Debian Codegraph @

CHANGELOG

raw
1 v0.6.6 (released January 4, 2024):
2
3 - Dropped support for end-of-life Python 3.7.
4 - Added support for Python 3.12.
5 - Raised nested node depth limit from 40 to 100 to match current MediaWiki. (#313)
6
7 v0.6.5 (released September 5, 2023):
8
9 - Dropped support for end-of-life Python 3.6.
10 - Added support for Python 3.11.
11 - Fixed parsing of leading zeros in named HTML entities. (#288)
12 - Fixed memory leak parsing tags. (#303)
13 - Fixed pickling SmartList objects. (#289)
14
15 v0.6.4 (released February 14, 2022):
16
17 - Dropped support for end-of-life Python 3.5.
18 - Added support for Python 3.10. (#278)
19 - Fixed a regression in v0.6.2 that broke parsing of nested wikilinks in file
20 captions. For now, the parser will interpret nested wikilinks in normal links
21 as well, even though this differs from MediaWiki. (#270)
22
23 v0.6.3 (released September 2, 2021):
24
25 - Added Linux AArch64 wheels. (#276)
26 - Fixed C integer conversion, manifesting as parsing errors on big-endian
27 platforms. (#277)
28
29 v0.6.2 (released May 16, 2021):
30
31 - Improved parsing of external links. (#232)
32 - Fixed parsing of nested wikilinks.
33 - Ported tests to pytest. (#237)
34 - Moved mwparserfromhell package to src/ dir.
35 - There was no 0.6.1 release due to a packaging error.
36
37 v0.6 (released December 21, 2020):
38
39 Thanks to everyone for their patience with this release!
40
41 - Breaking change: dropped support for end-of-life Python 2.7 and 3.4.
42 - Added support for Python 3.8 and 3.9.
43 - Added binary wheels for Linux and macOS.
44 - Updated Wikicode.matches() to recognize underscores as being equivalent
45 to spaces. (#216)
46 - Added a 'default' parameter to Template.get(), and implement dict-style item
47 access for template parameters. (#252)
48 - Fixed a rare parsing bug involving deeply nested style tags. (#224)
49 - Fixed parsing of section headings inside templates. (#233)
50 - Updated HTML tag definitions.
51 - Internal refactoring and cleanup.
52
53 v0.5.4 (released May 15, 2019):
54
55 - Fixed an unlikely crash in the C tokenizer when interrupted while parsing
56 a heading.
57
58 v0.5.3 (released March 30, 2019):
59
60 - Fixed manual construction of Node objects, previously unsupported. (#214)
61 - Fixed Wikicode transformation methods (replace(), remove(), etc.) when passed
62 an empty section as an argument. (#212)
63 - Fixed the parser getting stuck inside malformed tables. (#206)
64
65 v0.5.2 (released November 1, 2018):
66
67 - Dropped support for end-of-life Python versions 2.6, 3.2, 3.3. (#199, #204)
68 - Fixed signals getting stuck inside the C tokenizer until parsing finishes,
69 in pathological cases. (#206)
70 - Fixed <wbr> not being considered a single-only tag. (#200)
71 - Fixed a C tokenizer crash on Python 3.7 when compiled with assertions. (#208)
72 - Cleaned up some minor documentation issues. (#207)
73
74 v0.5.1 (released March 3, 2018):
75
76 - Improved behavior when adding parameters to templates (via Template.add())
77 with poorly formatted whitespace conventions. (#185)
78 - Fixed the parser getting stuck in deeply nested HTML tags with unclosed,
79 quoted attributes. (#190)
80
81 v0.5 (released June 23, 2017):
82
83 - Added Wikicode.contains() to determine whether a Node or Wikicode object is
84 contained within another Wikicode object.
85 - Added Wikicode.get_ancestors() and Wikicode.get_parent() to find all
86 ancestors and the direct parent of a Node, respectively.
87 - Fixed a long-standing performance issue with deeply nested, invalid syntax
88 (issue #42). The parser should be much faster on certain complex pages. The
89 "max cycle" restriction has also been removed, so some situations where
90 templates at the end of a page were being skipped are now resolved.
91 - Made Template.remove(keep_field=True) behave more reasonably when the
92 parameter is already empty.
93 - Added the keep_template_params argument to Wikicode.strip_code(). If True,
94 then template parameters will be preserved in the output.
95 - Wikicode objects can now be pickled properly (fixed infinite recursion error
96 on incompletely-constructed StringMixIn subclasses).
97 - Fixed Wikicode.matches()'s behavior on iterables besides lists and tuples.
98 - Fixed len() sometimes raising ValueError on empty node lists.
99 - Fixed a rare parsing bug involving self-closing tags inside the attributes of
100 unpaired tags.
101 - Fixed release script after changes to PyPI.
102
103 v0.4.4 (released December 30, 2016):
104
105 - Added support for Python 3.6.
106 - Fixed parsing bugs involving:
107 - wikitables nested in templates;
108 - wikitable error recovery when unable to recurse;
109 - templates nested in template parameters before other parameters.
110 - Fixed parsing file-like objects.
111 - Made builds deterministic.
112 - Documented caveats.
113
114 v0.4.3 (released October 29, 2015):
115
116 - Added Windows binaries for Python 3.5.
117 - Fixed edge cases involving wikilinks inside of external links and vice versa.
118 - Fixed a C tokenizer crash when a keyboard interrupt happens while parsing.
119
120 v0.4.2 (released July 30, 2015):
121
122 - Fixed setup script not including header files in releases.
123 - Fixed Windows binary uploads.
124
125 v0.4.1 (released July 30, 2015):
126
127 - The process for building Windows binaries has been fixed, and these should be
128 distributed along with new releases. Windows users can now take advantage of
129 C speedups without having a compiler of their own.
130 - Added support for Python 3.5.
131 - '<' and '>' are now disallowed in wikilink titles and template names. This
132 includes when denoting tags, but not comments.
133 - Fixed the behavior of preserve_spacing in Template.add() and keep_field in
134 Template.remove() on parameters with hidden keys.
135 - Removed _ListProxy.detach(). SmartLists now use weak references and their
136 children are garbage-collected properly.
137 - Fixed parser bugs involving:
138 - templates with completely blank names;
139 - templates with newlines and comments.
140 - Heavy refactoring and fixes to the C tokenizer, including:
141 - corrected a design flaw in text handling, allowing for substantial speed
142 improvements when parsing long strings of plain text;
143 - implemented new Python 3.3 PEP 393 Unicode APIs.
144 - Fixed various bugs in SmartList, including one that was causing memory issues
145 on 64-bit builds of Python 2 on Windows.
146 - Fixed some bugs in the release scripts.
147
148 v0.4 (released May 23, 2015):
149
150 - The parser now falls back on pure Python mode if C extensions cannot be
151 built. This fixes an issue that prevented some Windows users from installing
152 the parser.
153 - Added support for parsing wikicode tables (patches by David Winegar).
154 - Added a script to test for memory leaks in scripts/memtest.py.
155 - Added a script to do releases in scripts/release.sh.
156 - skip_style_tags can now be passed to mwparserfromhell.parse() (previously,
157 only Parser().parse() allowed it).
158 - The 'recursive' argument to Wikicode's filter methods now accepts a third
159 option, RECURSE_OTHERS, which recurses over all children except instances of
160 'forcetype' (for example, `code.filter_templates(code.RECURSE_OTHERS)`
161 returns all un-nested templates).
162 - The parser now understands HTML tag attributes quoted with single quotes.
163 When setting a tag attribute's value, quotes will be added if necessary. As
164 part of this, Attribute's 'quoted' attribute has been changed to 'quotes',
165 and is now either a string or None.
166 - Calling Template.remove() with a Parameter object that is not part of the
167 template now raises ValueError instead of doing nothing.
168 - Parameters with non-integer keys can no longer be created with
169 'showkey=False', nor have the value of this attribute be set to False later.
170 - _ListProxy.destroy() has been changed to _ListProxy.detach(), and now works
171 in a more useful way.
172 - If something goes wrong while parsing, ParserError will now be raised.
173 Previously, the parser would produce an unclear BadRoute exception or allow
174 an incorrect node tree to be build.
175 - Fixed parser bugs involving:
176 - nested tags;
177 - comments in template names;
178 - tags inside of <nowiki> tags.
179 - Added tests to ensure that parsed trees convert back to wikicode without
180 unintentional modifications.
181 - Added support for a NOWEB environment variable, which disables a unit test
182 that makes a web call.
183 - Test coverage has been improved, and some minor related bugs have been fixed.
184 - Updated and fixed some documentation.
185
186 v0.3.3 (released April 22, 2014):
187
188 - Added support for Python 2.6 and 3.4.
189 - Template.has() is now passed 'ignore_empty=False' by default instead of True.
190 This fixes a bug when adding parameters to templates with empty fields, and
191 is a breaking change if you rely on the default behavior.
192 - The 'matches' argument of Wikicode's filter methods now accepts a function
193 (taking one argument, a Node, and returning a bool) in addition to a regex.
194 - Re-added 'flat' argument to Wikicode.get_sections(), fixed the order in which
195 it returns sections, and made it faster.
196 - Wikicode.matches() now accepts a tuple or list of strings/Wikicode objects
197 instead of just a single string or Wikicode.
198 - Given the frequency of issues with the (admittedly insufficient) tag parser,
199 there's a temporary skip_style_tags argument to parse() that ignores '' and
200 ''' until these issues are corrected.
201 - Fixed a parser bug involving nested wikilinks and external links.
202 - C code cleanup and speed improvements.
203
204 v0.3.2 (released September 1, 2013):
205
206 - Added support for Python 3.2 (along with current support for 3.3 and 2.7).
207 - Renamed Template.remove()'s first argument from 'name' to 'param', which now
208 accepts Parameter objects in addition to parameter name strings.
209
210 v0.3.1 (released August 29, 2013):
211
212 - Fixed a parser bug involving URLs nested inside other markup.
213 - Fixed some typos.
214
215 v0.3 (released August 24, 2013):
216
217 - Added complete support for HTML Tags, including forms like <ref>foo</ref>,
218 <ref name="bar"/>, and wiki-markup tags like bold ('''), italics (''), and
219 lists (*, #, ; and :).
220 - Added support for ExternalLinks (http://example.com/ and
221 [http://example.com/ Example]).
222 - Wikicode's filter methods are now passed 'recursive=True' by default instead
223 of False. This is a breaking change if you rely on any filter() methods being
224 non-recursive by default.
225 - Added a matches() method to Wikicode for page/template name comparisons.
226 - The 'obj' param of Wikicode.insert_before(), insert_after(), replace(), and
227 remove() now accepts other Wikicode objects and strings representing parts of
228 wikitext, instead of just nodes. These methods also make all possible
229 substitutions instead of just one.
230 - Renamed Template.has_param() to has() for consistency with Template's other
231 methods; has_param() is now an alias.
232 - The C tokenizer extension now works on Python 3 in addition to Python 2.7.
233 - Various bugfixes, internal changes, and cleanup.
234
235 v0.2 (released June 20, 2013):
236
237 - The parser now fully supports Python 3 in addition to Python 2.7.
238 - Added a C tokenizer extension that is significantly faster than its Python
239 equivalent. It is enabled by default (if available) and can be toggled by
240 setting `mwparserfromhell.parser.use_c` to a boolean value.
241 - Added a complete set of unit tests covering parsing and wikicode
242 manipulation.
243 - Renamed Wikicode.filter_links() to filter_wikilinks() (applies to ifilter as
244 well).
245 - Added filter methods for Arguments, Comments, Headings, and HTMLEntities.
246 - Added 'before' param to Template.add(); renamed 'force_nonconformity' to
247 'preserve_spacing'.
248 - Added 'include_lead' param to Wikicode.get_sections().
249 - Removed 'flat' param from Wikicode.get_sections().
250 - Removed 'force_no_field' param from Template.remove().
251 - Added support for Travis CI.
252 - Added note about Windows build issue in the README.
253 - The tokenizer will limit itself to a realistic recursion depth to prevent
254 errors and unreasonably long parse times.
255 - Fixed how some nodes' attribute setters handle input.
256 - Fixed multiple bugs in the tokenizer's handling of invalid markup.
257 - Fixed bugs in the implementation of SmartList and StringMixIn.
258 - Fixed some broken example code in the README; other copyedits.
259 - Other bugfixes and code cleanup.
260
261 v0.1.1 (released September 21, 2012):
262
263 - Added support for Comments (<!-- foo -->) and Wikilinks ([[foo]]).
264 - Added corresponding ifilter_links() and filter_links() methods to Wikicode.
265 - Fixed a bug when parsing incomplete templates.
266 - Fixed strip_code() to affect the contents of headings.
267 - Various copyedits in documentation and comments.
268
269 v0.1 (released August 23, 2012):
270
271 - Initial release.