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
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
# coding: utf-8
# $Id: __init__.py 8672 2021-04-07 12:10:06Z milde $
# Author: David Goodger <goodger@python.org>
# Copyright: This module has been placed in the public domain.

"""
Miscellaneous utilities for the documentation utilities.
"""

__docformat__ = 'reStructuredText'

import sys
import os
import os.path
import re
import itertools
import warnings
import unicodedata
from docutils import ApplicationError, DataError, __version_info__
from docutils import nodes
from docutils.nodes import unescape
import docutils.io
from docutils.utils.error_reporting import ErrorOutput, SafeString

if sys.version_info >= (3, 0):
    unicode = str


class SystemMessage(ApplicationError):

    def __init__(self, system_message, level):
        Exception.__init__(self, system_message.astext())
        self.level = level


class SystemMessagePropagation(ApplicationError): pass


class Reporter(object):

    """
    Info/warning/error reporter and ``system_message`` element generator.

    Five levels of system messages are defined, along with corresponding
    methods: `debug()`, `info()`, `warning()`, `error()`, and `severe()`.

    There is typically one Reporter object per process.  A Reporter object is
    instantiated with thresholds for reporting (generating warnings) and
    halting processing (raising exceptions), a switch to turn debug output on
    or off, and an I/O stream for warnings.  These are stored as instance
    attributes.

    When a system message is generated, its level is compared to the stored
    thresholds, and a warning or error is generated as appropriate.  Debug
    messages are produced if the stored debug switch is on, independently of
    other thresholds.  Message output is sent to the stored warning stream if
    not set to ''.

    The Reporter class also employs a modified form of the "Observer" pattern
    [GoF95]_ to track system messages generated.  The `attach_observer` method
    should be called before parsing, with a bound method or function which
    accepts system messages.  The observer can be removed with
    `detach_observer`, and another added in its place.

    .. [GoF95] Gamma, Helm, Johnson, Vlissides. *Design Patterns: Elements of
       Reusable Object-Oriented Software*. Addison-Wesley, Reading, MA, USA,
       1995.
    """

    levels = 'DEBUG INFO WARNING ERROR SEVERE'.split()
    """List of names for system message levels, indexed by level."""

    # system message level constants:
    (DEBUG_LEVEL,
     INFO_LEVEL,
     WARNING_LEVEL,
     ERROR_LEVEL,
     SEVERE_LEVEL) = range(5)

    def __init__(self, source, report_level, halt_level, stream=None,
                 debug=False, encoding=None, error_handler='backslashreplace'):
        """
        :Parameters:
            - `source`: The path to or description of the source data.
            - `report_level`: The level at or above which warning output will
              be sent to `stream`.
            - `halt_level`: The level at or above which `SystemMessage`
              exceptions will be raised, halting execution.
            - `debug`: Show debug (level=0) system messages?
            - `stream`: Where warning output is sent.  Can be file-like (has a
              ``.write`` method), a string (file name, opened for writing),
              '' (empty string) or `False` (for discarding all stream messages)
              or `None` (implies `sys.stderr`; default).
            - `encoding`: The output encoding.
            - `error_handler`: The error handler for stderr output encoding.
        """

        self.source = source
        """The path to or description of the source data."""

        self.error_handler = error_handler
        """The character encoding error handler."""

        self.debug_flag = debug
        """Show debug (level=0) system messages?"""

        self.report_level = report_level
        """The level at or above which warning output will be sent
        to `self.stream`."""

        self.halt_level = halt_level
        """The level at or above which `SystemMessage` exceptions
        will be raised, halting execution."""

        if not isinstance(stream, ErrorOutput):
            stream = ErrorOutput(stream, encoding, error_handler)

        self.stream = stream
        """Where warning output is sent."""

        self.encoding = encoding or getattr(stream, 'encoding', 'ascii')
        """The output character encoding."""

        self.observers = []
        """List of bound methods or functions to call with each system_message
        created."""

        self.max_level = -1
        """The highest level system message generated so far."""

    def set_conditions(self, category, report_level, halt_level,
                       stream=None, debug=False):
        warnings.warn('docutils.utils.Reporter.set_conditions deprecated; '
                      'set attributes via configuration settings or directly',
                      DeprecationWarning, stacklevel=2)
        self.report_level = report_level
        self.halt_level = halt_level
        if not isinstance(stream, ErrorOutput):
            stream = ErrorOutput(stream, self.encoding, self.error_handler)
        self.stream = stream
        self.debug_flag = debug

    def attach_observer(self, observer):
        """
        The `observer` parameter is a function or bound method which takes one
        argument, a `nodes.system_message` instance.
        """
        self.observers.append(observer)

    def detach_observer(self, observer):
        self.observers.remove(observer)

    def notify_observers(self, message):
        for observer in self.observers:
            observer(message)

    def system_message(self, level, message, *children, **kwargs):
        """
        Return a system_message object.

        Raise an exception or generate a warning if appropriate.
        """
        # `message` can be a `string`, `unicode`, or `Exception` instance.
        if isinstance(message, Exception):
            message = SafeString(message)

        attributes = kwargs.copy()
        if 'base_node' in kwargs:
            source, line = get_source_line(kwargs['base_node'])
            del attributes['base_node']
            if source is not None:
                attributes.setdefault('source', source)
            if line is not None:
                attributes.setdefault('line', line)
                # assert source is not None, "node has line- but no source-argument"
        if not 'source' in attributes: # 'line' is absolute line number
            try: # look up (source, line-in-source)
                source, line = self.get_source_and_line(attributes.get('line'))
            except AttributeError:
                source, line = None, None
            if source is not None:
                attributes['source'] = source
            if line is not None:
                attributes['line'] = line
        # assert attributes['line'] is not None, (message, kwargs)
        # assert attributes['source'] is not None, (message, kwargs)
        attributes.setdefault('source', self.source)

        msg = nodes.system_message(message, level=level,
                                   type=self.levels[level],
                                   *children, **attributes)
        if self.stream and (level >= self.report_level
                            or self.debug_flag and level == self.DEBUG_LEVEL
                            or level >= self.halt_level):
            self.stream.write(msg.astext() + '\n')
        if level >= self.halt_level:
            raise SystemMessage(msg, level)
        if level > self.DEBUG_LEVEL or self.debug_flag:
            self.notify_observers(msg)
        self.max_level = max(level, self.max_level)
        return msg

    def debug(self, *args, **kwargs):
        """
        Level-0, "DEBUG": an internal reporting issue. Typically, there is no
        effect on the processing. Level-0 system messages are handled
        separately from the others.
        """
        if self.debug_flag:
            return self.system_message(self.DEBUG_LEVEL, *args, **kwargs)

    def info(self, *args, **kwargs):
        """
        Level-1, "INFO": a minor issue that can be ignored. Typically there is
        no effect on processing, and level-1 system messages are not reported.
        """
        return self.system_message(self.INFO_LEVEL, *args, **kwargs)

    def warning(self, *args, **kwargs):
        """
        Level-2, "WARNING": an issue that should be addressed. If ignored,
        there may be unpredictable problems with the output.
        """
        return self.system_message(self.WARNING_LEVEL, *args, **kwargs)

    def error(self, *args, **kwargs):
        """
        Level-3, "ERROR": an error that should be addressed. If ignored, the
        output will contain errors.
        """
        return self.system_message(self.ERROR_LEVEL, *args, **kwargs)

    def severe(self, *args, **kwargs):
        """
        Level-4, "SEVERE": a severe error that must be addressed. If ignored,
        the output will contain severe errors. Typically level-4 system
        messages are turned into exceptions which halt processing.
        """
        return self.system_message(self.SEVERE_LEVEL, *args, **kwargs)


class ExtensionOptionError(DataError): pass
class BadOptionError(ExtensionOptionError): pass
class BadOptionDataError(ExtensionOptionError): pass
class DuplicateOptionError(ExtensionOptionError): pass


def extract_extension_options(field_list, options_spec):
    """
    Return a dictionary mapping extension option names to converted values.

    :Parameters:
        - `field_list`: A flat field list without field arguments, where each
          field body consists of a single paragraph only.
        - `options_spec`: Dictionary mapping known option names to a
          conversion function such as `int` or `float`.

    :Exceptions:
        - `KeyError` for unknown option names.
        - `ValueError` for invalid option values (raised by the conversion
           function).
        - `TypeError` for invalid option value types (raised by conversion
           function).
        - `DuplicateOptionError` for duplicate options.
        - `BadOptionError` for invalid fields.
        - `BadOptionDataError` for invalid option data (missing name,
          missing data, bad quotes, etc.).
    """
    option_list = extract_options(field_list)
    option_dict = assemble_option_dict(option_list, options_spec)
    return option_dict

def extract_options(field_list):
    """
    Return a list of option (name, value) pairs from field names & bodies.

    :Parameter:
        `field_list`: A flat field list, where each field name is a single
        word and each field body consists of a single paragraph only.

    :Exceptions:
        - `BadOptionError` for invalid fields.
        - `BadOptionDataError` for invalid option data (missing name,
          missing data, bad quotes, etc.).
    """
    option_list = []
    for field in field_list:
        if len(field[0].astext().split()) != 1:
            raise BadOptionError(
                'extension option field name may not contain multiple words')
        name = str(field[0].astext().lower())
        body = field[1]
        if len(body) == 0:
            data = None
        elif len(body) > 1 or not isinstance(body[0], nodes.paragraph) \
              or len(body[0]) != 1 or not isinstance(body[0][0], nodes.Text):
            raise BadOptionDataError(
                  'extension option field body may contain\n'
                  'a single paragraph only (option "%s")' % name)
        else:
            data = body[0][0].astext()
        option_list.append((name, data))
    return option_list

def assemble_option_dict(option_list, options_spec):
    """
    Return a mapping of option names to values.

    :Parameters:
        - `option_list`: A list of (name, value) pairs (the output of
          `extract_options()`).
        - `options_spec`: Dictionary mapping known option names to a
          conversion function such as `int` or `float`.

    :Exceptions:
        - `KeyError` for unknown option names.
        - `DuplicateOptionError` for duplicate options.
        - `ValueError` for invalid option values (raised by conversion
           function).
        - `TypeError` for invalid option value types (raised by conversion
           function).
    """
    options = {}
    for name, value in option_list:
        convertor = options_spec[name]  # raises KeyError if unknown
        if convertor is None:
            raise KeyError(name)        # or if explicitly disabled
        if name in options:
            raise DuplicateOptionError('duplicate option "%s"' % name)
        try:
            options[name] = convertor(value)
        except (ValueError, TypeError) as detail:
            raise detail.__class__('(option: "%s"; value: %r)\n%s'
                                   % (name, value, ' '.join(detail.args)))
    return options


class NameValueError(DataError): pass


def decode_path(path):
    """
    Ensure `path` is Unicode. Return `nodes.reprunicode` object.

    Decode file/path string in a failsave manner if not already done.
    """
    # see also http://article.gmane.org/gmane.text.docutils.user/2905
    if isinstance(path, unicode):
        return path
    try:
        path = path.decode(sys.getfilesystemencoding(), 'strict')
    except AttributeError: # default value None has no decode method
        if not path:
            return nodes.reprunicode('')
        raise ValueError('`path` value must be a String or ``None``, not %r'
                         %path)
    except UnicodeDecodeError:
        try:
            path = path.decode('utf-8', 'strict')
        except UnicodeDecodeError:
            path = path.decode('ascii', 'replace')
    return nodes.reprunicode(path)


def extract_name_value(line):
    """
    Return a list of (name, value) from a line of the form "name=value ...".

    :Exception:
        `NameValueError` for invalid input (missing name, missing data, bad
        quotes, etc.).
    """
    attlist = []
    while line:
        equals = line.find('=')
        if equals == -1:
            raise NameValueError('missing "="')
        attname = line[:equals].strip()
        if equals == 0 or not attname:
            raise NameValueError(
                  'missing attribute name before "="')
        line = line[equals+1:].lstrip()
        if not line:
            raise NameValueError(
                  'missing value after "%s="' % attname)
        if line[0] in '\'"':
            endquote = line.find(line[0], 1)
            if endquote == -1:
                raise NameValueError(
                      'attribute "%s" missing end quote (%s)'
                      % (attname, line[0]))
            if len(line) > endquote + 1 and line[endquote + 1].strip():
                raise NameValueError(
                      'attribute "%s" end quote (%s) not followed by '
                      'whitespace' % (attname, line[0]))
            data = line[1:endquote]
            line = line[endquote+1:].lstrip()
        else:
            space = line.find(' ')
            if space == -1:
                data = line
                line = ''
            else:
                data = line[:space]
                line = line[space+1:].lstrip()
        attlist.append((attname.lower(), data))
    return attlist

def new_reporter(source_path, settings):
    """
    Return a new Reporter object.

    :Parameters:
        `source` : string
            The path to or description of the source text of the document.
        `settings` : optparse.Values object
            Runtime settings.
    """
    reporter = Reporter(
        source_path, settings.report_level, settings.halt_level,
        stream=settings.warning_stream, debug=settings.debug,
        encoding=settings.error_encoding,
        error_handler=settings.error_encoding_error_handler)
    return reporter

def new_document(source_path, settings=None):
    """
    Return a new empty document object.

    :Parameters:
        `source_path` : string
            The path to or description of the source text of the document.
        `settings` : optparse.Values object
            Runtime settings.  If none are provided, a default core set will
            be used.  If you will use the document object with any Docutils
            components, you must provide their default settings as well.  For
            example, if parsing rST, at least provide the rst-parser settings,
            obtainable as follows::

                settings = docutils.frontend.OptionParser(
                    components=(docutils.parsers.rst.Parser,)
                    ).get_default_values()
    """
    from docutils import frontend
    if settings is None:
        settings = frontend.OptionParser().get_default_values()
    source_path = decode_path(source_path)
    reporter = new_reporter(source_path, settings)
    document = nodes.document(settings, reporter, source=source_path)
    document.note_source(source_path, -1)
    return document

def clean_rcs_keywords(paragraph, keyword_substitutions):
    if len(paragraph) == 1 and isinstance(paragraph[0], nodes.Text):
        textnode = paragraph[0]
        for pattern, substitution in keyword_substitutions:
            match = pattern.search(textnode)
            if match:
                paragraph[0] = nodes.Text(pattern.sub(substitution, textnode))
                return

def relative_path(source, target):
    """
    Build and return a path to `target`, relative to `source` (both files).

    If there is no common prefix, return the absolute path to `target`.
    """
    source_parts = os.path.abspath(source or type(target)('dummy_file')
                                  ).split(os.sep)
    target_parts = os.path.abspath(target).split(os.sep)
    # Check first 2 parts because '/dir'.split('/') == ['', 'dir']:
    if source_parts[:2] != target_parts[:2]:
        # Nothing in common between paths.
        # Return absolute path, using '/' for URLs:
        return '/'.join(target_parts)
    source_parts.reverse()
    target_parts.reverse()
    while (source_parts and target_parts
           and source_parts[-1] == target_parts[-1]):
        # Remove path components in common:
        source_parts.pop()
        target_parts.pop()
    target_parts.reverse()
    parts = ['..'] * (len(source_parts) - 1) + target_parts
    return '/'.join(parts)

def get_stylesheet_reference(settings, relative_to=None):
    """
    Retrieve a stylesheet reference from the settings object.

    Deprecated. Use get_stylesheet_list() instead to
    enable specification of multiple stylesheets as a comma-separated
    list.
    """
    if settings.stylesheet_path:
        assert not settings.stylesheet, (
            'stylesheet and stylesheet_path are mutually exclusive.')
        if relative_to == None:
            relative_to = settings._destination
        return relative_path(relative_to, settings.stylesheet_path)
    else:
        return settings.stylesheet

# Return 'stylesheet' or 'stylesheet_path' arguments as list.
#
# The original settings arguments are kept unchanged: you can test
# with e.g. ``if settings.stylesheet_path:``
#
# Differences to ``get_stylesheet_reference``:
# * return value is a list
# * no re-writing of the path (and therefore no optional argument)
#   (if required, use ``utils.relative_path(source, target)``
#   in the calling script)
def get_stylesheet_list(settings):
    """
    Retrieve list of stylesheet references from the settings object.
    """
    assert not (settings.stylesheet and settings.stylesheet_path), (
            'stylesheet and stylesheet_path are mutually exclusive.')
    stylesheets = settings.stylesheet_path or settings.stylesheet or []
    # programmatically set default can be string or unicode:
    if not isinstance(stylesheets, list):
        stylesheets = [path.strip() for path in stylesheets.split(',')]
    # expand relative paths if found in stylesheet-dirs:
    return [find_file_in_dirs(path, settings.stylesheet_dirs)
            for path in stylesheets]

def find_file_in_dirs(path, dirs):
    """
    Search for `path` in the list of directories `dirs`.

    Return the first expansion that matches an existing file.
    """
    if os.path.isabs(path):
        return path
    for d in dirs:
        if d == '.':
            f = path
        else:
            d = os.path.expanduser(d)
            f = os.path.join(d, path)
        if os.path.exists(f):
            return f
    return path

def get_trim_footnote_ref_space(settings):
    """
    Return whether or not to trim footnote space.

    If trim_footnote_reference_space is not None, return it.

    If trim_footnote_reference_space is None, return False unless the
    footnote reference style is 'superscript'.
    """
    if settings.setdefault('trim_footnote_reference_space', None) is None:
        return getattr(settings, 'footnote_references', None) == 'superscript'
    else:
        return settings.trim_footnote_reference_space

def get_source_line(node):
    """
    Return the "source" and "line" attributes from the `node` given or from
    its closest ancestor.
    """
    while node:
        if node.source or node.line:
            return node.source, node.line
        node = node.parent
    return None, None

def escape2null(text):
    """Return a string with escape-backslashes converted to nulls."""
    parts = []
    start = 0
    while True:
        found = text.find('\\', start)
        if found == -1:
            parts.append(text[start:])
            return ''.join(parts)
        parts.append(text[start:found])
        parts.append('\x00' + text[found+1:found+2])
        start = found + 2               # skip character after escape

# `unescape` definition moved to `nodes` to avoid circular import dependency.

def split_escaped_whitespace(text):
    """
    Split `text` on escaped whitespace (null+space or null+newline).
    Return a list of strings.
    """
    strings = text.split('\x00 ')
    strings = [string.split('\x00\n') for string in strings]
    # flatten list of lists of strings to list of strings:
    return list(itertools.chain(*strings))

def strip_combining_chars(text):
    if isinstance(text, str) and sys.version_info < (3, 0):
        return text
    return u''.join([c for c in text if not unicodedata.combining(c)])

def find_combining_chars(text):
    """Return indices of all combining chars in  Unicode string `text`.

    >>> from docutils.utils import find_combining_chars
    >>> find_combining_chars(u'A t̆ab̆lĕ')
    [3, 6, 9]

    """
    if isinstance(text, str) and sys.version_info < (3, 0):
        return []
    return [i for i,c in enumerate(text) if unicodedata.combining(c)]

def column_indices(text):
    """Indices of Unicode string `text` when skipping combining characters.

    >>> from docutils.utils import column_indices
    >>> column_indices(u'A t̆ab̆lĕ')
    [0, 1, 2, 4, 5, 7, 8]

    """
    # TODO: account for asian wide chars here instead of using dummy
    # replacements in the tableparser?
    string_indices = list(range(len(text)))
    for index in find_combining_chars(text):
        string_indices[index] = None
    return [i for i in string_indices if i is not None]

east_asian_widths = {'W': 2,   # Wide
                     'F': 2,   # Full-width (wide)
                     'Na': 1,  # Narrow
                     'H': 1,   # Half-width (narrow)
                     'N': 1,   # Neutral (not East Asian, treated as narrow)
                     'A': 1}   # Ambiguous (s/b wide in East Asian context,
                               # narrow otherwise, but that doesn't work)
"""Mapping of result codes from `unicodedata.east_asian_widt()` to character
column widths."""

def column_width(text):
    """Return the column width of text.

    Correct ``len(text)`` for wide East Asian and combining Unicode chars.
    """
    if isinstance(text, str) and sys.version_info < (3, 0):
        return len(text)
    width = sum([east_asian_widths[unicodedata.east_asian_width(c)]
                 for c in text])
    # correction for combining chars:
    width -= len(find_combining_chars(text))
    return width

def uniq(L):
     r = []
     for item in L:
         if not item in r:
             r.append(item)
     return r

def unique_combinations(items, n):
    """Return `itertools.combinations`."""
    warnings.warn('docutils.utils.unique_combinations is deprecated; '
                  'use itertools.combinations directly.',
                      DeprecationWarning, stacklevel=2)
    return itertools.combinations(items, n)

def normalize_language_tag(tag):
    """Return a list of normalized combinations for a `BCP 47` language tag.

    Example:

    >>> from docutils.utils import normalize_language_tag
    >>> normalize_language_tag('de_AT-1901')
    ['de-at-1901', 'de-at', 'de-1901', 'de']
    >>> normalize_language_tag('de-CH-x_altquot')
    ['de-ch-x-altquot', 'de-ch', 'de-x-altquot', 'de']

    """
    # normalize:
    tag = tag.lower().replace('-', '_')
    # split (except singletons, which mark the following tag as non-standard):
    tag = re.sub(r'_([a-zA-Z0-9])_', r'_\1-', tag)
    subtags = [subtag for subtag in tag.split('_')]
    base_tag = (subtags.pop(0),)
    # find all combinations of subtags
    taglist = []
    for n in range(len(subtags), 0, -1):
        for tags in itertools.combinations(subtags, n):
            taglist.append('-'.join(base_tag+tags))
    taglist += base_tag
    return taglist


class DependencyList(object):

    """
    List of dependencies, with file recording support.

    Note that the output file is not automatically closed.  You have
    to explicitly call the close() method.
    """

    def __init__(self, output_file=None, dependencies=[]):
        """
        Initialize the dependency list, automatically setting the
        output file to `output_file` (see `set_output()`) and adding
        all supplied dependencies.
        """
        self.set_output(output_file)
        for i in dependencies:
            self.add(i)

    def set_output(self, output_file):
        """
        Set the output file and clear the list of already added
        dependencies.

        `output_file` must be a string.  The specified file is
        immediately overwritten.

        If output_file is '-', the output will be written to stdout.
        If it is None, no file output is done when calling add().
        """
        self.list = []
        if output_file:
            if output_file == '-':
                of = None
            else:
                of = output_file
            self.file = docutils.io.FileOutput(destination_path=of,
                                   encoding='utf8', autoclose=False)
        else:
            self.file = None

    def add(self, *filenames):
        """
        If the dependency `filename` has not already been added,
        append it to self.list and print it to self.file if self.file
        is not None.
        """
        for filename in filenames:
            if not filename in self.list:
                self.list.append(filename)
                if self.file is not None:
                    self.file.write(filename+'\n')

    def close(self):
        """
        Close the output file.
        """
        self.file.close()
        self.file = None

    def __repr__(self):
        try:
            output_file = self.file.name
        except AttributeError:
            output_file = None
        return '%s(%r, %s)' % (self.__class__.__name__, output_file, self.list)


release_level_abbreviations = {
    'alpha':     'a',
    'beta':      'b',
    'candidate': 'rc',
    'final':     '',}

def version_identifier(version_info=None):
    """
    Return a version identifier string built from `version_info`, a
    `docutils.VersionInfo` namedtuple instance or compatible tuple. If
    `version_info` is not provided, by default return a version identifier
    string based on `docutils.__version_info__` (i.e. the current Docutils
    version).
    """
    if version_info is None:
        version_info = __version_info__
    if version_info.micro:
        micro = '.%s' % version_info.micro
    else:
        # 0 is omitted:
        micro = ''
    releaselevel = release_level_abbreviations[version_info.releaselevel]
    if version_info.serial:
        serial = version_info.serial
    else:
        # 0 is omitted:
        serial = ''
    if version_info.release:
        dev = ''
    else:
        dev = '.dev'
    version = '%s.%s%s%s%s%s' % (
        version_info.major,
        version_info.minor,
        micro,
        releaselevel,
        serial,
        dev)
    return version