aboutsummaryrefslogtreecommitdiffstats
path: root/contrib/python/Jinja2/py2/jinja2/lexer.py
diff options
context:
space:
mode:
authorfloatdrop <floatdrop@yandex-team.ru>2022-02-10 16:47:15 +0300
committerDaniil Cherednik <dcherednik@yandex-team.ru>2022-02-10 16:47:15 +0300
commite63b84f1d39557d9e46ac380b1f388271894293c (patch)
tree338cdaff3fb027e030b847db66df06019a0e3149 /contrib/python/Jinja2/py2/jinja2/lexer.py
parentf60febb7ea449535e7b073c386c7ff0539637fc0 (diff)
downloadydb-e63b84f1d39557d9e46ac380b1f388271894293c.tar.gz
Restoring authorship annotation for <floatdrop@yandex-team.ru>. Commit 1 of 2.
Diffstat (limited to 'contrib/python/Jinja2/py2/jinja2/lexer.py')
-rw-r--r--contrib/python/Jinja2/py2/jinja2/lexer.py820
1 files changed, 410 insertions, 410 deletions
diff --git a/contrib/python/Jinja2/py2/jinja2/lexer.py b/contrib/python/Jinja2/py2/jinja2/lexer.py
index 552356a12d..075484b88b 100644
--- a/contrib/python/Jinja2/py2/jinja2/lexer.py
+++ b/contrib/python/Jinja2/py2/jinja2/lexer.py
@@ -1,26 +1,26 @@
-# -*- coding: utf-8 -*-
+# -*- coding: utf-8 -*-
"""Implements a Jinja / Python combination lexer. The ``Lexer`` class
is used to do some preprocessing. It filters out invalid operators like
the bitshift operators we don't allow in templates. It separates
template code and python code in expressions.
-"""
-import re
+"""
+import re
from ast import literal_eval
-from collections import deque
-from operator import itemgetter
-
+from collections import deque
+from operator import itemgetter
+
from ._compat import implements_iterator
from ._compat import intern
from ._compat import iteritems
from ._compat import text_type
from .exceptions import TemplateSyntaxError
from .utils import LRUCache
-
-# cache for the lexers. Exists in order to be able to have multiple
-# environments with the same lexer
-_lexer_cache = LRUCache(50)
-
-# static regular expressions
+
+# cache for the lexers. Exists in order to be able to have multiple
+# environments with the same lexer
+_lexer_cache = LRUCache(50)
+
+# static regular expressions
whitespace_re = re.compile(r"\s+", re.U)
newline_re = re.compile(r"(\r\n|\r|\n)")
string_re = re.compile(
@@ -40,22 +40,22 @@ float_re = re.compile(
""",
re.IGNORECASE | re.VERBOSE,
)
-
-try:
- # check if this Python supports Unicode identifiers
+
+try:
+ # check if this Python supports Unicode identifiers
compile("föö", "<unknown>", "eval")
-except SyntaxError:
+except SyntaxError:
# Python 2, no Unicode support, use ASCII identifiers
name_re = re.compile(r"[a-zA-Z_][a-zA-Z0-9_]*")
- check_ident = False
-else:
+ check_ident = False
+else:
# Unicode support, import generated re pattern and set flag to use
# str.isidentifier to validate during lexing.
from ._identifier import pattern as name_re
- check_ident = True
-
-# internal the tokens and keep references to them
+ check_ident = True
+
+# internal the tokens and keep references to them
TOKEN_ADD = intern("add")
TOKEN_ASSIGN = intern("assign")
TOKEN_COLON = intern("colon")
@@ -105,9 +105,9 @@ TOKEN_LINECOMMENT = intern("linecomment")
TOKEN_DATA = intern("data")
TOKEN_INITIAL = intern("initial")
TOKEN_EOF = intern("eof")
-
-# bind operators to token types
-operators = {
+
+# bind operators to token types
+operators = {
"+": TOKEN_ADD,
"-": TOKEN_SUB,
"/": TOKEN_DIV,
@@ -134,14 +134,14 @@ operators = {
"|": TOKEN_PIPE,
",": TOKEN_COMMA,
";": TOKEN_SEMICOLON,
-}
-
-reverse_operators = dict([(v, k) for k, v in iteritems(operators)])
+}
+
+reverse_operators = dict([(v, k) for k, v in iteritems(operators)])
assert len(operators) == len(reverse_operators), "operators dropped"
operator_re = re.compile(
"(%s)" % "|".join(re.escape(x) for x in sorted(operators, key=lambda x: -len(x)))
)
-
+
ignored_tokens = frozenset(
[
TOKEN_COMMENT_BEGIN,
@@ -156,12 +156,12 @@ ignored_tokens = frozenset(
ignore_if_empty = frozenset(
[TOKEN_WHITESPACE, TOKEN_DATA, TOKEN_COMMENT, TOKEN_LINECOMMENT]
)
-
-
-def _describe_token_type(token_type):
- if token_type in reverse_operators:
- return reverse_operators[token_type]
- return {
+
+
+def _describe_token_type(token_type):
+ if token_type in reverse_operators:
+ return reverse_operators[token_type]
+ return {
TOKEN_COMMENT_BEGIN: "begin of comment",
TOKEN_COMMENT_END: "end of comment",
TOKEN_COMMENT: "comment",
@@ -174,38 +174,38 @@ def _describe_token_type(token_type):
TOKEN_LINESTATEMENT_END: "end of line statement",
TOKEN_DATA: "template data / text",
TOKEN_EOF: "end of template",
- }.get(token_type, token_type)
-
-
-def describe_token(token):
- """Returns a description of the token."""
+ }.get(token_type, token_type)
+
+
+def describe_token(token):
+ """Returns a description of the token."""
if token.type == TOKEN_NAME:
- return token.value
- return _describe_token_type(token.type)
-
-
-def describe_token_expr(expr):
- """Like `describe_token` but for token expressions."""
+ return token.value
+ return _describe_token_type(token.type)
+
+
+def describe_token_expr(expr):
+ """Like `describe_token` but for token expressions."""
if ":" in expr:
type, value = expr.split(":", 1)
if type == TOKEN_NAME:
- return value
- else:
- type = expr
- return _describe_token_type(type)
-
-
-def count_newlines(value):
- """Count the number of newline characters in the string. This is
- useful for extensions that filter a stream.
- """
- return len(newline_re.findall(value))
-
-
-def compile_rules(environment):
- """Compiles all the rules from the environment into a list of rules."""
- e = re.escape
- rules = [
+ return value
+ else:
+ type = expr
+ return _describe_token_type(type)
+
+
+def count_newlines(value):
+ """Count the number of newline characters in the string. This is
+ useful for extensions that filter a stream.
+ """
+ return len(newline_re.findall(value))
+
+
+def compile_rules(environment):
+ """Compiles all the rules from the environment into a list of rules."""
+ e = re.escape
+ rules = [
(
len(environment.comment_start_string),
TOKEN_COMMENT_BEGIN,
@@ -221,9 +221,9 @@ def compile_rules(environment):
TOKEN_VARIABLE_BEGIN,
e(environment.variable_start_string),
),
- ]
-
- if environment.line_statement_prefix is not None:
+ ]
+
+ if environment.line_statement_prefix is not None:
rules.append(
(
len(environment.line_statement_prefix),
@@ -231,7 +231,7 @@ def compile_rules(environment):
r"^[ \t\v]*" + e(environment.line_statement_prefix),
)
)
- if environment.line_comment_prefix is not None:
+ if environment.line_comment_prefix is not None:
rules.append(
(
len(environment.line_comment_prefix),
@@ -239,169 +239,169 @@ def compile_rules(environment):
r"(?:^|(?<=\S))[^\S\r\n]*" + e(environment.line_comment_prefix),
)
)
-
- return [x[1:] for x in sorted(rules, reverse=True)]
-
-
-class Failure(object):
- """Class that raises a `TemplateSyntaxError` if called.
- Used by the `Lexer` to specify known errors.
- """
-
- def __init__(self, message, cls=TemplateSyntaxError):
- self.message = message
- self.error_class = cls
-
- def __call__(self, lineno, filename):
- raise self.error_class(self.message, lineno, filename)
-
-
-class Token(tuple):
- """Token class."""
-
- __slots__ = ()
- lineno, type, value = (property(itemgetter(x)) for x in range(3))
-
- def __new__(cls, lineno, type, value):
- return tuple.__new__(cls, (lineno, intern(str(type)), value))
-
- def __str__(self):
- if self.type in reverse_operators:
- return reverse_operators[self.type]
+
+ return [x[1:] for x in sorted(rules, reverse=True)]
+
+
+class Failure(object):
+ """Class that raises a `TemplateSyntaxError` if called.
+ Used by the `Lexer` to specify known errors.
+ """
+
+ def __init__(self, message, cls=TemplateSyntaxError):
+ self.message = message
+ self.error_class = cls
+
+ def __call__(self, lineno, filename):
+ raise self.error_class(self.message, lineno, filename)
+
+
+class Token(tuple):
+ """Token class."""
+
+ __slots__ = ()
+ lineno, type, value = (property(itemgetter(x)) for x in range(3))
+
+ def __new__(cls, lineno, type, value):
+ return tuple.__new__(cls, (lineno, intern(str(type)), value))
+
+ def __str__(self):
+ if self.type in reverse_operators:
+ return reverse_operators[self.type]
elif self.type == "name":
- return self.value
- return self.type
-
- def test(self, expr):
- """Test a token against a token expression. This can either be a
- token type or ``'token_type:token_value'``. This can only test
- against string values and types.
- """
- # here we do a regular string equality check as test_any is usually
- # passed an iterable of not interned strings.
- if self.type == expr:
- return True
+ return self.value
+ return self.type
+
+ def test(self, expr):
+ """Test a token against a token expression. This can either be a
+ token type or ``'token_type:token_value'``. This can only test
+ against string values and types.
+ """
+ # here we do a regular string equality check as test_any is usually
+ # passed an iterable of not interned strings.
+ if self.type == expr:
+ return True
elif ":" in expr:
return expr.split(":", 1) == [self.type, self.value]
- return False
-
- def test_any(self, *iterable):
- """Test against multiple token expressions."""
- for expr in iterable:
- if self.test(expr):
- return True
- return False
-
- def __repr__(self):
+ return False
+
+ def test_any(self, *iterable):
+ """Test against multiple token expressions."""
+ for expr in iterable:
+ if self.test(expr):
+ return True
+ return False
+
+ def __repr__(self):
return "Token(%r, %r, %r)" % (self.lineno, self.type, self.value)
-
-
-@implements_iterator
-class TokenStreamIterator(object):
- """The iterator for tokenstreams. Iterate over the stream
- until the eof token is reached.
- """
-
- def __init__(self, stream):
- self.stream = stream
-
- def __iter__(self):
- return self
-
- def __next__(self):
- token = self.stream.current
- if token.type is TOKEN_EOF:
- self.stream.close()
- raise StopIteration()
- next(self.stream)
- return token
-
-
-@implements_iterator
-class TokenStream(object):
- """A token stream is an iterable that yields :class:`Token`\\s. The
- parser however does not iterate over it but calls :meth:`next` to go
- one token ahead. The current active token is stored as :attr:`current`.
- """
-
- def __init__(self, generator, name, filename):
- self._iter = iter(generator)
- self._pushed = deque()
- self.name = name
- self.filename = filename
- self.closed = False
+
+
+@implements_iterator
+class TokenStreamIterator(object):
+ """The iterator for tokenstreams. Iterate over the stream
+ until the eof token is reached.
+ """
+
+ def __init__(self, stream):
+ self.stream = stream
+
+ def __iter__(self):
+ return self
+
+ def __next__(self):
+ token = self.stream.current
+ if token.type is TOKEN_EOF:
+ self.stream.close()
+ raise StopIteration()
+ next(self.stream)
+ return token
+
+
+@implements_iterator
+class TokenStream(object):
+ """A token stream is an iterable that yields :class:`Token`\\s. The
+ parser however does not iterate over it but calls :meth:`next` to go
+ one token ahead. The current active token is stored as :attr:`current`.
+ """
+
+ def __init__(self, generator, name, filename):
+ self._iter = iter(generator)
+ self._pushed = deque()
+ self.name = name
+ self.filename = filename
+ self.closed = False
self.current = Token(1, TOKEN_INITIAL, "")
- next(self)
-
- def __iter__(self):
- return TokenStreamIterator(self)
-
- def __bool__(self):
- return bool(self._pushed) or self.current.type is not TOKEN_EOF
-
- __nonzero__ = __bool__ # py2
-
+ next(self)
+
+ def __iter__(self):
+ return TokenStreamIterator(self)
+
+ def __bool__(self):
+ return bool(self._pushed) or self.current.type is not TOKEN_EOF
+
+ __nonzero__ = __bool__ # py2
+
@property
def eos(self):
"""Are we at the end of the stream?"""
return not self
-
- def push(self, token):
- """Push a token back to the stream."""
- self._pushed.append(token)
-
- def look(self):
- """Look at the next token."""
- old_token = next(self)
- result = self.current
- self.push(result)
- self.current = old_token
- return result
-
- def skip(self, n=1):
- """Got n tokens ahead."""
+
+ def push(self, token):
+ """Push a token back to the stream."""
+ self._pushed.append(token)
+
+ def look(self):
+ """Look at the next token."""
+ old_token = next(self)
+ result = self.current
+ self.push(result)
+ self.current = old_token
+ return result
+
+ def skip(self, n=1):
+ """Got n tokens ahead."""
for _ in range(n):
- next(self)
-
- def next_if(self, expr):
- """Perform the token test and return the token if it matched.
- Otherwise the return value is `None`.
- """
- if self.current.test(expr):
- return next(self)
-
- def skip_if(self, expr):
- """Like :meth:`next_if` but only returns `True` or `False`."""
- return self.next_if(expr) is not None
-
- def __next__(self):
- """Go one token ahead and return the old one.
-
- Use the built-in :func:`next` instead of calling this directly.
- """
- rv = self.current
- if self._pushed:
- self.current = self._pushed.popleft()
- elif self.current.type is not TOKEN_EOF:
- try:
- self.current = next(self._iter)
- except StopIteration:
- self.close()
- return rv
-
- def close(self):
- """Close the stream."""
+ next(self)
+
+ def next_if(self, expr):
+ """Perform the token test and return the token if it matched.
+ Otherwise the return value is `None`.
+ """
+ if self.current.test(expr):
+ return next(self)
+
+ def skip_if(self, expr):
+ """Like :meth:`next_if` but only returns `True` or `False`."""
+ return self.next_if(expr) is not None
+
+ def __next__(self):
+ """Go one token ahead and return the old one.
+
+ Use the built-in :func:`next` instead of calling this directly.
+ """
+ rv = self.current
+ if self._pushed:
+ self.current = self._pushed.popleft()
+ elif self.current.type is not TOKEN_EOF:
+ try:
+ self.current = next(self._iter)
+ except StopIteration:
+ self.close()
+ return rv
+
+ def close(self):
+ """Close the stream."""
self.current = Token(self.current.lineno, TOKEN_EOF, "")
- self._iter = None
- self.closed = True
-
- def expect(self, expr):
- """Expect a given token type and return it. This accepts the same
- argument as :meth:`jinja2.lexer.Token.test`.
- """
- if not self.current.test(expr):
- expr = describe_token_expr(expr)
- if self.current.type is TOKEN_EOF:
+ self._iter = None
+ self.closed = True
+
+ def expect(self, expr):
+ """Expect a given token type and return it. This accepts the same
+ argument as :meth:`jinja2.lexer.Token.test`.
+ """
+ if not self.current.test(expr):
+ expr = describe_token_expr(expr)
+ if self.current.type is TOKEN_EOF:
raise TemplateSyntaxError(
"unexpected end of template, expected %r." % expr,
self.current.lineno,
@@ -414,14 +414,14 @@ class TokenStream(object):
self.name,
self.filename,
)
- try:
- return self.current
- finally:
- next(self)
-
-
-def get_lexer(environment):
- """Return a lexer which is probably cached."""
+ try:
+ return self.current
+ finally:
+ next(self)
+
+
+def get_lexer(environment):
+ """Return a lexer which is probably cached."""
key = (
environment.block_start_string,
environment.block_end_string,
@@ -436,13 +436,13 @@ def get_lexer(environment):
environment.newline_sequence,
environment.keep_trailing_newline,
)
- lexer = _lexer_cache.get(key)
- if lexer is None:
- lexer = Lexer(environment)
- _lexer_cache[key] = lexer
- return lexer
-
-
+ lexer = _lexer_cache.get(key)
+ if lexer is None:
+ lexer = Lexer(environment)
+ _lexer_cache[key] = lexer
+ return lexer
+
+
class OptionalLStrip(tuple):
"""A special tuple for marking a point in the state that can have
lstrip applied.
@@ -456,53 +456,53 @@ class OptionalLStrip(tuple):
return super(OptionalLStrip, cls).__new__(cls, members)
-class Lexer(object):
- """Class that implements a lexer for a given environment. Automatically
- created by the environment class, usually you don't have to do that.
-
- Note that the lexer is not automatically bound to an environment.
- Multiple environments can share the same lexer.
- """
-
- def __init__(self, environment):
- # shortcuts
- e = re.escape
-
+class Lexer(object):
+ """Class that implements a lexer for a given environment. Automatically
+ created by the environment class, usually you don't have to do that.
+
+ Note that the lexer is not automatically bound to an environment.
+ Multiple environments can share the same lexer.
+ """
+
+ def __init__(self, environment):
+ # shortcuts
+ e = re.escape
+
def c(x):
return re.compile(x, re.M | re.S)
- # lexing rules for tags
- tag_rules = [
- (whitespace_re, TOKEN_WHITESPACE, None),
- (float_re, TOKEN_FLOAT, None),
- (integer_re, TOKEN_INTEGER, None),
- (name_re, TOKEN_NAME, None),
- (string_re, TOKEN_STRING, None),
+ # lexing rules for tags
+ tag_rules = [
+ (whitespace_re, TOKEN_WHITESPACE, None),
+ (float_re, TOKEN_FLOAT, None),
+ (integer_re, TOKEN_INTEGER, None),
+ (name_re, TOKEN_NAME, None),
+ (string_re, TOKEN_STRING, None),
(operator_re, TOKEN_OPERATOR, None),
- ]
-
- # assemble the root lexing rule. because "|" is ungreedy
- # we have to sort by length so that the lexer continues working
- # as expected when we have parsing rules like <% for block and
- # <%= for variables. (if someone wants asp like syntax)
- # variables are just part of the rules if variable processing
- # is required.
- root_tag_rules = compile_rules(environment)
-
- # block suffix if trimming is enabled
+ ]
+
+ # assemble the root lexing rule. because "|" is ungreedy
+ # we have to sort by length so that the lexer continues working
+ # as expected when we have parsing rules like <% for block and
+ # <%= for variables. (if someone wants asp like syntax)
+ # variables are just part of the rules if variable processing
+ # is required.
+ root_tag_rules = compile_rules(environment)
+
+ # block suffix if trimming is enabled
block_suffix_re = environment.trim_blocks and "\\n?" or ""
-
+
# If lstrip is enabled, it should not be applied if there is any
# non-whitespace between the newline and block.
self.lstrip_unless_re = c(r"[^ \t]") if environment.lstrip_blocks else None
-
- self.newline_sequence = environment.newline_sequence
- self.keep_trailing_newline = environment.keep_trailing_newline
-
- # global lexing rules
- self.rules = {
+
+ self.newline_sequence = environment.newline_sequence
+ self.keep_trailing_newline = environment.keep_trailing_newline
+
+ # global lexing rules
+ self.rules = {
"root": [
- # directives
+ # directives
(
c(
"(.*?)(?:%s)"
@@ -524,11 +524,11 @@ class Lexer(object):
OptionalLStrip(TOKEN_DATA, "#bygroup"),
"#bygroup",
),
- # data
+ # data
(c(".+"), TOKEN_DATA, None),
- ],
- # comments
- TOKEN_COMMENT_BEGIN: [
+ ],
+ # comments
+ TOKEN_COMMENT_BEGIN: [
(
c(
r"(.*?)((?:\-%s\s*|%s)%s)"
@@ -542,9 +542,9 @@ class Lexer(object):
"#pop",
),
(c("(.)"), (Failure("Missing end of comment tag"),), None),
- ],
- # blocks
- TOKEN_BLOCK_BEGIN: [
+ ],
+ # blocks
+ TOKEN_BLOCK_BEGIN: [
(
c(
r"(?:\-%s\s*|%s)%s"
@@ -559,8 +559,8 @@ class Lexer(object):
),
]
+ tag_rules,
- # variables
- TOKEN_VARIABLE_BEGIN: [
+ # variables
+ TOKEN_VARIABLE_BEGIN: [
(
c(
r"\-%s\s*|%s"
@@ -574,8 +574,8 @@ class Lexer(object):
)
]
+ tag_rules,
- # raw block
- TOKEN_RAW_BEGIN: [
+ # raw block
+ TOKEN_RAW_BEGIN: [
(
c(
r"(.*?)((?:%s(\-|\+|))\s*endraw\s*(?:\-%s\s*|%s%s))"
@@ -590,121 +590,121 @@ class Lexer(object):
"#pop",
),
(c("(.)"), (Failure("Missing end of raw directive"),), None),
- ],
- # line statements
- TOKEN_LINESTATEMENT_BEGIN: [
+ ],
+ # line statements
+ TOKEN_LINESTATEMENT_BEGIN: [
(c(r"\s*(\n|$)"), TOKEN_LINESTATEMENT_END, "#pop")
]
+ tag_rules,
- # line comments
- TOKEN_LINECOMMENT_BEGIN: [
+ # line comments
+ TOKEN_LINECOMMENT_BEGIN: [
(
c(r"(.*?)()(?=\n|$)"),
(TOKEN_LINECOMMENT, TOKEN_LINECOMMENT_END),
"#pop",
)
],
- }
-
- def _normalize_newlines(self, value):
- """Called for strings and template data to normalize it to unicode."""
- return newline_re.sub(self.newline_sequence, value)
-
- def tokenize(self, source, name=None, filename=None, state=None):
+ }
+
+ def _normalize_newlines(self, value):
+ """Called for strings and template data to normalize it to unicode."""
+ return newline_re.sub(self.newline_sequence, value)
+
+ def tokenize(self, source, name=None, filename=None, state=None):
"""Calls tokeniter + tokenize and wraps it in a token stream."""
- stream = self.tokeniter(source, name, filename, state)
- return TokenStream(self.wrap(stream, name, filename), name, filename)
-
- def wrap(self, stream, name=None, filename=None):
- """This is called with the stream as returned by `tokenize` and wraps
- every token in a :class:`Token` and converts the value.
- """
- for lineno, token, value in stream:
- if token in ignored_tokens:
- continue
+ stream = self.tokeniter(source, name, filename, state)
+ return TokenStream(self.wrap(stream, name, filename), name, filename)
+
+ def wrap(self, stream, name=None, filename=None):
+ """This is called with the stream as returned by `tokenize` and wraps
+ every token in a :class:`Token` and converts the value.
+ """
+ for lineno, token, value in stream:
+ if token in ignored_tokens:
+ continue
elif token == TOKEN_LINESTATEMENT_BEGIN:
token = TOKEN_BLOCK_BEGIN
elif token == TOKEN_LINESTATEMENT_END:
token = TOKEN_BLOCK_END
- # we are not interested in those tokens in the parser
+ # we are not interested in those tokens in the parser
elif token in (TOKEN_RAW_BEGIN, TOKEN_RAW_END):
- continue
+ continue
elif token == TOKEN_DATA:
- value = self._normalize_newlines(value)
+ value = self._normalize_newlines(value)
elif token == "keyword":
- token = value
+ token = value
elif token == TOKEN_NAME:
- value = str(value)
- if check_ident and not value.isidentifier():
- raise TemplateSyntaxError(
+ value = str(value)
+ if check_ident and not value.isidentifier():
+ raise TemplateSyntaxError(
"Invalid character in identifier", lineno, name, filename
)
elif token == TOKEN_STRING:
- # try to unescape string
- try:
+ # try to unescape string
+ try:
value = (
self._normalize_newlines(value[1:-1])
.encode("ascii", "backslashreplace")
.decode("unicode-escape")
)
- except Exception as e:
+ except Exception as e:
msg = str(e).split(":")[-1].strip()
- raise TemplateSyntaxError(msg, lineno, name, filename)
+ raise TemplateSyntaxError(msg, lineno, name, filename)
elif token == TOKEN_INTEGER:
value = int(value.replace("_", ""))
elif token == TOKEN_FLOAT:
# remove all "_" first to support more Python versions
value = literal_eval(value.replace("_", ""))
elif token == TOKEN_OPERATOR:
- token = operators[value]
- yield Token(lineno, token, value)
-
- def tokeniter(self, source, name, filename=None, state=None):
- """This method tokenizes the text and returns the tokens in a
- generator. Use this method if you just want to tokenize a template.
- """
- source = text_type(source)
- lines = source.splitlines()
- if self.keep_trailing_newline and source:
+ token = operators[value]
+ yield Token(lineno, token, value)
+
+ def tokeniter(self, source, name, filename=None, state=None):
+ """This method tokenizes the text and returns the tokens in a
+ generator. Use this method if you just want to tokenize a template.
+ """
+ source = text_type(source)
+ lines = source.splitlines()
+ if self.keep_trailing_newline and source:
for newline in ("\r\n", "\r", "\n"):
- if source.endswith(newline):
+ if source.endswith(newline):
lines.append("")
- break
+ break
source = "\n".join(lines)
- pos = 0
- lineno = 1
+ pos = 0
+ lineno = 1
stack = ["root"]
if state is not None and state != "root":
assert state in ("variable", "block"), "invalid state"
stack.append(state + "_begin")
- statetokens = self.rules[stack[-1]]
- source_length = len(source)
- balancing_stack = []
+ statetokens = self.rules[stack[-1]]
+ source_length = len(source)
+ balancing_stack = []
lstrip_unless_re = self.lstrip_unless_re
newlines_stripped = 0
line_starting = True
-
- while 1:
- # tokenizer loop
- for regex, tokens, new_state in statetokens:
- m = regex.match(source, pos)
- # if no match we try again with the next rule
- if m is None:
- continue
-
- # we only match blocks and variables if braces / parentheses
- # are balanced. continue parsing with the lower rule which
- # is the operator rule. do this only if the end tags look
- # like operators
+
+ while 1:
+ # tokenizer loop
+ for regex, tokens, new_state in statetokens:
+ m = regex.match(source, pos)
+ # if no match we try again with the next rule
+ if m is None:
+ continue
+
+ # we only match blocks and variables if braces / parentheses
+ # are balanced. continue parsing with the lower rule which
+ # is the operator rule. do this only if the end tags look
+ # like operators
if balancing_stack and tokens in (
TOKEN_VARIABLE_END,
TOKEN_BLOCK_END,
TOKEN_LINESTATEMENT_END,
):
- continue
-
- # tuples support more options
- if isinstance(tokens, tuple):
+ continue
+
+ # tuples support more options
+ if isinstance(tokens, tuple):
groups = m.groups()
if isinstance(tokens, OptionalLStrip):
@@ -738,37 +738,37 @@ class Lexer(object):
if not lstrip_unless_re.search(text, l_pos):
groups = (text[:l_pos],) + groups[1:]
- for idx, token in enumerate(tokens):
- # failure group
- if token.__class__ is Failure:
- raise token(lineno, filename)
- # bygroup is a bit more complex, in that case we
- # yield for the current token the first named
- # group that matched
+ for idx, token in enumerate(tokens):
+ # failure group
+ if token.__class__ is Failure:
+ raise token(lineno, filename)
+ # bygroup is a bit more complex, in that case we
+ # yield for the current token the first named
+ # group that matched
elif token == "#bygroup":
- for key, value in iteritems(m.groupdict()):
- if value is not None:
- yield lineno, key, value
+ for key, value in iteritems(m.groupdict()):
+ if value is not None:
+ yield lineno, key, value
lineno += value.count("\n")
- break
- else:
+ break
+ else:
raise RuntimeError(
"%r wanted to resolve "
"the token dynamically"
" but no group matched" % regex
)
- # normal group
- else:
+ # normal group
+ else:
data = groups[idx]
- if data or token not in ignore_if_empty:
- yield lineno, token, data
+ if data or token not in ignore_if_empty:
+ yield lineno, token, data
lineno += data.count("\n") + newlines_stripped
newlines_stripped = 0
-
- # strings as token just are yielded as it.
- else:
- data = m.group()
- # update brace/parentheses balance
+
+ # strings as token just are yielded as it.
+ else:
+ data = m.group()
+ # update brace/parentheses balance
if tokens == TOKEN_OPERATOR:
if data == "{":
balancing_stack.append("}")
@@ -777,12 +777,12 @@ class Lexer(object):
elif data == "[":
balancing_stack.append("]")
elif data in ("}", ")", "]"):
- if not balancing_stack:
+ if not balancing_stack:
raise TemplateSyntaxError(
"unexpected '%s'" % data, lineno, name, filename
)
- expected_op = balancing_stack.pop()
- if expected_op != data:
+ expected_op = balancing_stack.pop()
+ if expected_op != data:
raise TemplateSyntaxError(
"unexpected '%s', "
"expected '%s'" % (data, expected_op),
@@ -790,56 +790,56 @@ class Lexer(object):
name,
filename,
)
- # yield items
- if data or tokens not in ignore_if_empty:
- yield lineno, tokens, data
+ # yield items
+ if data or tokens not in ignore_if_empty:
+ yield lineno, tokens, data
lineno += data.count("\n")
-
+
line_starting = m.group()[-1:] == "\n"
- # fetch new position into new variable so that we can check
- # if there is a internal parsing error which would result
- # in an infinite loop
- pos2 = m.end()
-
- # handle state changes
- if new_state is not None:
- # remove the uppermost state
+ # fetch new position into new variable so that we can check
+ # if there is a internal parsing error which would result
+ # in an infinite loop
+ pos2 = m.end()
+
+ # handle state changes
+ if new_state is not None:
+ # remove the uppermost state
if new_state == "#pop":
- stack.pop()
- # resolve the new state by group checking
+ stack.pop()
+ # resolve the new state by group checking
elif new_state == "#bygroup":
- for key, value in iteritems(m.groupdict()):
- if value is not None:
- stack.append(key)
- break
- else:
+ for key, value in iteritems(m.groupdict()):
+ if value is not None:
+ stack.append(key)
+ break
+ else:
raise RuntimeError(
"%r wanted to resolve the "
"new state dynamically but"
" no group matched" % regex
)
- # direct state name given
- else:
- stack.append(new_state)
- statetokens = self.rules[stack[-1]]
- # we are still at the same position and no stack change.
- # this means a loop without break condition, avoid that and
- # raise error
- elif pos2 == pos:
+ # direct state name given
+ else:
+ stack.append(new_state)
+ statetokens = self.rules[stack[-1]]
+ # we are still at the same position and no stack change.
+ # this means a loop without break condition, avoid that and
+ # raise error
+ elif pos2 == pos:
raise RuntimeError(
"%r yielded empty string without stack change" % regex
)
- # publish new function and start again
- pos = pos2
- break
- # if loop terminated without break we haven't found a single match
- # either we are at the end of the file or we have a problem
- else:
- # end of text
- if pos >= source_length:
- return
- # something went wrong
+ # publish new function and start again
+ pos = pos2
+ break
+ # if loop terminated without break we haven't found a single match
+ # either we are at the end of the file or we have a problem
+ else:
+ # end of text
+ if pos >= source_length:
+ return
+ # something went wrong
raise TemplateSyntaxError(
"unexpected char %r at %d" % (source[pos], pos),
lineno,