Source

sphinx / sphinx / builders / __init__.py

  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
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
# -*- coding: utf-8 -*-
"""
    sphinx.builders
    ~~~~~~~~~~~~~~~

    Builder superclass for all builders.

    :copyright: Copyright 2007-2013 by the Sphinx team, see AUTHORS.
    :license: BSD, see LICENSE for details.
"""

import os
from os import path

try:
    import multiprocessing
    import threading
except ImportError:
    multiprocessing = threading = None

from docutils import nodes

from sphinx.util.osutil import SEP, relative_uri
from sphinx.util.console import bold, purple, darkgreen, term_width_line

# side effect: registers roles and directives
from sphinx import roles
from sphinx import directives


class Builder(object):
    """
    Builds target formats from the reST sources.
    """

    # builder's name, for the -b command line options
    name = ''
    # builder's output format, or '' if no document output is produced
    format = ''
    # doctree versioning method
    versioning_method = 'none'
    # allow parallel write_doc() calls
    allow_parallel = False

    def __init__(self, app):
        self.env = app.env
        self.env.set_versioning_method(self.versioning_method)
        self.srcdir = app.srcdir
        self.confdir = app.confdir
        self.outdir = app.outdir
        self.doctreedir = app.doctreedir
        if not path.isdir(self.doctreedir):
            os.makedirs(self.doctreedir)

        self.app = app
        self.warn = app.warn
        self.info = app.info
        self.config = app.config
        self.tags = app.tags
        self.tags.add(self.format)
        self.tags.add(self.name)
        self.tags.add("format_%s" % self.format)
        self.tags.add("builder_%s" % self.name)

        # images that need to be copied over (source -> dest)
        self.images = {}

        self.init()

    # helper methods
    def init(self):
        """Load necessary templates and perform initialization.  The default
        implementation does nothing.
        """
        pass

    def create_template_bridge(self):
        """Return the template bridge configured."""
        if self.config.template_bridge:
            self.templates = self.app.import_object(
                self.config.template_bridge, 'template_bridge setting')()
        else:
            from sphinx.jinja2glue import BuiltinTemplateLoader
            self.templates = BuiltinTemplateLoader()

    def get_target_uri(self, docname, typ=None):
        """Return the target URI for a document name.

        *typ* can be used to qualify the link characteristic for individual
        builders.
        """
        raise NotImplementedError

    def get_relative_uri(self, from_, to, typ=None):
        """Return a relative URI between two source filenames.

        May raise environment.NoUri if there's no way to return a sensible URI.
        """
        return relative_uri(self.get_target_uri(from_),
                            self.get_target_uri(to, typ))

    def get_outdated_docs(self):
        """Return an iterable of output files that are outdated, or a string
        describing what an update build will build.

        If the builder does not output individual files corresponding to
        source files, return a string here.  If it does, return an iterable
        of those files that need to be written.
        """
        raise NotImplementedError

    def old_status_iterator(self, iterable, summary, colorfunc=darkgreen,
                            stringify_func=lambda x: x):
        l = 0
        for item in iterable:
            if l == 0:
                self.info(bold(summary), nonl=1)
                l = 1
            self.info(colorfunc(stringify_func(item)) + ' ', nonl=1)
            yield item
        if l == 1:
            self.info()

    # new version with progress info
    def status_iterator(self, iterable, summary, colorfunc=darkgreen, length=0,
                        stringify_func=lambda x: x):
        if length == 0:
            for item in self.old_status_iterator(iterable, summary, colorfunc,
                                                 stringify_func):
                yield item
            return
        l = 0
        summary = bold(summary)
        for item in iterable:
            l += 1
            s = '%s[%3d%%] %s' % (summary, 100*l/length,
                                  colorfunc(stringify_func(item)))
            if self.app.verbosity:
                s += '\n'
            else:
                s = term_width_line(s)
            self.info(s, nonl=1)
            yield item
        if l > 0:
            self.info()

    supported_image_types = []

    def post_process_images(self, doctree):
        """Pick the best candidate for all image URIs."""
        for node in doctree.traverse(nodes.image):
            if '?' in node['candidates']:
                # don't rewrite nonlocal image URIs
                continue
            if '*' not in node['candidates']:
                for imgtype in self.supported_image_types:
                    candidate = node['candidates'].get(imgtype, None)
                    if candidate:
                        break
                else:
                    self.warn(
                        'no matching candidate for image URI %r' % node['uri'],
                        '%s:%s' % (node.source, getattr(node, 'line', '')))
                    continue
                node['uri'] = candidate
            else:
                candidate = node['uri']
            if candidate not in self.env.images:
                # non-existing URI; let it alone
                continue
            self.images[candidate] = self.env.images[candidate][1]

    # build methods

    def build_all(self):
        """Build all source files."""
        self.build(None, summary='all source files', method='all')

    def build_specific(self, filenames):
        """Only rebuild as much as needed for changes in the *filenames*."""
        # bring the filenames to the canonical format, that is,
        # relative to the source directory and without source_suffix.
        dirlen = len(self.srcdir) + 1
        to_write = []
        suffix = self.config.source_suffix
        for filename in filenames:
            filename = path.normpath(path.abspath(filename))
            if not filename.startswith(self.srcdir):
                self.warn('file %r given on command line is not under the '
                          'source directory, ignoring' % filename)
                continue
            if not (path.isfile(filename) or path.isfile(filename + suffix)):
                self.warn('file %r given on command line does not exist, '
                          'ignoring' % filename)
                continue
            filename = filename[dirlen:]
            if filename.endswith(suffix):
                filename = filename[:-len(suffix)]
            filename = filename.replace(path.sep, SEP)
            to_write.append(filename)
        self.build(to_write, method='specific',
                   summary='%d source files given on command '
                   'line' % len(to_write))

    def build_update(self):
        """Only rebuild what was changed or added since last build."""
        to_build = self.get_outdated_docs()
        if isinstance(to_build, str):
            self.build(['__all__'], to_build)
        else:
            to_build = list(to_build)
            self.build(to_build,
                       summary='targets for %d source files that are '
                       'out of date' % len(to_build))

    def build(self, docnames, summary=None, method='update'):
        """Main build method.

        First updates the environment, and then calls :meth:`write`.
        """
        if summary:
            self.info(bold('building [%s]: ' % self.name), nonl=1)
            self.info(summary)

        updated_docnames = set()
        # while reading, collect all warnings from docutils
        warnings = []
        self.env.set_warnfunc(lambda *args: warnings.append(args))
        self.info(bold('updating environment: '), nonl=1)
        msg, length, iterator = self.env.update(self.config, self.srcdir,
                                                self.doctreedir, self.app)
        self.info(msg)
        for docname in self.status_iterator(iterator, 'reading sources... ',
                                            purple, length):
            updated_docnames.add(docname)
            # nothing further to do, the environment has already
            # done the reading
        for warning in warnings:
            self.warn(*warning)
        self.env.set_warnfunc(self.warn)

        doccount = len(updated_docnames)
        self.info(bold('looking for now-outdated files... '), nonl=1)
        for docname in self.env.check_dependents(updated_docnames):
            updated_docnames.add(docname)
        outdated = len(updated_docnames) - doccount
        if outdated:
            self.info('%d found' % outdated)
        else:
            self.info('none found')

        if updated_docnames:
            # save the environment
            from sphinx.application import ENV_PICKLE_FILENAME
            self.info(bold('pickling environment... '), nonl=True)
            self.env.topickle(path.join(self.doctreedir, ENV_PICKLE_FILENAME))
            self.info('done')

            # global actions
            self.info(bold('checking consistency... '), nonl=True)
            self.env.check_consistency()
            self.info('done')
        else:
            if method == 'update' and not docnames:
                self.info(bold('no targets are out of date.'))
                return

        # another indirection to support builders that don't build
        # files individually
        self.write(docnames, list(updated_docnames), method)

        # finish (write static files etc.)
        self.finish()
        status = (self.app.statuscode == 0 and 'succeeded'
                                           or 'finished with problems')
        if self.app._warncount:
            self.info(bold('build %s, %s warning%s.' %
                           (status, self.app._warncount,
                            self.app._warncount != 1 and 's' or '')))
        else:
            self.info(bold('build %s.' % status))

    def write(self, build_docnames, updated_docnames, method='update'):
        if build_docnames is None or build_docnames == ['__all__']:
            # build_all
            build_docnames = self.env.found_docs
        if method == 'update':
            # build updated ones as well
            docnames = set(build_docnames) | set(updated_docnames)
        else:
            docnames = set(build_docnames)

        # add all toctree-containing files that may have changed
        for docname in list(docnames):
            for tocdocname in self.env.files_to_rebuild.get(docname, []):
                if tocdocname in self.env.found_docs:
                    docnames.add(tocdocname)
        docnames.add(self.config.master_doc)

        self.info(bold('preparing documents... '), nonl=True)
        self.prepare_writing(docnames)
        self.info('done')

        warnings = []
        self.env.set_warnfunc(lambda *args: warnings.append(args))
        # check for prerequisites to parallel build
        # (parallel only works on POSIX, because the forking impl of
        # multiprocessing is required)
        if not (multiprocessing and
                self.app.parallel > 1 and
                self.allow_parallel and
                os.name == 'posix'):
            self._write_serial(sorted(docnames), warnings)
        else:
            # number of subprocesses is parallel-1 because the main process
            # is busy loading doctrees and doing write_doc_serialized()
            self._write_parallel(sorted(docnames), warnings,
                                 nproc=self.app.parallel - 1)
        self.env.set_warnfunc(self.warn)

    def _write_serial(self, docnames, warnings):
        for docname in self.status_iterator(
                docnames, 'writing output... ', darkgreen, len(docnames)):
            doctree = self.env.get_and_resolve_doctree(docname, self)
            self.write_doc_serialized(docname, doctree)
            self.write_doc(docname, doctree)
        for warning in warnings:
            self.warn(*warning)

    def _write_parallel(self, docnames, warnings, nproc):
        def write_process(docs):
            try:
                for docname, doctree in docs:
                    self.write_doc(docname, doctree)
            except KeyboardInterrupt:
                pass  # do not print a traceback on Ctrl-C
            finally:
                for warning in warnings:
                    self.warn(*warning)

        def process_thread(docs):
            p = multiprocessing.Process(target=write_process, args=(docs,))
            p.start()
            p.join()
            semaphore.release()

        # allow only "nproc" worker processes at once
        semaphore = threading.Semaphore(nproc)
        # list of threads to join when waiting for completion
        threads = []

        # warm up caches/compile templates using the first document
        firstname, docnames = docnames[0], docnames[1:]
        doctree = self.env.get_and_resolve_doctree(firstname, self)
        self.write_doc_serialized(firstname, doctree)
        self.write_doc(firstname, doctree)
        # for the rest, determine how many documents to write in one go
        ndocs = len(docnames)
        chunksize = min(ndocs // nproc, 10)
        nchunks, rest = divmod(ndocs, chunksize)
        if rest:
            nchunks += 1
        # partition documents in "chunks" that will be written by one Process
        chunks = [docnames[i*chunksize:(i+1)*chunksize] for i in range(nchunks)]
        for docnames in self.status_iterator(
                chunks, 'writing output... ', darkgreen, len(chunks),
                lambda chk: '%s .. %s' % (chk[0], chk[-1])):
            docs = []
            for docname in docnames:
                doctree = self.env.get_and_resolve_doctree(docname, self)
                self.write_doc_serialized(docname, doctree)
                docs.append((docname, doctree))
            # start a new thread to oversee the completion of this chunk
            semaphore.acquire()
            t = threading.Thread(target=process_thread, args=(docs,))
            t.setDaemon(True)
            t.start()
            threads.append(t)

        # make sure all threads have finished
        self.info(bold('waiting for workers... '))#, nonl=True)
        for t in threads:
            t.join()

    def prepare_writing(self, docnames):
        raise NotImplementedError

    def write_doc(self, docname, doctree):
        raise NotImplementedError

    def write_doc_serialized(self, docname, doctree):
        """Handle parts of write_doc that must be called in the main process
        if parallel build is active.
        """
        pass

    def finish(self):
        """Finish the building process.

        The default implementation does nothing.
        """
        pass

    def cleanup(self):
        """Cleanup any resources.

        The default implementation does nothing.
        """
        pass

    def get_builder_config(self, option, default):
        """Return a builder specific option.

        This method allows customization of common builder settings by
        inserting the name of the current builder in the option key.
        If the key does not exist, use default as builder name.
        """
        # At the moment, only XXX_use_index is looked up this way.
        # Every new builder variant must be registered in Config.config_values.
        try:
            optname = '%s_%s' % (self.name, option)
            return getattr(self.config, optname)
        except AttributeError:
            optname = '%s_%s' % (default, option)
            return getattr(self.config, optname)

BUILTIN_BUILDERS = {
    'html':       ('html', 'StandaloneHTMLBuilder'),
    'dirhtml':    ('html', 'DirectoryHTMLBuilder'),
    'singlehtml': ('html', 'SingleFileHTMLBuilder'),
    'pickle':     ('html', 'PickleHTMLBuilder'),
    'json':       ('html', 'JSONHTMLBuilder'),
    'web':        ('html', 'PickleHTMLBuilder'),
    'htmlhelp':   ('htmlhelp', 'HTMLHelpBuilder'),
    'devhelp':    ('devhelp', 'DevhelpBuilder'),
    'qthelp':     ('qthelp', 'QtHelpBuilder'),
    'epub':       ('epub', 'EpubBuilder'),
    'latex':      ('latex', 'LaTeXBuilder'),
    'text':       ('text', 'TextBuilder'),
    'man':        ('manpage', 'ManualPageBuilder'),
    'texinfo':    ('texinfo', 'TexinfoBuilder'),
    'changes':    ('changes', 'ChangesBuilder'),
    'linkcheck':  ('linkcheck', 'CheckExternalLinksBuilder'),
    'websupport': ('websupport', 'WebSupportBuilder'),
    'gettext':    ('gettext', 'MessageCatalogBuilder'),
    'xml':        ('xml', 'XMLBuilder'),
    'pseudoxml':  ('xml', 'PseudoXMLBuilder'),
}