5 :Author: David Goodger; open to all Docutils developers
6 :Contact: goodger@python.org
9 :Copyright: This document has been placed in the public domain.
11 .. _Docutils: http://docutils.sourceforge.net/
14 Bugs in Docutils?!? Yes, we do have a few. Some are old-timers that
15 tend to stay in the shadows and don't bother anybody. Once in a while
16 new bugs are born. From time to time some bugs (new and old) crawl
17 out into the light and must be dealt with. Icky.
19 This document describes how to report a bug, and lists known bugs.
27 If you think you've discovered a bug, please read through these
28 guidelines before reporting it.
30 First, make sure it's a new bug:
32 * Please check the list of `known bugs`_ below and the `SourceForge
33 Bug Tracker`_ to see if it has already been reported.
35 * Are you using the very latest version of Docutils? The bug may have
36 already been fixed. Please get the latest version of Docutils from
37 the repository_ or from the current snapshot_ and check again. Even
38 if your bug has not been fixed, others probably have, and you're
39 better off with the most up-to-date code.
41 If you don't have time to check the latest snapshot, please report
42 the bug anyway. We'd rather tell you that it's already fixed than
43 miss reports of unfixed bugs.
45 * If Docutils does not behave the way you expect, look in the
46 documentation_ (don't forget the FAQ_!) and `mailing list archives`_
47 for evidence that it should behave the way you expect.
49 If you're not sure, please ask on the Docutils-users_ mailing list
52 If it's a new bug, the most important thing you can do is to write a
53 simple description and a recipe that reproduces the bug. Try to
54 create a minimal document that demonstrates the bug. The easier you
55 make it to understand and track down the bug, the more likely a fix
58 Now you're ready to write the bug report. Please include:
60 * A clear description of the bug. Describe how you expected Docutils
61 to behave, and contrast that with how it actually behaved. While
62 the bug may seem obvious to you, it may not be so obvious to someone
63 else, so it's best to avoid a guessing game.
65 * A complete description of the environment in which you reproduced
68 - Your operating system & version.
69 - The version of Python (``python -V``).
70 - The version of Docutils (use the "-V" option to most Docutils
72 - Any private modifications you made to Docutils.
73 - Anything else that could possibly be relevant. Err on the side
74 of too much information, rather than too little.
76 * A literal transcript of the *exact* command you ran, and the *exact*
77 output. Use the "--traceback" option to get a complete picture.
79 * The exact input and output files. Better to attach complete files
80 to your bug report than to include just a summary or excerpt.
82 * If you also want to include speculation as to the cause, and even a
83 patch to fix the bug, that would be great!
85 The best place to send your bug report is to the `SourceForge Bug
86 Tracker`_. That way, it won't be misplaced or forgotten. In fact, an
87 open bug report on SourceForge is a constant irritant that begs to be
92 (This section was inspired by the `Subversion project's`__ BUGS__
95 __ http://subversion.tigris.org/
96 __ http://svn.collab.net/viewcvs/svn/trunk/BUGS?view=markup
98 .. _CVS: http://sourceforge.net/cvs/?group_id=38414
99 .. _repository: docs/dev/repository.html
100 .. _snapshot: http://docutils.sourceforge.net/#download
101 .. _documentation: docs/
103 .. _mailing list archives: http://docutils.sf.net/#mailing-lists
104 .. _Docutils-users: docs/user/mailing-lists.html#docutils-users
105 .. _SourceForge Bug Tracker:
106 http://sourceforge.net/tracker/?group_id=38414&atid=422030
112 Also see the `SourceForge Bug Tracker`_.
114 * The HTML writer generates invalid XHTML for _`centered images with
117 .. image:: example.png
119 :target: http://docutils.sourceforge.net/
121 This results in ``<a><div><img /></div></a>``, which is invalid.
123 * .. _error reporting:
125 Calling rst2s5.py with a non-existent theme (``--theme
126 does_not_exist``) or a non-existent language (``--language zz``)
127 causes exceptions. Such errors should be handled more gracefully.
129 * There's a problem with _`hyperlink references in substitutions`::
133 .. |rst| replace:: reStructuredText_
135 .. _reStructuredText: http://docutils.sourceforge.net/rst.html
137 There is a preferred alternative::
141 .. |rst| replace:: reStructuredText
142 .. _rst: http://docutils.sourceforge.net/rst.html
144 This only works for the case where the entire substitution is a
145 reference, not when the reference is embedded as part of the
148 The problem pattern has never been tested for, perhaps because of
149 the alternative. Unit tests should be added.
151 This was not a bug in Docutils 0.3.7 (`reference
152 <http://thread.gmane.org/gmane.text.docutils.user/2997/focus=2997>`__).
154 * The "stylesheet" setting (a URL, to be used verbatim) should be
155 allowed to be combined with "embed_stylesheet". The stylesheet data
156 should be read in using urllib. There was an assumption that a
157 stylesheet to be embedded should exist as a file on the local
158 system, and only the "stylesheet_path" setting should be used.
160 * ``utils.relative_path()`` sometimes returns absolute _`paths on
161 Windows` (like ``C:/test/foo.css``) where it could have chosen a
164 Furthermore, absolute pathnames are inserted verbatim, like
165 ``href="C:/test/foo.css"`` instead of
166 ``href="file:///C:/test/foo.css"``.
168 For details, see `this posting by Alan G. Isaac
169 <http://article.gmane.org/gmane.text.docutils.user/1569>`_.
171 * _`Line numbers` in system messages are inconsistent in the parser.
173 - In text inserted by the "include" directive, errors are often not
174 reported with the correct "source" or "line" numbers. Perhaps all
175 Reporter calls need "source" and "line" keyword arguments.
176 Elements' .line assignments should be checked. (Assign to .source
177 too? Add a set_info method? To what?) There's a test in
178 test/test_parsers/test_rst/test_directives/test_include.py.
180 - Some line numbers in elements are not being set properly
181 (explicitly), just implicitly/automatically. See rev. 1.74 of
182 docutils/parsers/rst/states.py for an example of how to set.
184 - The line numbers of definition list items are wrong::
186 $ rst2pseudoxml.py --expose-internal-attribute line
194 <document source="<stdin>">
196 <definition_list_item internal:line="3">
200 <paragraph internal:line="2">
203 <definition_list_item internal:line="6">
207 <paragraph internal:line="6">
213 Quite a few nodes are getting a "None" source attribute as well. In
214 particular, see the bodies of definition lists.
216 * Footnote label "5" should be "4" when processing the following
219 ref [#abc]_ [#]_ [1]_ [#4]_
228 <document source="<stdin>">
231 <footnote_reference auto="1" ids="id1" refid="abc">
234 <footnote_reference auto="1" ids="id2" refid="id5">
237 <footnote_reference ids="id3" refid="id6">
240 <footnote_reference auto="1" ids="id4" refid="id7">
242 <footnote auto="1" backrefs="id1" ids="abc" names="abc">
247 <footnote auto="1" backrefs="id2" ids="id5" names="3">
252 <footnote backrefs="id3" ids="id6" names="1">
257 <footnote auto="1" backrefs="id4" ids="id7" names="4">
263 * IDs are based on names. Explicit hyperlink targets have priority
264 over implicit targets. But if an explicit target comes after an
265 implicit target with the same name, the ID of the first (implicit)
266 target remains based on the implicit name. Since HTML fragment
267 identifiers based on the IDs, the first target keeps the name. For
280 text with a reference to contents_ and section_
284 This paragraph is explicitly targeted with the name "section".
286 When processed to HTML, the 2 internal hyperlinks (to "contents" &
287 "section") will work fine, but hyperlinks from outside the document
288 using ``href="...#contents"`` and ``href="...#section"`` won't work.
289 Such external links will connect to the implicit targets (table of
290 contents and "Section" title) instead of the explicit targets
291 ("Subsection" title and last paragraph).
293 Hyperlink targets with duplicate names should be assigned new IDs
294 unrelated to the target names (i.e., "id"-prefix serial IDs).
296 * The "contents" ID of the local table of contents in
297 ``test/functional/expected/standalone_rst_pseudoxml.txt`` is lost in
299 ``test/functional/expected/standalone_rst_html4css1.html``.
301 * _`Blank first columns` in simple tables with explicit row separators
302 silently swallow their input. They should at least produce system
303 error messages. But, with explicit row separators, the meaning is
304 unambiguous and ought to be supported::
306 ============== ==========
307 Table with row separators
308 ============== ==========
310 -------------- ----------
312 -------------- ----------
314 -------------- ----------
316 ============== ==========
318 Added a commented-out test case to
319 test/test_parsers/test_rst/test_SimpleTableParser.py.
321 * _`Footnote references with hyperlink targets` cause a possibly
322 invalid node tree and make the HTML writer crash::
328 <document source="<stdin>">
330 <footnote_reference ids="id1" refuri="URI">
332 <target ids="id2" names="1" refuri="URI">
334 * Anonymous references have "name" attributes. Should they? Are they
335 used? See ``test/test_parsers/test_rst/test_inline_markup.py``.
337 * <reference> elements have a "name" attribute, not "names". The
338 attribute should be "names"; this is an inconsistency.
340 * Setting the _`halt level` should implicitly set the report level to
341 at most the halt level; otherwise Docutils might halt without
342 providing a descriptive error message::
344 ~ $ rst2pseudoxml.py --halt=1 --no-traceback
346 Exiting due to level-1 (INFO) system message.
352 indent-tabs-mode: nil
353 sentence-end-double-space: t