_1: { + state->error = 0x20; + state->reason = "`on_method_complete` callback error"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_method: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_method(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_invoke_llhttp__on_method_complete; + return s_error; + } + goto s_n_llhttp__internal__n_invoke_llhttp__on_method_complete; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_update_type: { + switch (llhttp__internal__c_update_type(state, p, endp)) { + default: + goto s_n_llhttp__internal__n_span_end_llhttp__on_method; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_store_method: { + switch (llhttp__internal__c_store_method(state, p, endp, match)) { + default: + goto s_n_llhttp__internal__n_invoke_update_type; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_88: { + state->error = 0x8; + state->reason = "Invalid word encountered"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_span_end_llhttp__on_method_1: { + const unsigned char* start; + int err; + + start = state->_span_pos0; + state->_span_pos0 = NULL; + err = llhttp__on_method(state, start, p); + if (err != 0) { + state->error = err; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_invoke_update_type_1; + return s_error; + } + goto s_n_llhttp__internal__n_invoke_update_type_1; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_update_type_2: { + switch (llhttp__internal__c_update_type(state, p, endp)) { + default: + goto s_n_llhttp__internal__n_span_start_llhttp__on_method_1; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_pause_23: { + state->error = 0x15; + state->reason = "on_message_begin pause"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_invoke_load_type; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error: { + state->error = 0x10; + state->reason = "`on_message_begin` callback error"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_llhttp__on_message_begin: { + switch (llhttp__on_message_begin(state, p, endp)) { + case 0: + goto s_n_llhttp__internal__n_invoke_load_type; + case 21: + goto s_n_llhttp__internal__n_pause_23; + default: + goto s_n_llhttp__internal__n_error; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_pause_24: { + state->error = 0x15; + state->reason = "on_reset pause"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_n_llhttp__internal__n_invoke_update_finish; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_error_92: { + state->error = 0x1f; + state->reason = "`on_reset` callback error"; + state->error_pos = (const char*) p; + state->_current = (void*) (intptr_t) s_error; + return s_error; + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_llhttp__on_reset: { + switch (llhttp__on_reset(state, p, endp)) { + case 0: + goto s_n_llhttp__internal__n_invoke_update_finish; + case 21: + goto s_n_llhttp__internal__n_pause_24; + default: + goto s_n_llhttp__internal__n_error_92; + } + /* UNREACHABLE */; + abort(); + } + s_n_llhttp__internal__n_invoke_load_initial_message_completed: { + switch (llhttp__internal__c_load_initial_message_completed(state, p, endp)) { + case 1: + goto s_n_llhttp__internal__n_invoke_llhttp__on_reset; + default: + goto s_n_llhttp__internal__n_invoke_update_finish; + } + /* UNREACHABLE */; + abort(); + } +} + +int llhttp__internal_execute(llhttp__internal_t* state, const char* p, const char* endp) { + llparse_state_t next; + + /* check lingering errors */ + if (state->error != 0) { + return state->error; + } + + /* restart spans */ + if (state->_span_pos0 != NULL) { + state->_span_pos0 = (void*) p; + } + + next = llhttp__internal__run(state, (const unsigned char*) p, (const unsigned char*) endp); + if (next == s_error) { + return state->error; + } + state->_current = (void*) (intptr_t) next; + + /* execute spans */ + if (state->_span_pos0 != NULL) { + int error; + + error = ((llhttp__internal__span_cb) state->_span_cb0)(state, state->_span_pos0, (const char*) endp); + if (error != 0) { + state->error = error; + state->error_pos = endp; + return error; + } + } + + return 0; +}
\ No newline at end of file diff --git a/contrib/restricted/llhttp/ya.make b/contrib/restricted/llhttp/ya.make new file mode 100644 index 0000000000..749a6b7ce2 --- /dev/null +++ b/contrib/restricted/llhttp/ya.make @@ -0,0 +1,27 @@ +# Generated by devtools/yamaker from nixpkgs 23.05. + +LIBRARY() + +LICENSE(MIT) + +LICENSE_TEXTS(.yandex_meta/licenses.list.txt) + +VERSION(9.0.1) + +ORIGINAL_SOURCE(https://github.com/nodejs/llhttp/archive/release/v9.0.1.tar.gz) + +ADDINCL( + GLOBAL contrib/restricted/llhttp/include +) + +NO_COMPILER_WARNINGS() + +NO_RUNTIME() + +SRCS( + src/api.c + src/http.c + src/llhttp.c +) + +END() diff --git a/ydb/public/sdk/python/ya.make b/ydb/public/sdk/python/ya.make index c05e9d4fb9..3fe3056ec6 100644 --- a/ydb/public/sdk/python/ya.make +++ b/ydb/public/sdk/python/ya.make @@ -1,9 +1,15 @@ PY23_LIBRARY() +# use direct include py2/py3 for skip protobuf peerdirs from contrib/python/ydb IF (PYTHON2) - PEERDIR(ydb/public/sdk/python2) + PEERDIR(contrib/python/ydb/py2) ELSE() - PEERDIR(ydb/public/sdk/python3) + PEERDIR(contrib/python/ydb/py3) ENDIF() +PEERDIR( + ydb/public/api/grpc + ydb/public/api/grpc/draft +) + END() diff --git a/ydb/public/sdk/python2/ydb/ydb_version.py b/ydb/public/sdk/python2/ydb/ydb_version.py deleted file mode 100644 index 24acd1ff32..0000000000 --- a/ydb/public/sdk/python2/ydb/ydb_version.py +++ /dev/null @@ -1 +0,0 @@ -VERSION = "2.10.0" diff --git a/ydb/public/sdk/python3/ydb/_grpc/common/__init__.py b/ydb/public/sdk/python3/ydb/_grpc/common/__init__.py deleted file mode 100644 index 689535172c..0000000000 --- a/ydb/public/sdk/python3/ydb/_grpc/common/__init__.py +++ /dev/null @@ -1,6 +0,0 @@ -import sys - -from ydb.public.api.grpc import * # noqa -sys.modules["ydb._grpc.common"] = sys.modules["ydb.public.api.grpc"] -from ydb.public.api import protos # noqa -sys.modules["ydb._grpc.common.protos"] = sys.modules["ydb.public.api.protos"] diff --git a/ydb/public/sdk/python3/ydb/_topic_common/test_helpers.py b/ydb/public/sdk/python3/ydb/_topic_common/test_helpers.py deleted file mode 100644 index 084146f8fb..0000000000 --- a/ydb/public/sdk/python3/ydb/_topic_common/test_helpers.py +++ /dev/null @@ -1,76 +0,0 @@ -import asyncio -import time -import typing - -from .._grpc.grpcwrapper.common_utils import IToProto, IGrpcWrapperAsyncIO - - -class StreamMock(IGrpcWrapperAsyncIO): - from_server: asyncio.Queue - from_client: asyncio.Queue - _closed: bool - - def __init__(self): - self.from_server = asyncio.Queue() - self.from_client = asyncio.Queue() - self._closed = False - - async def receive(self) -> typing.Any: - if self._closed: - raise Exception("read from closed StreamMock") - - item = await self.from_server.get() - if item is None: - raise StopAsyncIteration() - if isinstance(item, Exception): - raise item - return item - - def write(self, wrap_message: IToProto): - if self._closed: - raise Exception("write to closed StreamMock") - self.from_client.put_nowait(wrap_message) - - def close(self): - if self._closed: - return - - self._closed = True - self.from_server.put_nowait(None) - - -class WaitConditionError(Exception): - pass - - -async def wait_condition( - f: typing.Callable[[], bool], - timeout: typing.Optional[typing.Union[float, int]] = None, -): - """ - timeout default is 1 second - if timeout is 0 - only counter work. It userful if test need fast timeout for condition (without wait full timeout) - """ - if timeout is None: - timeout = 1 - - minimal_loop_count_for_wait = 1000 - - start = time.monotonic() - counter = 0 - while (time.monotonic() - start < timeout) or counter < minimal_loop_count_for_wait: - counter += 1 - if f(): - return - await asyncio.sleep(0) - - raise WaitConditionError("Bad condition in test") - - -async def wait_for_fast( - awaitable: typing.Union[typing.Awaitable, typing.Coroutine], - timeout: typing.Optional[typing.Union[float, int]] = None, -): - fut = asyncio.ensure_future(awaitable) - await wait_condition(lambda: fut.done(), timeout) - return fut.result() diff --git a/ydb/public/sdk/ya.make b/ydb/public/sdk/ya.make index 0cb53806e9..2507276520 100644 --- a/ydb/public/sdk/ya.make +++ b/ydb/public/sdk/ya.make @@ -1,6 +1,4 @@ RECURSE( cpp python - python2 - python3 ) diff --git a/ydb/tests/oss/ydb_sdk_import/ya.make b/ydb/tests/oss/ydb_sdk_import/ya.make index 24cc17d79e..254ed37111 100644 --- a/ydb/tests/oss/ydb_sdk_import/ya.make +++ b/ydb/tests/oss/ydb_sdk_import/ya.make @@ -6,9 +6,9 @@ PY_SRCS( IF (PYTHON2) ENV(PYTHON2_YDB_IMPORT='yes') - PEERDIR(ydb/public/sdk/python2) + PEERDIR(ydb/public/sdk/python) ELSE() - PEERDIR(ydb/public/sdk/python3) + PEERDIR(ydb/public/sdk/python) ENDIF() PEERDIR( diff --git a/ydb/tests/tools/ydb_serializable/lib/ya.make b/ydb/tests/tools/ydb_serializable/lib/ya.make index a0b5200d7b..291aa43200 100644 --- a/ydb/tests/tools/ydb_serializable/lib/ya.make +++ b/ydb/tests/tools/ydb_serializable/lib/ya.make @@ -2,7 +2,7 @@ PY3_LIBRARY() PEERDIR( ydb/tests/library - ydb/public/sdk/python3 + ydb/public/sdk/python ) PY_SRCS(__init__.py) |