forked from attardi/wikiextractor
-
Notifications
You must be signed in to change notification settings - Fork 0
/
ChangeLog
517 lines (343 loc) · 17.6 KB
/
ChangeLog
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
2017-01-15 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (process_dump): use text_type to decide whether
to use decode('utf-8').)
2016-10-29 Giuseppe Attardi <[email protected]>
* setup.py: use scripts instead of console_scripts.
* WikiExtractor.py (pages_from): handle self closing <text />.
2016-08-31 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (findMatchingBraces): fixed ambiguous cases
like {{{!}} {{!}}}
2016-08-30 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (Extractor.wiki2text): drop templates before
dropping tables.
(Extractor.templateParams): fixed RE for separating var from value.
(main): removed option xml_namespaces, since we extract only
articles, i.e. those with namespace 0.
(Extractor.clean): removed option --escapedoc, use --html instead.
* WikiExtractor.py (Extractor.transform): Keep <nowiki> or else
further expansion will be triggered.
2016-08-29 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (callParserFunction): reworked order of
evaluation of template parameters, according to
https://www.mediawiki.org/wiki/Help:Templates#Order_of_evaluation
Template parameters are fully evaluated before they are passed to the template.
The branching parser function are an exception, since evaluation
depends on values tested.
* WikiExtractor.py (callParserFunction): reworked handling of
parser function.
Handle expansion of:
{{numero romano|1999}}
where Template:Numero romano = {{{{{|safesubst:}}}#invoke:Numero romano|main}}
and parameter body has no arguments, by getting them from frame.
Instead expansion of:
{{Str endswith|1999|a.C.}}
where Template:Str endswith = {{#ifeq:{{#invoke:String|sub|s={{{1|}}}| -{{#invoke:String|len|s={{{2}}}}} |ignore_errors=true}}|{{{2|}}}|sì}}
uses parameters in template.
* WikiExtractor.py (Extractor.extract): Adds an h1 tag for title if --html is
passed. Patch contributed by okb1100 on GitHub.
* WikiExtractor.py (modules): added partial emulation of modules String and Roman.
)
2016-08-26 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (nowiki): handle <nowiki>...</nowiki>.
2016-08-22 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (Extractor.extract): fixed UTF-8 encoding with
option -a.
2016-08-19 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (Extractor): fixed misspelled keepLists.
2016-08-11 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (Extract): made escape_doc a class variable.
2016-06-20 Seth Cleveland <[email protected]>
* WikiExtractor.py: Fix string encoding error for python 2 and log
extract exceptions.
2016-06-19 Giuseppe Attardi <[email protected]>
* WikiExtractor.py: support for Python 3 by [email protected]
2016-03-23 Seth Cleveland <[email protected]>
* WikiExtractor.py: add filtering options -- min text length, xml
namespace, and disambig pages. Added option to print the revision
id with each document.
2016-03-23 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (compact): properly emit section headers when Extractor.keepSections.
2016-03-19 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (main): restored option --sections.
2016-03-12 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (findBalanced): fix to bad single tuple
parameter to findBalanced.
2016-03-06 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (process_dump): handle templates with stdin.
2016-02-20 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (reduce_process): tell mapper to wait when
spool gets too long.
2016-02-19 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (extract_process): use out.truncate() instread
of out.close()
2016-02-15 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (Extractor.clean): turned into method.
(process_dump): control spool length to avoid filing up memory.
2016-02-14 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (load_templates): save also <id> in templates.
(pages_from): common iterator for extracting pages from dump, used
both for analyzing pages, templates and single articles.
2016-02-13 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (reduce_process): close file here.
* extractPage.py (process_data): allow range of ids.
2016-02-12 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (reduce_process): moved here creation of OutputSplitter.
(compact): Extractor.keepLists allows preserving lists in output.
(main): added new option --lists for preserving lists in output.
(compact): rest lislLevel entering new section.
(ignoredTags): removed 'div' from here, since it is in discardedTags.
(ignoredTags): moved 'sub' and 'sup' to discardedTags.
2016-02-11 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (if_empty): added implementation of Lua module
for "Template:If empty".
(compact): preserve lists also in text mode.
(process_dump): log ID and title of pages.
2016-02-10 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (process_dump): detect templates by element
<ns>10</ns> rather than by colon in title.
(Extractor.templateParams): restore space in RE [^= ]:
m = re.match(' *([^= ]*?) *=(.*)', param, re.DOTALL)
2016-02-04 Giuseppe Attardi <[email protected]>
* cirrus-extract.py: added.
* README.md: added mention of Cirrus extract.
2015-11-20 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (makeExternalLink): fixed.
(main): dropped confusing option --section.
2015-10-29 Giuseppe Attardi <[email protected]>
* WikiExtractor.py: /usr/bin/env
2015-09-29 Giuseppe Attardi <[email protected]>
* WikiExtractor.py: fixed argparse default.
(load_templates): save also modules, for a future release capaple
of applying them.
2015-09-14 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (Extractor.templateParams): allow space in key
(process_dump): terminate by joining processes rather than queues.
* WikiExtractor.py (process_dump): eliminated ordering_queue in
favor of termination signals to jobs_queue.
2015-09-13 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (process_dump): queue.put() dropped second arg
True, since it is the default.
(process_dump): queue renamed to jobs_queue.
(main): restored default bytes to 1M.
Dropped confusing option to write to a single file: use -o - for that.
2015-08-30 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (main): check presence of title element in
single article.
(load_templates): reconstruct the template namespace from the
title of the first template in the saved templates.
(define_template): match also #redirect as used in French.
2015-06-02 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (Extractor.expandTemplate): extend frame before
subst(), since there may be recursion in default parameter value,
e.g. {{OTRS|celebrative|date=April 2015}} in article 21637542 in
enwiki.
2015-05-06 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (main): fixed arg.namespaces.
(compact): use fillvalue=' ' in izip_longest.
2015-04-26 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (clean): use re.U when matching \W or chinese
characters will be lost.
2015-04-25 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (sharp_expr): use unicode() instread od str()
or else chinese article 596814 fails.
(splitParameters): use findMatchingBraces istead of findBalanced,
to properly handle {{{4|{{{{{subst|}}}CURRENTYEAR}}}}
(Extractor.extract): handle currentyear, currentmonth and currentday
2015-04-23 Giuseppe Attardi <[email protected]>
* WikiExtractor.py: make UTF-8 the default encoding
2015-04-22 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (replaceInternalLinks): function for replacing
internal links, modeled after MediaWiki original.
(replaceExternalLinks): revised taking into account the former.
(replaceInternalLinks): fix to nested iterator.
2015-04-20 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (findBalanced): dropped optional arguments.
(make_anchor_tag): dont use splitParameters() since we must
consider also single [, which do not count ib parameters.
(sharp_switch): use split() to split at fist = sign.
(main): grouped command options.
(main): removed option -B.
(Template): class used to represent templates. It provides a method
for parameter subsitution.
Templats are parsed on demand when used and stored in a cache.
2015-04-19 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (clean): use dropNext to drop discardElements.
(discardElements): added div.
(compact): avoid duplicated header line with optio --sections.
(compact): skip empty list items.
(main): changed logger format.
(ignoredTags): added abbr.
(clean): handle extension SyntaxHighlight.
(main): use % parameters in logging.
(main): added option --html, that subsumes --links and --sections
for producing HTML output instead of plain text.
(Extractor): moved here variables keepLinks, keepSections, toHTML.
(compact): modified to produce HTML lists.
2015-04-18 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (compact): strip lists of initial characters.
2015-04-17 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (sharp_switch): fixed handling of default.
(uc, lc, ucfirst, lcfirst): parser functions.
2015-04-16 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (MagicWords): dealing with MagicWords.
This required rewriting code as methods of class Extractor, since
some MagicWords are document related and are being handled in
different threads, hence thay canot be globals.
(findMatchingBraces): fix to ambiguities resolution.
(clean): fixed dealing with trail for make_anchor-tag()
(substParameter): expand defaultValue only if used.
(selfClosing_tag_patterns): avoid matching besides tag end..
2015-04-15 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (expandTemplates): increase depth only when
calling expandTemplate()
(define_template): removed \n in onlyincludeAccumulator, drop
<noinclude> always.
(sharp_invoke): restored support for #invoke, by adding parameter
frame to expandTemplate.
(main): allow specifying G in --bytes.
(make_anchor_tag): urlencode link.
(wikiLink): properly match anchor.
(make_anchor_tag): use splitParameters to separate parts of link.
(splitParameters): include pair [], since arguments may contain
wikilinks.
2015-04-14 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (clean): dropped removal of preformatted lines,
since it is hard to distinguich them, since templates may
introdice lines with starting blanks.
(discardElements) added 'small'.
(ignoredTags) removed 'small'.
(make_anchor_tag): fixed RE for wikiLink.
(sharp_expr): added infix operators.
(Infix): support for infix operators.
(Extractor.extract): moved here logging of document being processed.
(clean): rewritten handling of wikilinks since using RE is to slow.
(maxTemplateRecursionLevels): increased to 30.
2015-04-13 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (findMatchingBraces): rewritten to handle
ambiguities.
(substParameter): only evaluate name and default.
(main): fixed option --article.
2015-04-12 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (ExtractorThread): enabled multithread version.
(findMatchingBraces): handle isolated braces.
(expandTemplates): recurse on result from expandTemplate().
2015-04-11 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (sharp_switch): deal properly with #default.
(OutputSplitter): update to new-style classes
* WikiExtractor.py (selfClosingTags): added nowiki.
* WikiExtractor.py (bold_italic, bold): allow single quote inside,
e.g. '''[[Chinese New Year|New Year's Eve]]'''.
* WikiExtractor.py (templateParams): fix pattern to match
parameter name.
* WikiExtractor.py (substParameter): use splitParameters()
* WikiExtractor.py (main): added --no-templates option.
* WikiExtractor.py (substParameter): added parameter param_depth
to control depth of parameter expansion, separately from depth,
used for template expansion.
2015-04-10 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (callParserFunction): return '' also in case of
failure.
2015-04-09 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (expandTemplates): replaced frame parameter with
depth, used to limit max template recursion.
2015-04-07 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (main): added --debug option.
2015-01-24 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (splitParameters): rewritten template
processing by performing proper parsing of all balanced
expressions in templates invocation and expansion, using iterator
findBalanced().
2015-01-18 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (expandTemplates): template expansion now working.
2015-01-11 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (externalLink): replaced .* with appropriate
[^x]* here and elsewhere.
2015-01-10 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (main): added option --article for processing a
single article.
(main): get dump rm file rather than frpm stdin, so that
preprocessing does not need to save data to temp file.
2014-02-25 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (ignoreTag): make / optional.
2013-12-15 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (clean): added template expansion
2013-10-14 Giuseppe Attardi <[email protected]>
* WikiExtractor.py: added wiktionary and wikt to the namespaces
(used e.g. in http:https://en.wikipedia.org/wiki?curid=12)
2013-05-09 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (main): handle properly keepLinks option.
2013-04-05 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (compact): keep lines ending with ':'.
2013-04-02 Giuseppe Attardi <[email protected]>
* WikiExtractor.py: obtain prefix from dump.
2013-01-27 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (WikiDocument): add newline after <doc>.
Release version 2.3.
2012-12-30 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (process_data): added patch by Humberto
Pereira, who claims a 10x improvement in speed.
(main): added option to set acceptedNamespaces
2012-11-01 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (get_url): create URL from Id instead than from title.
2012-06-28 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (OutputSplitter.reserve): added method to
invoke before writing.
(WikiDocument): use reserve() before writing whole page.
(main): added version number and option -v.
2012-05-17 Giuseppe Attardi <[email protected]>
* WikiExtractor.py (main): added option to preserve sections as
HTML headers and lists as <LI>.
2012-05-08 Giuseppe Attardi <[email protected]>
* WikiExtractor.py: Released version 2.0.
* test/test.xml: added sample to test hard cases for extractor.
* WikiExtractor.py (dropNested): Completely rewritten to be more
compliant to WikiMedia Markup language.
Use proper parsing fuctions to handle nested structures.
Improved performance by reducing creation of lists and strings.
Use htmlentitydefs instead of hand crafted list.
Added parameter -b to set URL for site.
Extensive use of RegExpr instead of specific string tests.
Deal with preformatted text.
Added parameter accepetedNamespaces to select namespaces to retain
in page titles or wiki links.
TODO:
1. handle Template expansion. See WikiPrep
(http:https://www.cs.technion.ac.il/~gabr/resources/code/wikiprep/)
2. Use full parser in order to better deal with nested and
unbalanced expressions.
2012-02-15 Stefano Dei Rossi <[email protected]>
* WikiExtractor.py (WikiExtractor): replaced with a simple space
instead of u'\u00A0'.
2012-02-10 Giuseppe Attardi <[email protected]>
* WikiExtractor.py: added Copyright.
2011-11-03 Antonio Fuschetto <[email protected]>
* WikiExtractor.py: updated version to 1.6 (Oct 17).
2011-10-17 Giuseppe Attardi <[email protected]>
* WikiExtractor.py: turned prefix into a parameter.
2011-07-29 Antonio Fuschetto <[email protected]>
* WikiExtractor.py (init): fixed bugs in apostrophe_bold_pattern and
apostrophe_italic_pattern.
2011-07-28 Antonio Fuschetto <[email protected]>
* WikiExtractor.py (__garbage_namespaces): added "file" namespace to
remove list.
2011-07-10 Antonio Fuschetto <[email protected]>
* WikiExtractor.py (get_wiki_document_url): changed the handling of
URL prefix (anchors don't use prefix but a relative URLs).
2011-06-26 Antonio Fuschetto <[email protected]>
* WikiExtractor.py (extract_document): changed the handling of
wikilinks, adding an anchor tag for each link with a reference to the
Wikipedia document.
* WikiExtractor.py (WikiExtractor): changed the handling of
placeholders: from "[Formula 12]" to "formula_12".
2011-04-06 Antonio Fuschetto <[email protected]>
* WikiExtractor.py (init): fixed bugs in apostrophe_bold_pattern and
apostrophe_italic_pattern.
* WikiExtractor.py (compact): drop lines ending with ':'
(these are sentences preceding list items); fixed some bugs.
* WikiExtractor.py: released version 1.1.
2011-03-12 Antonio Fuschetto <[email protected]>
* wiki-extractor.py (main): removed the sentence splitting option.
* wiki-extractor.py: fixed some bugs; released version 1.0; changed
filename to "Wiki-Extractor.py" according to Tanl module names.
2011-03-01 Antonio Fuschetto <[email protected]>
* wiki-extractor.py (main): added cross platform path management.