summaryrefslogtreecommitdiff
path: root/giscanner/annotationparser.py
blob: ba70c3ea0d22048d911146d9092075c8a76ad634 (plain)
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
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
# -*- Mode: Python -*-
# GObject-Introspection - a framework for introspecting GObject libraries
# Copyright (C) 2008  Johan Dahlin
#
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License
# as published by the Free Software Foundation; either version 2
# of the License, or (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software
# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
# 02110-1301, USA.
#

# AnnotationParser - parses gtk-doc annotations

import re

from .odict import odict

# All gtk-doc comments needs to start with this:
_COMMENT_HEADER = '*\n '

# Tags - annotations applyed to comment blocks
TAG_VFUNC = 'virtual'
TAG_SINCE = 'since'
TAG_DEPRECATED = 'deprecated'
TAG_RETURNS = 'returns'
TAG_ATTRIBUTES = 'attributes'
TAG_RENAME_TO = 'rename to'
TAG_TYPE = 'type'
TAG_TRANSFER = 'transfer'
TAG_UNREF_FUNC = 'unref func'
TAG_REF_FUNC = 'ref func'
TAG_SET_VALUE_FUNC = 'set value func'
TAG_GET_VALUE_FUNC = 'get value func'

# Options - annotations for parameters and return values
OPT_ALLOW_NONE = 'allow-none'
OPT_ARRAY = 'array'
OPT_ELEMENT_TYPE = 'element-type'
OPT_IN = 'in'
OPT_INOUT = 'inout'
OPT_INOUT_ALT = 'in-out'
OPT_OUT = 'out'
OPT_SCOPE = 'scope'
OPT_TRANSFER = 'transfer'
OPT_TYPE = 'type'
OPT_CLOSURE = 'closure'
OPT_DESTROY = 'destroy'
OPT_SKIP = 'skip'
OPT_FOREIGN = 'foreign'

# Array options - array specific annotations
OPT_ARRAY_FIXED_SIZE = 'fixed-size'
OPT_ARRAY_LENGTH = 'length'
OPT_ARRAY_ZERO_TERMINATED = 'zero-terminated'

OPT_SCOPE_ASYNC = 'async'
OPT_SCOPE_CALL = 'call'
OPT_SCOPE_NOTIFIED = 'notified'

class DocBlock(object):

    def __init__(self, name):
        self.name = name
        self.options = {}
        self.value = None
        self.tags = odict()
        self.comment = None
        self.params = []
        self.filename = None
        self.lineno = -1

    def __repr__(self):
        return '<DocBlock %r %r>' % (self.name, self.options)

    def get(self, name):
        return self.tags.get(name)


class DocTag(object):

    def __init__(self, block, name):
        self.block = block
        self.name = name
        self.options = {}
        self.comment = None
        self.value = ''
        self.filename = None
        self.lineno = -1

    def __repr__(self):
        return '<DocTag %r %r>' % (self.name, self.options)

class DocOption(object):

    def __init__(self, tag, option):
        self.tag = tag
        self._array = []
        self._dict = {}
        for p in option.split(' '):
            if '=' in p:
                name, value = p.split('=', 1)
            else:
                name = p
                value = None
            self._dict[name] = value
            if value is None:
                self._array.append(name)
            else:
                self._array.append((name, value))

    def __repr__(self):
        return '<DocOption %r>' % (self._array, )

    def one(self):
        assert len(self._array) == 1
        return self._array[0]

    def flat(self):
        return self._array

    def all(self):
        return self._dict


class AnnotationParser(object):
    WHITESPACE_RE = re.compile(r'^\s*$')
    ASCII_TEXT_RE = re.compile(r'\s*[A-Za-z]+')
    OPTION_RE = re.compile(r'\([A-Za-z]+[^(]*\)')
    RETURNS_RE = re.compile(r'^return(s?)( value)?:', re.IGNORECASE)

    def __init__(self):
        self._blocks = {}

    def parse(self, comments):
        for comment in comments:
            self._parse_comment(comment)
        return self._blocks

    def _parse_comment(self, cmt):
        # We're looking for gtk-doc comments here, they look like this:
        # /**
        #   * symbol:
        #
        # Or, alternatively, with options:
        # /**
        #   * symbol: (name value) ...
        #
        # symbol is currently one of:
        #  - function: gtk_widget_show
        #  - signal:   GtkWidget::destroy
        #  - property: GtkWidget:visible
        #
        comment, filename, lineno = cmt
        comment = comment.lstrip()
        if not comment.startswith(_COMMENT_HEADER):
            return
        comment = comment[len(_COMMENT_HEADER):]
        comment = comment.strip()
        if not comment.startswith('* '):
            return
        comment = comment[2:]

        pos = comment.find('\n ')
        if pos == -1:
            return
        block_header = comment[:pos]
        block_header = block_header.strip()
        cpos = block_header.find(': ')
        block_name = block_header
        if cpos:
            block_name = block_name[:cpos]
        block = DocBlock(block_name)
        block.lineno = lineno
        block.filename = filename
        if cpos:
            block.options = self.parse_options(block, block_header[cpos+2:])
        comment_lines = []
        parsing_parameters = True
        last_param_tag = None

        # Second phase: parse parameters, return values, Tag: format
        # annotations.
        #
        # Valid lines look like:
        # * @foo: some comment here
        # * @baz: (inout): This has an annotation
        # * @bar: (out) (allow-none): this is a long parameter comment
        # *  that gets wrapped to the next line.
        # *
        # * Some documentation for the function.
        # *
        # * Returns: (transfer none): A value

        # offset of the first doctag in relation to the start of
        # the docblock, we parsed /** and the xxx: lines already
        lineno = 2
        for line in comment[pos+1:].split('\n'):
            line = line.lstrip()
            if (not line.startswith('*') or
                self.WHITESPACE_RE.match(line[1:])):
                # As soon as we find a line that's just whitespace,
                # we're done parsing the parameters.
                parsing_parameters = False
                lineno += 1
                continue

            line = line[1:].lstrip()

            # Look for a parameter or return value.  Both of these can
            # have parenthesized options.
            first_colonspace_index = line.find(': ')
            is_parameter = line.startswith('@')
            is_return_value = self.RETURNS_RE.search(line)
            if ((is_parameter or is_return_value)
                and first_colonspace_index > 0):
                if is_parameter:
                    argname = line[1:first_colonspace_index]
                else:
                    argname = TAG_RETURNS
                tag = DocTag(block, argname)
                tag.filename = block.filename
                tag.lineno = block.lineno + lineno
                second_colon_index = line.rfind(':')
                found_options = False
                if second_colon_index > first_colonspace_index:
                    value_line = \
                      line[first_colonspace_index+2:second_colon_index]
                    if self.OPTION_RE.search(value_line):
                        # The OPTION_RE is a little bit heuristic.  If
                        # we found two colons, we scan inside for something
                        # that looks like (foo).
                        # *Ideally* we'd change the gtk-doc format to
                        # require double colons, and then there'd be
                        # no ambiguity.  I.e.:
                        # @foo:: Some documentation here
                        # But that'd be a rather incompatible change.
                        found_options = True
                        tag.comment = line[second_colon_index+1:].strip()
                        tag.options = self.parse_options(tag, value_line)
                if not found_options:
                    # We didn't find any options, so just take the whole thing
                    # as documentation.
                    tag.comment = line[first_colonspace_index+2:].strip()
                block.tags[argname] = tag
                last_param_tag = tag
                if is_parameter:
                    block.params.append(argname)
            elif (not is_parameter) and parsing_parameters and last_param_tag:
                # We need to handle continuation lines on parameters.  The
                # conditional above - if a line doesn't start with '@', we're
                # not yet in the documentation block for the whole function,
                # and we've seen at least one parameter.
                last_param_tag.comment += (' ' + line.strip())
            elif first_colonspace_index > 0:
                # The line is of the form "Tag: some value here", like:
                # Since: 0.8
                tag_name = line[:first_colonspace_index]
                if self.ASCII_TEXT_RE.match(tag_name):
                    tag_name = tag_name.lower()
                    tag = DocTag(block, tag_name)
                    tag.value = line[first_colonspace_index+2:]
                    tag.filename = block.filename
                    tag.lineno = block.lineno + lineno
                    block.tags[tag_name] = tag
                else:
                    comment_lines.append(line)
            elif (not is_parameter):
                comment_lines.append(line)
            lineno += 1
        block.comment = '\n'.join(comment_lines)
        self._blocks[block.name] = block

    @classmethod
    def parse_options(cls, tag, value):
        # (foo)
        # (bar opt1 opt2...)
        opened = -1
        options = {}
        last = None
        for i, c in enumerate(value):
            if c == '(' and opened == -1:
                opened = i+1
            if c == ')' and opened != -1:
                segment = value[opened:i]
                parts = segment.split(' ', 1)
                if len(parts) == 2:
                    name, option = parts
                elif len(parts) == 1:
                    name = parts[0]
                    option = None
                else:
                    raise AssertionError
                if option is not None:
                    option = DocOption(tag, option)
                options[name] = option
                last = i + 2
                opened = -1

        return options