2 # -*- coding: utf-8 -*-
3 # Authors: David Goodger <goodger@python.org>; Ueli Schlaepfer; Günter Milde
4 # Maintainer: docutils-develop@lists.sourceforge.net
5 # Copyright: This module has been placed in the public domain.
8 Transforms needed by most or all documents:
10 - `Decorations`: Generate a document's header & footer.
11 - `Messages`: Placement of system messages stored in
12 `nodes.document.transform_messages`.
13 - `TestMessages`: Like `Messages`, used on test runs.
14 - `FinalReferences`: Resolve remaining references.
17 __docformat__
= 'reStructuredText'
22 from docutils
import nodes
, utils
23 from docutils
.transforms
import TransformError
, Transform
24 from docutils
.utils
import smartquotes
26 class Decorations(Transform
):
29 Populate a document's decoration element (header, footer).
32 default_priority
= 820
35 header_nodes
= self
.generate_header()
37 decoration
= self
.document
.get_decoration()
38 header
= decoration
.get_header()
39 header
.extend(header_nodes
)
40 footer_nodes
= self
.generate_footer()
42 decoration
= self
.document
.get_decoration()
43 footer
= decoration
.get_footer()
44 footer
.extend(footer_nodes
)
46 def generate_header(self
):
49 def generate_footer(self
):
50 # @@@ Text is hard-coded for now.
51 # Should be made dynamic (language-dependent).
52 # @@@ Use timestamp from the `SOURCE_DATE_EPOCH`_ environment variable
54 # See https://sourceforge.net/p/docutils/patches/132/
55 # and https://reproducible-builds.org/specs/source-date-epoch/
56 settings
= self
.document
.settings
57 if settings
.generator
or settings
.datestamp
or settings
.source_link \
58 or settings
.source_url
:
60 if settings
.source_link
and settings
._source \
61 or settings
.source_url
:
62 if settings
.source_url
:
63 source
= settings
.source_url
65 source
= utils
.relative_path(settings
._destination
,
68 nodes
.reference('', 'View document source',
71 if settings
.datestamp
:
72 datestamp
= time
.strftime(settings
.datestamp
, time
.gmtime())
73 text
.append(nodes
.Text('Generated on: ' + datestamp
+ '.\n'))
74 if settings
.generator
:
76 nodes
.Text('Generated by '),
77 nodes
.reference('', 'Docutils', refuri
=
78 'http://docutils.sourceforge.net/'),
80 nodes
.reference('', 'reStructuredText', refuri
='http://'
81 'docutils.sourceforge.net/rst.html'),
82 nodes
.Text(' source.\n')])
83 return [nodes
.paragraph('', '', *text
)]
88 class ExposeInternals(Transform
):
91 Expose internal attributes if ``expose_internals`` setting is set.
94 default_priority
= 840
96 def not_Text(self
, node
):
97 return not isinstance(node
, nodes
.Text
)
100 if self
.document
.settings
.expose_internals
:
101 for node
in self
.document
.traverse(self
.not_Text
):
102 for att
in self
.document
.settings
.expose_internals
:
103 value
= getattr(node
, att
, None)
104 if value
is not None:
105 node
['internal:' + att
] = value
108 class Messages(Transform
):
111 Place any system messages generated after parsing into a dedicated section
115 default_priority
= 860
118 unfiltered
= self
.document
.transform_messages
119 threshold
= self
.document
.reporter
.report_level
121 for msg
in unfiltered
:
122 if msg
['level'] >= threshold
and not msg
.parent
:
125 section
= nodes
.section(classes
=['system-messages'])
126 # @@@ get this from the language module?
127 section
+= nodes
.title('', 'Docutils System Messages')
129 self
.document
.transform_messages
[:] = []
130 self
.document
+= section
133 class FilterMessages(Transform
):
136 Remove system messages below verbosity threshold.
139 default_priority
= 870
142 for node
in self
.document
.traverse(nodes
.system_message
):
143 if node
['level'] < self
.document
.reporter
.report_level
:
144 node
.parent
.remove(node
)
147 class TestMessages(Transform
):
150 Append all post-parse system messages to the end of the document.
152 Used for testing purposes.
155 default_priority
= 880
158 for msg
in self
.document
.transform_messages
:
163 class StripComments(Transform
):
166 Remove comment elements from the document tree (only if the
167 ``strip_comments`` setting is enabled).
170 default_priority
= 740
173 if self
.document
.settings
.strip_comments
:
174 for node
in self
.document
.traverse(nodes
.comment
):
175 node
.parent
.remove(node
)
178 class StripClassesAndElements(Transform
):
181 Remove from the document tree all elements with classes in
182 `self.document.settings.strip_elements_with_classes` and all "classes"
183 attribute values in `self.document.settings.strip_classes`.
186 default_priority
= 420
189 if not (self
.document
.settings
.strip_elements_with_classes
190 or self
.document
.settings
.strip_classes
):
192 # prepare dicts for lookup (not sets, for Python 2.2 compatibility):
193 self
.strip_elements
= dict(
195 for key
in (self
.document
.settings
.strip_elements_with_classes
197 self
.strip_classes
= dict(
198 [(key
, None) for key
in (self
.document
.settings
.strip_classes
200 for node
in self
.document
.traverse(self
.check_classes
):
201 node
.parent
.remove(node
)
203 def check_classes(self
, node
):
204 if isinstance(node
, nodes
.Element
):
205 for class_value
in node
['classes'][:]:
206 if class_value
in self
.strip_classes
:
207 node
['classes'].remove(class_value
)
208 if class_value
in self
.strip_elements
:
211 class SmartQuotes(Transform
):
214 Replace ASCII quotation marks with typographic form.
216 Also replace multiple dashes with em-dash/en-dash characters.
219 default_priority
= 850
221 def __init__(self
, document
, startnode
):
222 Transform
.__init
__(self
, document
, startnode
=startnode
)
223 self
.unsupported_languages
= set()
225 def get_tokens(self
, txtnodes
):
226 # A generator that yields ``(texttype, nodetext)`` tuples for a list
227 # of "Text" nodes (interface to ``smartquotes.educate_tokens()``).
229 texttype
= {True: 'literal', # "literal" text is not changed:
231 for txtnode
in txtnodes
:
232 nodetype
= texttype
[isinstance(txtnode
.parent
,
238 yield (nodetype
, txtnode
.astext())
242 smart_quotes
= self
.document
.settings
.smart_quotes
246 alternative
= smart_quotes
.startswith('alt')
247 except AttributeError:
249 # print repr(alternative)
251 document_language
= self
.document
.settings
.language_code
253 # "Educate" quotes in normal text. Handle each block of text
254 # (TextElement node) as a unit to keep context around inline nodes:
255 for node
in self
.document
.traverse(nodes
.TextElement
):
256 # skip preformatted text blocks and special elements:
257 if isinstance(node
, (nodes
.FixedTextElement
, nodes
.Special
)):
259 # nested TextElements are not "block-level" elements:
260 if isinstance(node
.parent
, nodes
.TextElement
):
263 # list of text nodes in the "text block":
264 txtnodes
= [txtnode
for txtnode
in node
.traverse(nodes
.Text
)
265 if not isinstance(txtnode
.parent
,
266 nodes
.option_string
)]
268 # language: use typographical quotes for language "lang"
269 lang
= node
.get_language_code(document_language
)
270 # use alternative form if `smart-quotes` setting starts with "alt":
272 if '-x-altquot' in lang
:
273 lang
= lang
.replace('-x-altquot', '')
276 # drop subtags missing in quotes:
277 for tag
in utils
.normalize_language_tag(lang
):
278 if tag
in smartquotes
.smartchars
.quotes
:
281 else: # language not supported: (keep ASCII quotes)
282 if lang
not in self
.unsupported_languages
:
283 self
.document
.reporter
.warning('No smart quotes '
284 'defined for language "%s".'%lang
, base_node
=node
)
285 self
.unsupported_languages
.add(lang
)
288 # Iterator educating quotes in plain text:
289 # '2': set all, using old school en- and em- dash shortcuts
290 teacher
= smartquotes
.educate_tokens(self
.get_tokens(txtnodes
),
291 attr
='qDe', language
=lang
)
293 for txtnode
, newtext
in zip(txtnodes
, teacher
):
294 txtnode
.parent
.replace(txtnode
, nodes
.Text(newtext
))
296 self
.unsupported_languages
= set() # reset