aboutsummaryrefslogtreecommitdiffstats
path: root/contrib/python/fonttools/fontTools/feaLib/lexer.py
blob: 4b6499d06f591efbdb5e603cbcd256840695f431 (plain) (blame)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
from fontTools.feaLib.error import FeatureLibError, IncludedFeaNotFound
from fontTools.feaLib.location import FeatureLibLocation
import re
import os

try:
    import cython
except ImportError:
    # if cython not installed, use mock module with no-op decorators and types
    from fontTools.misc import cython


class Lexer(object):
    NUMBER = "NUMBER"
    HEXADECIMAL = "HEXADECIMAL"
    OCTAL = "OCTAL"
    NUMBERS = (NUMBER, HEXADECIMAL, OCTAL)
    FLOAT = "FLOAT"
    STRING = "STRING"
    NAME = "NAME"
    FILENAME = "FILENAME"
    GLYPHCLASS = "GLYPHCLASS"
    CID = "CID"
    SYMBOL = "SYMBOL"
    COMMENT = "COMMENT"
    NEWLINE = "NEWLINE"
    ANONYMOUS_BLOCK = "ANONYMOUS_BLOCK"

    CHAR_WHITESPACE_ = " \t"
    CHAR_NEWLINE_ = "\r\n"
    CHAR_SYMBOL_ = ",;:-+'{}[]<>()="
    CHAR_DIGIT_ = "0123456789"
    CHAR_HEXDIGIT_ = "0123456789ABCDEFabcdef"
    CHAR_LETTER_ = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz"
    CHAR_NAME_START_ = CHAR_LETTER_ + "_+*:.^~!\\"
    CHAR_NAME_CONTINUATION_ = CHAR_LETTER_ + CHAR_DIGIT_ + "_.+*:^~!/-"

    RE_GLYPHCLASS = re.compile(r"^[A-Za-z_0-9.\-]+$")

    MODE_NORMAL_ = "NORMAL"
    MODE_FILENAME_ = "FILENAME"

    def __init__(self, text, filename):
        self.filename_ = filename
        self.line_ = 1
        self.pos_ = 0
        self.line_start_ = 0
        self.text_ = text
        self.text_length_ = len(text)
        self.mode_ = Lexer.MODE_NORMAL_

    def __iter__(self):
        return self

    def next(self):  # Python 2
        return self.__next__()

    def __next__(self):  # Python 3
        while True:
            token_type, token, location = self.next_()
            if token_type != Lexer.NEWLINE:
                return (token_type, token, location)

    def location_(self):
        column = self.pos_ - self.line_start_ + 1
        return FeatureLibLocation(self.filename_ or "<features>", self.line_, column)

    def next_(self):
        self.scan_over_(Lexer.CHAR_WHITESPACE_)
        location = self.location_()
        start = self.pos_
        text = self.text_
        limit = len(text)
        if start >= limit:
            raise StopIteration()
        cur_char = text[start]
        next_char = text[start + 1] if start + 1 < limit else None

        if cur_char == "\n":
            self.pos_ += 1
            self.line_ += 1
            self.line_start_ = self.pos_
            return (Lexer.NEWLINE, None, location)
        if cur_char == "\r":
            self.pos_ += 2 if next_char == "\n" else 1
            self.line_ += 1
            self.line_start_ = self.pos_
            return (Lexer.NEWLINE, None, location)
        if cur_char == "#":
            self.scan_until_(Lexer.CHAR_NEWLINE_)
            return (Lexer.COMMENT, text[start : self.pos_], location)

        if self.mode_ is Lexer.MODE_FILENAME_:
            if cur_char != "(":
                raise FeatureLibError("Expected '(' before file name", location)
            self.scan_until_(")")
            cur_char = text[self.pos_] if self.pos_ < limit else None
            if cur_char != ")":
                raise FeatureLibError("Expected ')' after file name", location)
            self.pos_ += 1
            self.mode_ = Lexer.MODE_NORMAL_
            return (Lexer.FILENAME, text[start + 1 : self.pos_ - 1], location)

        if cur_char == "\\" and next_char in Lexer.CHAR_DIGIT_:
            self.pos_ += 1
            self.scan_over_(Lexer.CHAR_DIGIT_)
            return (Lexer.CID, int(text[start + 1 : self.pos_], 10), location)
        if cur_char == "@":
            self.pos_ += 1
            self.scan_over_(Lexer.CHAR_NAME_CONTINUATION_)
            glyphclass = text[start + 1 : self.pos_]
            if len(glyphclass) < 1:
                raise FeatureLibError("Expected glyph class name", location)
            if not Lexer.RE_GLYPHCLASS.match(glyphclass):
                raise FeatureLibError(
                    "Glyph class names must consist of letters, digits, "
                    "underscore, period or hyphen",
                    location,
                )
            return (Lexer.GLYPHCLASS, glyphclass, location)
        if cur_char in Lexer.CHAR_NAME_START_:
            self.pos_ += 1
            self.scan_over_(Lexer.CHAR_NAME_CONTINUATION_)
            token = text[start : self.pos_]
            if token == "include":
                self.mode_ = Lexer.MODE_FILENAME_
            return (Lexer.NAME, token, location)
        if cur_char == "0" and next_char in "xX":
            self.pos_ += 2
            self.scan_over_(Lexer.CHAR_HEXDIGIT_)
            return (Lexer.HEXADECIMAL, int(text[start : self.pos_], 16), location)
        if cur_char == "0" and next_char in Lexer.CHAR_DIGIT_:
            self.scan_over_(Lexer.CHAR_DIGIT_)
            return (Lexer.OCTAL, int(text[start : self.pos_], 8), location)
        if cur_char in Lexer.CHAR_DIGIT_:
            self.scan_over_(Lexer.CHAR_DIGIT_)
            if self.pos_ >= limit or text[self.pos_] != ".":
                return (Lexer.NUMBER, int(text[start : self.pos_], 10), location)
            self.scan_over_(".")
            self.scan_over_(Lexer.CHAR_DIGIT_)
            return (Lexer.FLOAT, float(text[start : self.pos_]), location)
        if cur_char == "-" and next_char in Lexer.CHAR_DIGIT_:
            self.pos_ += 1
            self.scan_over_(Lexer.CHAR_DIGIT_)
            if self.pos_ >= limit or text[self.pos_] != ".":
                return (Lexer.NUMBER, int(text[start : self.pos_], 10), location)
            self.scan_over_(".")
            self.scan_over_(Lexer.CHAR_DIGIT_)
            return (Lexer.FLOAT, float(text[start : self.pos_]), location)
        if cur_char in Lexer.CHAR_SYMBOL_:
            self.pos_ += 1
            return (Lexer.SYMBOL, cur_char, location)
        if cur_char == '"':
            self.pos_ += 1
            self.scan_until_('"')
            if self.pos_ < self.text_length_ and self.text_[self.pos_] == '"':
                self.pos_ += 1
                # strip newlines embedded within a string
                string = re.sub("[\r\n]", "", text[start + 1 : self.pos_ - 1])
                return (Lexer.STRING, string, location)
            else:
                raise FeatureLibError("Expected '\"' to terminate string", location)
        raise FeatureLibError("Unexpected character: %r" % cur_char, location)

    def scan_over_(self, valid):
        p = self.pos_
        while p < self.text_length_ and self.text_[p] in valid:
            p += 1
        self.pos_ = p

    def scan_until_(self, stop_at):
        p = self.pos_
        while p < self.text_length_ and self.text_[p] not in stop_at:
            p += 1
        self.pos_ = p

    def scan_anonymous_block(self, tag):
        location = self.location_()
        tag = tag.strip()
        self.scan_until_(Lexer.CHAR_NEWLINE_)
        self.scan_over_(Lexer.CHAR_NEWLINE_)
        regexp = r"}\s*" + tag + r"\s*;"
        split = re.split(regexp, self.text_[self.pos_ :], maxsplit=1)
        if len(split) != 2:
            raise FeatureLibError(
                "Expected '} %s;' to terminate anonymous block" % tag, location
            )
        self.pos_ += len(split[0])
        return (Lexer.ANONYMOUS_BLOCK, split[0], location)


class IncludingLexer(object):
    """A Lexer that follows include statements.

    The OpenType feature file specification states that due to
    historical reasons, relative imports should be resolved in this
    order:

    1. If the source font is UFO format, then relative to the UFO's
       font directory
    2. relative to the top-level include file
    3. relative to the parent include file

    We only support 1 (via includeDir) and 2.
    """

    def __init__(self, featurefile, *, includeDir=None):
        """Initializes an IncludingLexer.

        Behavior:
            If includeDir is passed, it will be used to determine the top-level
            include directory to use for all encountered include statements. If it is
            not passed, ``os.path.dirname(featurefile)`` will be considered the
            include directory.
        """

        self.lexers_ = [self.make_lexer_(featurefile)]
        self.featurefilepath = self.lexers_[0].filename_
        self.includeDir = includeDir

    def __iter__(self):
        return self

    def next(self):  # Python 2
        return self.__next__()

    def __next__(self):  # Python 3
        while self.lexers_:
            lexer = self.lexers_[-1]
            try:
                token_type, token, location = next(lexer)
            except StopIteration:
                self.lexers_.pop()
                continue
            if token_type is Lexer.NAME and token == "include":
                fname_type, fname_token, fname_location = lexer.next()
                if fname_type is not Lexer.FILENAME:
                    raise FeatureLibError("Expected file name", fname_location)
                # semi_type, semi_token, semi_location = lexer.next()
                # if semi_type is not Lexer.SYMBOL or semi_token != ";":
                #    raise FeatureLibError("Expected ';'", semi_location)
                if os.path.isabs(fname_token):
                    path = fname_token
                else:
                    if self.includeDir is not None:
                        curpath = self.includeDir
                    elif self.featurefilepath is not None:
                        curpath = os.path.dirname(self.featurefilepath)
                    else:
                        # if the IncludingLexer was initialized from an in-memory
                        # file-like stream, it doesn't have a 'name' pointing to
                        # its filesystem path, therefore we fall back to using the
                        # current working directory to resolve relative includes
                        curpath = os.getcwd()
                    path = os.path.join(curpath, fname_token)
                if len(self.lexers_) >= 5:
                    raise FeatureLibError("Too many recursive includes", fname_location)
                try:
                    self.lexers_.append(self.make_lexer_(path))
                except FileNotFoundError as err:
                    raise IncludedFeaNotFound(fname_token, fname_location) from err
            else:
                return (token_type, token, location)
        raise StopIteration()

    @staticmethod
    def make_lexer_(file_or_path):
        if hasattr(file_or_path, "read"):
            fileobj, closing = file_or_path, False
        else:
            filename, closing = file_or_path, True
            fileobj = open(filename, "r", encoding="utf-8-sig")
        data = fileobj.read()
        filename = getattr(fileobj, "name", None)
        if closing:
            fileobj.close()
        return Lexer(data, filename)

    def scan_anonymous_block(self, tag):
        return self.lexers_[-1].scan_anonymous_block(tag)


class NonIncludingLexer(IncludingLexer):
    """Lexer that does not follow `include` statements, emits them as-is."""

    def __next__(self):  # Python 3
        return next(self.lexers_[0])