Debian Codegraph @

ChangeLog

raw ◀ ▶
1 Debian patch touch this file
•1 2017-03-25 Dimitrij Mijoski <dmjpp at hotm>:
2 * Version 1.6.1
3 * Library changes:
4 * Performance improvements in suggest()
5 * Fixes regressions for Hungarian related to compounding.
6 * Fixes regressions for Korean related to ICONV.
7 * Command line tool:
8 * Added Tajik translation
9 * Fix regarding serching of OOo dicts installed in user folder
10 * Manpages:
11 * Fix microsoft-cp1251 to cp1251. Dicts should not use the first.
12 * Typos.
13
14 2016-12-22 Dimitrij Mijoski <dmjpp at hotm>:
15 * Version 1.6.0
16 * Changes in the library:
17 * Performance improvement in ngsuggest(), suggestions should be faster.
18 * Revert MAXWORDLEN to 100 as in 1.3.3 for performance reasons.
19 * MAXWORDLEN can be set during build time with -D defines.
20 * Fix crash when word with 102 consecutive X is spelled.
21 * Changes in the command line tool:
22 * -D shows all loaded dictionares insted of only the first.
23 * -D properly lists all available dictionaries on Windows.
24
25 2016-11-30 Dimitrij Mijoski <dmjpp at hotm>:
26 * Version 1.5.4
27 * Fixes the command COMPOUNDSYLLABLE used in Hungarian dictionary.
28
29 2016-11-28 Dimitrij Mijoski <dmjpp at hotm>:
30 * Version 1.5.3
31 * Removed a #include from hunspell.hxx that was creating trouble
32
33 2016-11-27 Dimitrij Mijoski <dmjpp at hotm>:
34 * Version 1.5.2
35 * Reverted full backward compatibility with 1.4 public API, again
36
37 2016-11-27 Dimitrij Mijoski <dmjpp at hotm>:
38 * Version 1.5.1
39 * Reverted full backward compatibility with 1.4 public API
40
41 2016-11-26 Dimitrij Mijoski <dmjpp at hotm>:
42 * Version 1.5
43 * Fixed the changelog notice bellow about the API. I had a mistake.
44
45 2016-11-18 Dimitrij Mijoski <dmjpp at hotm>:
46 * Version 1.5
47 * Lot of stability fixes
48 * Fixed compilation errors on various systems (Windows, FreeBSD)
49 * Small performance improvement compared to 1.4.0
50 * The C++ API is updated to use modern C++ types (string, vector).
51 Backward compatibility is kept for most of the functions except for
52 the following:
53 get_wordchars();
54 get_version();
55 input_conv(string, string);
56 removed get_csconv();
57
58 2016-04-29 Caolán McNamara <caolanm at LibO>:
59 * deprecate old api and add new one
60 old one remains implemented in terms of new one
61 and will eventually be removed
62 * shrink exposed api down to just hunspell.hxx
63 * next major release is likely to require C++11
64
65 2016-04-15 Caolán McNamara <caolanm at LibO>:
66 * generally using std::string and std::vector internally
67
68 2016-04-13 Caolán McNamara <caolanm at LibO>:
69 * gh#371 drop experimental code
70
71 2015-09-11 Caolán McNamara <caolanm at LibO>:
72 * rhbz#1261421 crash on mashing hangul korean keyboard
73
74 2014-12-03 Németh László <nemeth at numbertext dot org>:
75 * tools/hunspell.cxx: security fixes of the Hunspell executable
76 - secure file name handling, the problem (checking
77 OpenDocument files with malicious file names)
78 reported by Eric Sesterhenn
79 - using tmpnam() only with system("mkdir tempname && ...")
80
81 2014-10-17 Caolán McNamara <caolanm at LibO>:
82 * sf#245 Feature from Anish Patil -S mode
83 to show suggestions for completion of
84 correctly spelled words
85 * sf#248 Fix manpage about how to include
86
87 2014-10-16 Caolán McNamara <caolanm at LibO>:
88 * rhbz#915448, sf#57, sf#185 report character offset
89 and not byte offset in ispell mode
90 * sf#56 segv in experimental mode
91 * sf#228 don't translate init string
92
93 2014-09-22 Németh László <nemeth at numbertext dot org>:
94 * fix crash in morphological analysis of the Hungarian
95 compound word 'művészegyéniség', reported by Gáspár Sinai
96
97 2014-08-26 Németh László <nemeth at numbertext dot org>:
98 * unmunch separates flags of prefixes from the word,
99 bug reported by Daniel Naber
100
101 2014-08-05 Németh László <nemeth at numbertext dot org>:
102 * moz#318040 Mozzilla accepts abbreviations without dots
103 * myfopen(): add _wfullpath to expand relative parts of absolute paths
104
105 2014-07-16 Caolán McNamara <caolanm at LibO>:
106 * moz#675553 Switch from PRBool to bool
107 * moz#690892 replace PR_TRUE/PR_FALSE with true/false
108 * Silence the warning about empty while body loop in clang
109 * moz#777292 Make nsresult an enum
110 * moz#579517 Use stdint types in gecko
111 * moz#784776 consistently use FLAG_NULL
112 * moz#927728 Convert PRUnichar to char16_t
113 * moz#943268 Remove nsCharsetAlias and nsCharsetConverterManager
114 * Don't include config.h in license.hunspell if MOZILLA_CLIENT is set
115
116 2014-06-26 Caolán McNamara <caolanm at LibO>:
117 * clang scan-build: Allocator sizeof operand mismatch
118 * clang scan-build: other low hanging warnings
119 * clang scan-build: significant warnings
120
121 2014-06-02 Németh László <nemeth at numbertext dot org>:
122 * escape spaces in paths of ODF files
123
124 2014-05-28 Németh László <nemeth at numbertext dot org>:
125 * add long path/Unicode path support in WIN32 environment:
126 - hunspell#233 (reported by mahak gark) and LibreOffice fdo#48017
127 * flat ODF support, eg.:
128 hunspell doc.fodt
129 cat doc.fodt | hunspell -l -O
130 * new options:
131 - -X (XML) input format
132 - -O (ODF or flat ODF) input format
133 - --check-apostrophe: check and force Unicode apostrophe usage
134 (ASCII or Unicode apostrophe has to be in the
135 WORDCHARS section of the affix file)
136 * fix ODF support:
137 - break 1-line XML of ODT documents at </style:style>, too,
138 not only at </text:p> (limiting tokenization problems, when
139 fgets stops within an XML tag)
140 - show ODF file path on the UI instead of the temporary file
141 * fix XML support:
142 - ', ", &, < and > in replacements converted to XML entities
143 - recognize &apos at tokenization, depending from WORDCHARS
144 - &apos; in tokens converted to ' before spell checking and
145 in the output of the pipe interface
146 * better apostrophe usage:
147 - WORDCHARS only with one of the Unicode or ASCII apostrophe
148 results extended word tokenization: both of them will be part of
149 the words (if they are inside: eg. word's, but not words').
150 - convert Unicode apostrophes to ASCII ones for 8-bit dictionaries
151 (eg. English dictionaries), or for UTF-8 dictionaries only
152 with ASCII apostrophe supports (eg. French dictionaries).
153 * updated manual:
154 - hunspell.4 renamed to hunspell.5, see
155 hunspell#241 reported by Cristopher Yeleighton
156 - updated translations
157 - note about long/Unicode paths in WIN32 (hunspell.3)
158
159 2014-04-25 Németh László <nemeth at numbertext dot org>:
160 * OpenDocument support, eg.
161 hunspell *.odt
162 hunspell -l *.odt
163 * always load default personal dictionary (fix
164 filtering bad words - reduce this word list - using
165 it as a personal dictionary workflow)
166 * fix parsing/URL recognition problem (bad tokens
167 with aposthrophes)
168
169 2013-07-25 [email protected]
170 * moz#897255 Wasted work in line_uniq
171 * moz#897780 Wasted work in SuggestMgr::twowords
172
173 2013-07-25 Caolán McNamara <caolanm at LibO>:
174 * hunspell#167 layout problems with long lines
175 - based on the original fix by xorho
176 adapted to HEAD
177 * rhbz#925562 upgrade config.guess for aarch64
178
179 2013-07-24 [email protected]
180 * moz#896301 Wasted work in SfxEntry::checkword
181 * moz#896844 Wasted work in AffixMgr::defcpd_check
182
183 2013-06-13 Konstantin Khlebniko
184 * #49 HashMgr::add_word computes wrong size for struct hentry
185
186 2013-06-13 Ville Skyttä
187 * #53 Man page syntax fixes
188
189 2013-04-19 John Thomson <john thomson at SIL>
190 * win_api: add remove() of Hunspell API (hun#3606435)
191
192 2013-04-19 Rouslan Solomokhin <at sf.net>
193 * fix crash in suggestions for 99-character long words
194 by extending arrays of SuggestMgr::forgotchar_*
195 (hun#3595024, also http://crbug.com/130128),
196 thanks to also Paweł Hajdan to report the patch
197
198 2013-04-01 Caolán McNamara <caolanm at LibO>:
199 * hunspell: -Werror=undef
200
201 2013-03-13 Caolán McNamara <caolanm at LibO>:
202 * rhbz#918938 crash in interaction with danish thesaurus
203
204 2012-09-18 Németh László <nemeth at numbertext dot org>:
205 * src/hunspell/affixmgr.*: - fix morphological analysis of
206 compound words (hun#3544994, reported by Dávid Nemeskey, fdo#55045)
207
208 2012-06-29 Caolán McNamara <caolanm at LibO>:
209 * fix various coverity warnings
210
211 2012-01-10 Ehsan Akhgari <ehsan at mozilla dot com>
212 * moz#710940 Firefox Crash [@ AffixMgr::parse_file(char const*, char
213 const*) ]
214
215 2011-12-16 Jared Wein <jwein at mozilla dot com>
216 * moz#710967 Incorrect argument passed to strncmp in
217 AffixMgr::parse_convtable
218
219 2011-12-06 Caolán McNamara <caolanm at LibO>:
220 * rhbz#759647 fixed tempname of hunSPELL.bak collides with other users
221 when multiple edits in one dir
222
223 2011-10-13 Caolán McNamara <caolanm at LibO>:
224 * moz#694002 crash in hunspell affixmgr on exit with bad .aff
225 * leak in hunspell affixmgr with bad .aff
226
227 2011-09-19 Caolán McNamara <caolanm at LibO>:
228 * make libparsers.a not installed thanks to Tomáš Chvátal
229
230 2011-06-23 Caolán McNamara <caolanm at LibO>:
231 * fix some windows compiler warnings
232
233 2011-05-24 Németh László <nemeth at numbertext dot org>:
234 * src/hunspell/affixmgr.*: allow twofold suffixes in compounds
235 by extended version of Arno Teigseth's patch, see hun#3288562.
236 - new option for this feature: COMPOUNDMORESUFFIXES
237
238 2011-02-16 Németh László <nemeth at numbertext dot org>:
239 * src/*/Makefile.am: fix library versioning, the probem reported by
240 Rene Engerhald and Simon Brouwer.
241
242 * man/hunspell.4: new version based on the revised version of Ruud Baars
243
244 2011-02-02 Németh László <nemeth at OOo>:
245 * suggestngr.cxx: fix ngram PHONE suggestion for input words with
246 diacritics using UTF-8 encoded dictionaries (add byte length to the
247 8-bit phonet() argument instead of character length)
248
249 * suggestmgr.cxx: fix missing csconv problem with UTF-8 encoding
250 dictionares, when the input contains non-BMP characters
251 - tests/utf8_nonbmp.sug: test file
252
253 * suggestmgr.cxx: mixed and keyboard based character suggestions
254 don't forbid ngram suggestion search (optimized tests/suggestiontest)
255
256 * affixmgr.cxx: fix hun#2999225: interfering compounding mechanisms,
257 tested on Dutch word list and reported by Ruud Baars
258
259 * affixmgr.cxx: allomorph fix for hun#2970240 (Hungarian
260 compound "vadász+gép" was analyzed as vad+ász+gép, and rejected
261 by the ss->s rep rule (verb "vadássz"), but the analysis
262 didn't continue for the longer word parts (vadász+gép).
263
264 * csutil.cxx: add lang code "az_AZ", "hu_HU", "tr_TR" for back
265 compatibility (fixing Azeri and Turkish casing conversion, also
266 Hungarian compound handling)
267
268 * affixmgr.cxx: fix morphological analysis
269
270 2011-01-26 Németh László <nemeth at OOo>:
271 * affixmgr.cxx: fix for moz#626195 (memcheck problem with FULLSTRIP).
272
273 * affixmgr.*, suggestmgr.cxx: FORBIDWARN parameter (see manual)
274
275 2011-01-24 Németh László <nemeth at OOo>:
276 * suffixmgr.cxx: fix bad suggestion of forbidden compound words, eg.
277 "termijndoel" with the Dutch dictionary. Reported by Ruud Baars.
278
279 * latexparser.cxx: fix double apostrophe TeX quoation mark tokenization
280 (hun#3119776), reported by Wybodekker at SF.net.
281
282 * tests/suggestiontest/*: multilanguage and single Hunspell version, see README
283 * tests/suggestiontest/prepare2: for make -f Makefile.orig single
284
285 2011-01-22 Németh László <nemeth at OOo>:
286 * affixmgr.*, suggestmgr.*: new features
287 ONLYMAXDIFF: remove all bad ngram suggestions (default mode keeps one)
288 NONGRAMSUGGEST: similar to NOSUGGEST, but it forbids to use the word
289 in ngram based (more, than 1-character distance) suggestions.
290
291 2011-01-21 Németh László <nemeth at OOo>:
292 * suggestmgr.*: limit wild suggestions (hun#2970237 by Ruud Baars)
293 - limited compound word suggestions
294 - improved and limited ngram based suggestions
295 * tests/*.sug: modified test files
296 - feature MAXCPDSUGS:
297 MAXCPDSUGS 0 : no compound suggestion, suggested by
298 Finn Gruwier Larsen in hunfeat#2836033
299 MAXCPDSUGS n : max. ~n compound suggestions
300 - feature MAXDIFF: differency limit for ngram suggestions: 0-10
301 eg. MAXDIFF 5: normal (default) limit
302 MAXDIFF 0: only one ngram suggestion
303 MAXDIFF 10: ~maxngramsugs ngram suggestions
304
305 * affixmgr.*, hunspell.*: add flag FORCEUCASE (hun#2999228), force
306 capitalization of compound words, see Hunspell 4 manual),
307 suggested by Ruud Baars
308 test/forceucase.*: test files
309
310 * affixmgr.*, hunspell.*: add flag WARN (hun#1808861), optional warning feature
311 for rare words, suggested by Ruud Baars
312 tests/warn: test files
313 * tools/hunspell.cxx: add option -r for optional filtering of rare words
314
315 * affixmgr.cxx: fix hun#3161359 (gcc warnings) reported by Ryan VanderMeulen.
316
317 2011-01-17 Németh László <nemeth at OOo>:
318 * suggestmgr.cxx: fix hun#3158994 and hun#3159027 (missing csconv table
319 using awkward 8bit capitalization of UTF-8 encoded dictionary words with PHONE
320 suggestion, reported by benjarobin and dicollecte at SF.net).
321
322 2011-01-13 Németh László <nemeth at OOo>:
323 * affixmgr.cxx: ONLYINCOMPOUND fix for hun#2999224 (fogemorphene
324 was allowed in end position of compoundings). Reported by Ruud Baars.
325 * tests/onlyincompound2.*: test files
326
327 2011-01-10 Ingo H. de Boer <idb_winshell at SF.net>:
328 * win_api/{hunspell,libhunspell, testparser}.vcproj: updated project
329 files for the library and the executables. Compiling problem
330 also reported by Don Walker.
331
332 2011-01-06 Németh László <nemeth at OOo>:
333 * affixmgr.cxx: fix freedesktop#32850 (program halt during Hungarian
334 spell checking of the word "6csillagocska6", reported by András Tímár)
335
336 * tools/hunspell.cxx: add Mac OS X Hunspell dictionary paths, asked by
337 Vidar Gundersen in hunfeat#3142010
338
339 2011-01-05 Caolán McNamara <cmc at OOo>:
340 * moz#620626 NS_UNICHARUTIL_CID doesn't support
341 case conversion
342
343 2011-01-03 Németh László <nemeth at OOo>:
344 * NEWS and THANKS: update for release 1.2.13
345
346 2010-12-20 Németh László <nemeth at OOo>:
347 * affixmgr.cxx: hun#3140784
348
349 2010-12-16 Németh László <nemeth at OOo>:
350 * affixmgr.cxx:
351 - improved fix of hun#2970242 (supporting
352 zero affixes, reported by Ruud Baars
353 - tests/opentaal_cpdpat{,2}: test files
354
355 - switching off default BREAK parameters by BREAK 0,
356 reported by Ruud Baars
357
358 - hun#2999225: interfering compounding mechanisms, reported by Ruud Baars
359
360 2010-12-11 Németh László <nemeth at OOo>:
361 * affixmgr.cxx: fix hun#2970242 (CHECKCOMPOUNDPATTERN only with flags),
362 the bug reported by Ruud Baars
363 * tests/2970242.*: test files
364
365 * tests/2970240.*: test files for CHECKCOMPOUNDPATTERN fix (check all
366 boundaries in compound words, fixed by the previous CHECKCOMPOUNDREP
367 fix), the bug reported by Ruud Baars
368
369 * win_api/Makefile.cygwin: update
370
371 2010-12-09 Caolán McNamara <cmc at OOo>:
372 * moz#617953 fix leak
373
374 2010-11-08 Caolán McNamara <cmc at OOo>:
375 * rhbz#650503 crash in arabic dictionary
376
377 2010-11-05 Caolán McNamara <cmc at OOo>:
378 * rhbz#648740 don't warn on empty flagvector
379
380 2010-11-03 Caolán McNamara <cmc at OOo>:
381 * logically we shouldn't need a csconv table in utf-8 mode
382
383 2010-10-27 Németh László <nemeth at OOo>:
384 * hun#3000055 (requested by Ruud Baars) add REP boundary specifiation:
385 REP ^word$ xxxx
386 REP ^wordstarting xxxx
387 REP wordending$ xxxx
388
389 * hun#3008434 (requested by Adrián Chaves Fernández) and
390 hun#3018929 (requested by Ruud Baars): REP with more than 2 words:
391 REP morethantwo more_than_two
392
393 * suggestmgr.cxx: fix incomplete suggestion list for capitalized words,
394 eg. missing Machtstrijd->Machtsstrijd in the Dutch dictionary
395 (reported by Ruud Bars)
396
397 * tests, man: related updates
398
399 2010-10-12 Caolán McNamara <cmc at OOo>:
400 * moz#603311 HashMgr::load_tables leaks dict when decode_flags fails
401 * fix mem leak found with new tests
402 * hun#3084340 allow underscores in html entity names
403
404 2010-10-07 Németh László <nemeth at OOo>:
405 * affixmgr.cxx:
406 - hun#2970239 fix bad suggestion of forbidden compound words
407 - hun#2999224 fix keepcase feature on compound words (only partial
408 fix for COMPOUNDRULE based compounding)
409 - fix checkcompoundrep feature in compound words (check all boundaries,
410 not only the last one)
411 Problems reported by Ruud Baars.
412
413 * tests/opentaal_forbiddenword[12]*, tests/opentaal_keepcase*:
414 new test files for the previous fixes
415 * tests/checkcompoundrep: extended test file.
416
417 2010-09-05 Caolán McNamara <cmc at OOo>:
418 * moz#583582 fix double buffer gcc fortify issue
419
420 2010-08-13 Caolán McNamara <cmc at OOo>:
421 * moz#586671 AffixMgr::parse_convtable leaks pattern/pattern2 if it
422 can't create both
423 * moz#586686 tidy up get_xml_list and friends
424
425 2010-08-10 Caolán McNamara <cmc at OOo>:
426 * hun#3022860 fix remove duplicate code
427
428 2010-07-17 Caolán McNamara <cmc at OOo>:
429 * remove ununsed get_default_enc and avoid potential misrecognition of
430 three letter language ids
431 * normalize encoding names before lookup
432
433 2010-07-05 Caolán McNamara <cmc at OOo>:
434 * hun#2286060 add Hangul syllables to unicode tables
435
436 2010-06-26 Caolán McNamara <cmc at OOo>:
437 * moz#571728 keep new[]/delete[] wrappers in sync for embedded in moz
438 case
439
440 2010-06-13 Caolán McNamara <cmc at OOo>:
441 * moz#571728 keep new[]/delete[] wrappers in sync for embedded in moz
442 case
443
444 2010-06-02 Caolán McNamara <cmc at OOo>:
445 * moz#569611 compile cleanly under win64
446
447 2010-05-22 Caolán McNamara <cmc at OOo>:
448 * moz#525581 apply mozilla's current preferred get_current_cs impl
449
450 2010-05-17 Németh László <nemeth at OOo>:
451 * affixmgr.cxx: fix bad limitation of parenthesized flags at
452 COMPOUNDRULEs. Windows crash reported by Ruud Baars and Simon Brouwer.
453
454 2010-05-05 Caolán McNamara <cmc at OOo>:
455 * rhbz#589326 malloc of int that should have been of char**
456 * hun#2997388 fix ironic misspellings
457
458 2010-04-28 Caolán McNamara <cmc at OOo>:
459 * moz#550942 get_xml_list doesn't handle failure from get_xml_par
460
461 2010-04-27 Caolán McNamara <cmc at OOo>:
462 * moz#465612 mozilla-specific code leaks
463 * moz#430900 phone is dereferenced before oom check
464 * moz#418348 ckey_utf alloc is used unchecked in SuggestMgr::badcharkey_utf
465 * CID#1487 pointer "rl" dereferenced before NULL check
466 * CID#1464 Returned without freeing storage "ptr"
467 * CID#1459 Avoid duplicate strchr
468 * CID#1443 Avoid any chance of dereferencing *slst
469 * CID#1442 Unsafe to have a null morph
470 * CID#1440 Avoid null filenames
471 * CID#1302 Dereferencing NULL value "apostrophe"
472 * CID#1441 Avoid deferencing null ppfx
473
474 2010-04-16 Caolán McNamara <cmc at OOo>:
475 * hun#2344123 fix U)ncap in utf-8 locale
476 * fix up hunspell text UI and lines wider than terminal
477
478 2010-04-15 Caolán McNamara <cmc at OOo>:
479 * hun#2613701 fix small leak in FileMgr::FileMgr
480 * fix small leak in tools/hunspell
481 * hun#2871300 avoid crash if def and words are NULL
482 * hun#2904479 fix length of hzip file
483 * hun#2986756 mingw build fix
484 * hun#2986756 fix double-free
485 * hun#2059896 fix crash in interactive mode without nls
486 * hun#2917914 add some extra words to the latexparser
487 * make some structs static
488 * C-api has duped symbol names
489 * regenerate gettext/intl with recent version
490 * hun#2796772 build a .dll under MinGW
491 * rhbz#502387 allow cross-compiling for MinGW target
492 * hun#2467643 update .vcproj files to include replist.?xx
493 * unify visiblity/dll_export support across platforms
494 * hun#2831289 sizeof(short) typo
495 * hun#2986756 add -u3 gcc style output
496
497 2010-04-14 Caolán McNamara <cmc at OOo>:
498 * hun#2813804 fix segfault on hu_HU stemming
499
500 2010-04-13 Caolán McNamara <cmc at OOo>:
501 * hun#2806689 fix ironic misspellings
502 * hun#2836240 add Italian translations
503
504 2010-04-09 Caolán McNamara <cmc at OOo>:
505 * fix titchy possible leak in command-line spellchecker
506
507 2010-04-07 Caolán McNamara <cmc at OOo>:
508 * hun#2973827 apply win64 patch
509 * hun#2005643 fix broken mystrdup
510
511 2010-03-04 Caolán McNamara <cmc at OOo>:
512 * ooo#107768 fix crash in long strings in spellml mode
513 * hun#1999737 add some malloc checks
514 * hun#1999769 drop old buffer on realloc failure
515 * hun#2005643 tidy string functions
516 * hun#2005643 micro-opt
517 * hun#2006077 free strings on failed dict parse
518 * hun#2110783 ispell-alike verbose mode implementation
519
520 2010-03-03 Németh László <nemeth at OOo>:
521 * hunspell/(affixmgr, suggestmgr).cxx: add character sequence
522 support for MAP suggestion, using parenthesized character groups
523 in the syntax, eg. MAP ß(ss).
524 * man/hunspell.4, tests/map*: documentation and test files
525
526 2010-02-25 Németh László <nemeth at OOo>:
527 * hunspell/hunspell.cxx: add recursion limit for BREAK (fix OOo Issue 106267)
528
529 * hunspell/hunspell.cxx: fix crash in morphological analysis of
530 capitalized words with ending dashes
531
532 * affixmgr.cxx: fix morphological analysis of long numbers combined with dash,
533 eg. 45-00000045 (reported by [email protected]).
534
535 2010-02-23 Caolán McNamara <cmc at OOo>:
536 * hun#2314461 improve ispell-alike mode
537 * hun#2784983 improve default language detection
538 * hun#2812045 fix some compiler warnings
539 * hun#2910695 survive missing HOME dir
540 * hun#2934195 fix suggestmgr crash
541 * hun#2921129 remove unused variables
542 * hun#2826164 make sure make check uses the in-tree libhunspell
543 * bump toolchain to support --disable-rpath
544 * hun#2843984 fix coverity warning
545 * hun#2843986 fix coverity warning
546 * hun#2077630 add iconv lib
547 * make gcc strict-aliasing warning free
548 * make cppcheck warning free
549
550 2008-11-01 Németh László <nemeth at OOo>:
551 * replist.*, hunspell.cxx, affixmgr.cxx: new input and output
552 conversion support, see ICONV and OCONV keywords in the Hunspell(4)
553 manual page and the test examples. The input/output conversion
554 problem of syllabic languages reported by Daniel Yacob and
555 Shewangizaw Gulilat.
556 - tests/{iconv,oconv}.*: test examples
557
558 * tools/wordforms: word generation script for dictionary developers
559 (Hunspell version of the unmunch program)
560
561 * hunspell/hunspell.cxx: extended BREAK feature: ^ and $ mean in break
562 patterns the beginning and end of the word.
563 - tests/BREAK.*: modified examples.
564
565 * hunspell/hunspell.cxx: set default break at hyphen characters.
566 The associated problem reported by S Page in Hunspell Bug 2174061.
567 See Mozilla Bug ID 355178 and OOo Issue 64400, too.
568 - tests/breakdefault.*: test data
569 The following definition is equivalent of the default word break:
570
571 BREAK 3
572 BREAK -
573 BREAK ^-
574 BREAK -$
575
576 * affixmgr.cxx: SIMPLIFIEDTRIPLE is a new affix file keyword to allow
577 simplified forms of the compound words with triple repeating letters.
578 It is useful for Swedish and Norwegian languages.
579
580 * affixmgr.cxx: extend CHECKCOMPOUNDPATTERN to support
581 alternations of compound words for example by sandhi
582 feature of Indian and other languages. The problem reported
583 by Kiran Chittella associated with Telugu writing system
584 (see Telugu example in tests/checkcompoundpattern4.test).
585 The new optional field of CHECKCOMPOUNDPATTERN definition is the
586 replacement of the compound boundary defined by the previous fields:
587 CHECKCOMPOUNDPATTERN ff f ff
588 means ff|f compound boundary has been replaced by "ff", like in
589 the (prereform) German Schiffahrt (Schiff+fahrt).
590 - CHECKCOMPOUNDPATTERN supports also optional flag conditions now:
591 CHECKCOMPOUNDPATTERN ff/A f/B ff
592 means that the first word of the compound needs flag "A" and
593 the second word of the compound needs flag "B" to the operation.
594
595 * tools/hunspell.cxx: add empty lines as separators to the output of
596 the stemming and morphological analysis.
597
598 * affixmgr.cxx: fix condition checking algorithm. Bad suggestion
599 generation reported by Mehmet Akin in SF.net Bug 2124186 with help of
600 Eleonora Goldman.
601
602 * affixmgr,cxx: fix COMPOUNDWORDMAX feature. The problem and its
603 code details reported by Göran Andersson under SF.net Bug ID 2138001.
604
605 * csutil.cxx: fix bad conditional code for Mozilla compilation.
606 Patch by Serge Gautherie. The problem reported by Ryan VanderMeulen.
607
608 * hunspell/hunspell.cxx: add missing ngram suggestion for HUHINITCAP
609 (capitalized mixed case) words.
610
611 * w_char.hxx: use GCC conditions for GCC related code. Patch by
612 Ryan VanderMeulen.
613
614 * affixmgr.cxx: check morphological description in morphgen()
615 (fix potential program fault by incomplete morphological
616 description of affix rules)
617
618 * src/win_api: config.h: switch on warning messages on Windows
619
620 * tools/affixcompress: extended help for -h (use LC_ALL=C sort
621 for input word list)
622
623 * man/hunspell.4: updated manual:
624 - new and modified features (SIMPLIFIEDTRIPLE, ICONV, OCONV,
625 BREAK, CHECKCOMPOUNDPATTERN).
626 - note about costs of zero affixes, suggested by Olivier Ronez.
627
628 * hunspell/hunspell.cxx: remove deprecated word breaking codes.
629
630 2008-08-15 Németh László <nemeth at OOo>:
631 * affentry.cxx: add FULLSTRIP option. With FULLSTRIP, affix rules can
632 strip full words, not only one less characters. Suggested by
633 Davide Prina and other developers in OOo Issue 80145.
634 * tests/fullstrip.*: Test data based on Davide Prina's example.
635 * tools/unmunch.cxx: modified for FULLSTRIP.
636
637 * affixmgr.cxx: COMPOUNDRULE now works with long and numerical flag
638 types by parenthesized flags. Syntax: (flag)*, (flag)(flag)?(flag)*.
639 * tests/compoundrule[78].*: tests with parenthesized COMPOUNDRULE
640 definitions.
641
642 * suggestmgr.cxx: modified badchar*(), forgotchar*() and extrachar*()
643 1-character distance suggestion algorithms: search a TRY character
644 in all position instead of all TRY characters in a character position
645 (it can give more readable suggestion order, also better suggestions
646 in the first positions, when TRY characters are sorted by frequency.)
647 For example, suggestions for "moze":
648 ooze, doze, Roze, maze, more etc. (Hunspell 1.2.6),
649 maze, more, mote, ooze, mole etc. (Hunspell 1.2.7).
650
651 * suggestmgr.cxx: extended compound word checking for better COMPOUNDRULE
652 related suggestions, for example English ordinal numbers: 121323th ->
653 121323rd (it needs also a th->rd REP definition).
654
655 * phonet.cxx: cast unsigned char parameter of isdigit() and fix
656 isalpha by myisalpha() (potential problems in Windows environment).
657 Reported by Thomas Lange in OOo Issue 92736.
658
659 * hunspell/csutil.*,hunspell/{affentry,affixmgr,hunspell,suggestmgr}.cxx:
660 fix potential buffer overloading under morphological analysis by the
661 new mystrcat() function. Reported by Molnár Andor (dolhpy at true
662 dot hu) in SF.net Bug 2026203.
663
664 * affixmgr.cxx: add recursion limit to defcpd(). Fix OOo Issue 76067:
665 crash-like deceleration by checking hexadecimal numbers with long FFF
666 sequence (combinatory explosion by the en_US words "f" and "ff").
667 Missing fix reported by Mathias Bauer.
668
669 * affixmgr.cxx: fix the difference in the Unicode and non-Unicode
670 parts of cpdcase_check(). Bug report by Brett Wilson.
671
672 * filemgr.*, affixmgr.cxx, csutil.*, hashmgr.*: warning messages now
673 contain line numbers (use --with-warnings configure option for
674 warning messages).
675
676 * hunspell.cxx: analyze(): fix case conversion of stemming and
677 morphological analysis of UTF-8 encoded input. Reported by Ferenc Godó.
678
679 * tools/hunspell.cxx: fix LaTeX Unicode support in filter mode.
680 Reported by Jan Seeger in SF.net Bug 2039990.
681
682 * affixmgr.hxx: 0.5 or in 64 bit environment, 1 MB (virtual) memory
683 saving using only the requested size for sFlag and pFlag arrays.
684 Bug report by Brett Wilson.
685
686 * affixmgr.cxx,tools/hunspell.cxx: get_version() returns with full
687 VERSION affix parameter instead of its first word. Fixes for
688 Hunspell's header. Some problems with Hunspell header reported in
689 SF.net Bug 2043080.
690
691 2008-07-15 Németh László <nemeth at OOo>:
692 * affentry.cxx: fixes of the affix rule matching algorithm (affected
693 only the sk_SK dictionary from all OpenOffice.org dictionaries):
694 - fix dot pattern + accented letters matching (in non Unicode encoding)
695 - word-length conditions work again
696 * tests/condition.*: extended test for the fix.
697
698 * hashmgr.cxx: load multiword expressions: spaces may be parts
699 of the dictionary words again (but spaces also work as morphological
700 field separators: word word2 -> "word word2", word po:noun -> "word").
701 * man/hunspell.4: updated manual
702
703 * tools/hunspell.cxx: add iconv character conversion support to
704 stemming and morphological analysis
705
706 * tools/hunspell.cxx: add /usr/share/myspell/dicts search path for
707 Ubuntu support
708
709 2008-07-09 Németh László <nemeth at OOo>:
710 * affentry.cxx: fixes of the affix rule matching algorithm:
711 - right ASCII character handling in bracket expression;
712 - fault-tolerant nextchar() for bad rules.
713 Problem with the en_GB dictionary and nextchar() with a detailed
714 code analysis reported by John Winters in SF.net Bug ID 2012753.
715 * tests/condition.*: extended test for the fix.
716
717 * hunspell/hunspell.*, parsers/*, tools/hunspell.cxx: fix compiler
718 warnings (deprecated const-free char consts)
719
720 * win_api/hunspelldll.*: add hunspell_free_list(), the problem
721 reported by Laurier Mercer.
722
723 2008-06-30 Török László <torok_laszlo at users dot SF dot net>:
724 * tests/affixmgr.cxx: fix morphological analysis: strcat() on
725 an uninitialized char array in suffix_check_morph().
726
727 2008-06-18 Németh László <nemeth at OOo>:
728 * src/hunspell/affixmgr.cxx: fix GCC compiler warnings
729 (comparisons with string literal results in unspecified behaviour).
730 The problem reported by Ladislav Michnovič.
731
732 2008-06-17 Németh László <nemeth at OOo>:
733 * src/hunspell/{hunspell.cxx,hunspell.h}: add free_list() to the C and
734 C++ interface to deallocate suggestion lists. The problem
735 reported by Laurie Mercer and Christophe Paris.
736 * csutil.cxx: fix freelist() to deallocate non-NULL list, when n = 0.
737 * tools/{analyze,example,chmorph,hunspell}.cxx: use free_list().
738
739 * tools/hunspell.cxx: fix only --with-readline compiling problem.
740 Reported by Volkov Peter in SF.net Bug 1995842.
741
742 * man/hunspell.3,hunspell.hxx: fix analyze and generate examples in
743 the manual and comments (using char*** parameter instead of char**).
744
745 * tools/example.cxx: fix suggestion example.
746
747 2008-06-17 Németh László <nemeth at OOo>:
748 * affentry.cxx: fix the new affix rule matching algorithm of
749 Hunspell 1.2. Arabic dictionary problem reported by Khaled Hosny
750 in SF.net Bug ID 1975530. Mohamed Kebdani also sent a
751 prepared test data.
752 * tests/{1975530,condition*}: tests for the fix
753
754 2008-06-13 Ingo H. de Boer <idb_winshell at SF.net>:
755 * src/hunspell/{affixmgr.cxx,hunspell.cxx}: add missing type
756 cast to strstr() calls for VC8 compatibility.
757
758 2008-06-13 Németh László <nemeth at OOo>:
759 * suggestmgr.cxx: add also part1-part2 suggestion with dash
760 for bad part1part2 word forms, suggested by Ruud Baars.
761 For example, now suggestion of "parttime": "part time"
762 and "part-time".
763 NOTE: this feature will work only when the TRY definition
764 contains "-" or the letter "a".
765
766 * hunspell.cxx: new XML API in spell() and suggest() (see hunspell(3)).
767
768 * src/hunspell/*: fixes for OpenOffice.org build environment.
769
770 * man/{hunspell.3,hzip.1,hunzip.1}: add new manual pages for
771 Hunspell programming API and dictionary compression and
772 encryption utilities.
773
774 * src/hunspell/*: handle failed mystrdup() calls and other potential
775 insufficient memory problems. The problem reported by Elio Voci
776 in OpenOffice.org Issue 90604 and others.
777
778 * src/tools/affixmgr.cxx: restore original behaviour of get_wordchars
779 without conditional code. Problem reported by Ingo H. de Boer
780 in SF.net Bug 1763105.
781
782 * win_api/hunspelldll.h: put_word() renamed to add() in the (old)
783 Windows DLL API bug reported in SF.net Bug 1943236. Also reported
784 by Bartkó Zoltán.
785
786 * tools/hunspell.cxx: fix chench() for environments without
787 native language support (ENABLE_NLS 0 in config.h),
788 PHP system_exec() bug reported by Michel Weimerskirch in
789 SF.net Bug 1951087.
790
791 * hunspell.cxx, affixmgr.cxx: remove "result" from the
792 (result && *result) conditions, when "result" is a static variable.
793 The problem and a possible solution reported by Ladislav Michnovič.
794
795 * affixmgr.cxx: parse_affix(): print line instead of NULL in
796 the warning message, when affix class header is bad.
797 The problem reported by Ladislav Michnovič.
798
799 2008-06-01 Christian Lohmaier <cloph at OOo>
800 * configure.ac: patch to fix --with-readline, --with-ui logic.
801 Reported in the SF.net Bug 981395.
802
803 2008-05-04: Volkov Peter <volkov_peter at users sourceforge net>
804 * configure.ac: fix LibTool 2.22 incompatibility by removing
805 unused LT_* macros. Report and patch in SF.net Bug 1957383.
806 The problem reported and fixed by Ladislav Michnovič, too.
807
808 2008-04-23: Ladislav Michnovič <lmichnovic at suse cz>
809 * hunspell.pc.in: fix wrongly set directories.
810
811 2008-04-12 Németh László <nemeth at OOo>:
812 * src/tools/hunspell.cxx:
813 - Multilingual spell checking and special dictionary support with -d.
814 Multilingual spell checking suggested by Khaled Hosny (SF.net
815 Bug 1834280). Example for the new syntax:
816
817 -d en_US,en_geo,en_med,de_DE,de_med
818
819 en_US and de_DE are base dictionaries, and en_geo, en_med, de_med
820 are special dictionaries (dictionaries without affix file).
821 Special dictionaries are optional extension of the base dictionaries.
822 There is no explicit naming convention for special dictionaries,
823 only the ".dic" extension: dictionaries without affix file will
824 be an extension of the preceding base dictionary. First dictionary
825 in -d parameter must have an affix file (it must be a base
826 dictionary).
827
828 - new options for debugging, morphological analysis and stemming:
829 -m: morphological analysis or flag debug mode (without affix
830 rule data it signs the flag of the affix rules)
831 -s: stemming mode
832 -D: show also available dictionaries and search path
833 (suggested by Aaron Digulla in SF.net Bug 1902133)
834
835 - add missing refresh() to print bad words before the slower suggestion
836 search in UI (better user experience)
837
838 - fix tabulator problems (reported by ugli-kid-joe AT sf DOT net)
839
840 - fix different encoding of dic and input, and suggestions
841
842 - add per mille sign to LANG hu_HU section.
843
844 - rewrite program messages. Concatenating multiple printfs for
845 easier translation suggested by András Tímár and Gábor Kelemen.
846
847 * src/hunspell/csutil.cxx: set static encds variable. Patch by
848 Rene Engerhald. SF.net Bug 1896207 and 1939988.
849
850 * src/hunspell/w_char.hxx,csutil.hxx: reorganizing
851 w_char typedef and HENTRY_DATA, HENTRY_FIND consts
852
853 * src/hunspell/hunzip.cxx: fopen(): using rb options instead of r (fix
854 for Windows)
855
856 * src/tools/affixmgr.cxx: restore original behaviour of get_wordchars
857 in an #ifdef WINSHELL section. Problem reported by Ingo H. de Boer
858 in SF.net Bug 1763105.
859
860 * src/tools/chmorph.cxx: remove the experimental modifications
861
862 * src/tools/hzip.c: fopen(): using wb options instead of w (fix
863 for Windows)
864
865 * src/tools/hunzip.cxx: add missing MOZILLA_CLIENT. Reported
866 by Ryan VanderMeulen.
867
868 * man/*, man/hu/*: updated manual
869
870 * man/hunspell.4: fix formatting problem (missing header)
871
872 * tools/makealias: now works with the extra data fields.
873
874 * phonet.cxx: use HASHSIZE const
875
876 * tests/rep.aff: fix REP count
877
878 * src/win_api/Makefile.cygwin, README: native Windows compilation
879 in Cygwin environment without cygwin1.dll dependency (see README
880 for compiling instructions).
881
882 2008-04-08 Roland Smith <rsmith AT xs4all DOT nl>:
883 * src/parsers/latexparser.cxx: fix PATTERN_LEN for AMD64 and
884 other platforms with different struct padding (SF.net Bug 1937995).
885
886 2008-04-03 Kelemen Gábor <kelemeng AT gnome DOT hu>:
887 * po/POTFILES.in: fix path of the source file
888
889 * po/Makevars: add --from-code=UTF-8 gettext option
890
891 * hunspell.cxx: add comments for shortkey translation
892
893 2008-02-04 Flemming Frandsen <flfr AT stibo DOT com>
894 * src/hunspell.h: fix Windows DLL support
895 - this patch also reported by Zoltán Bartkó.
896
897 2008-01-30 Mark McClain <marc_mcclain AT users DOT sf DOT net>
898 * src/hunspell.cxx: stem(): fix function call side effect
899 for PPC platform (SF.net Bug 1882105).
900
901 2008-01-30 Németh László <nemeth at OOo>:
902 * hunspell.cxx, csutil.cxx, hunspelldll.c: fix
903 SF.et Bug 1851246, patch also by Ingo H. de Boer.
904
905 * hunspell.h: fix SF.net Bug 1856572 (C prototype problem),
906 patch by Mark de Does.
907
908 * hunspell.pc.in: fix SF.net Bug 1857450 wrong prefix, reported
909 by Mark de Does.
910
911 * hunspell.pc.in: reset numbering scheme: libhunspell-1.2.
912 Fix SF.net Bug 1857512 reported by Mark de Does,
913 also by Rene Engelhard.
914
915 * csutil.cxx: patches for ARM platform, signed_chars.dpatch
916 by Rene Engelhard and arm_structure_alignment.dpatch by
917 Steinar H. Gunderson <[email protected]>
918
919 * hunzip.*, hzip.c: new hzip compression format
920
921 * tools/affixcompressor: affix compressor utility (similar to
922 munch, but it generates affix table automatically), works
923 with million-words dictionaries of agglutinative languages.
924
925 * README: fix problems reported by Pham Ngoc Khanh.
926
927 * csutil.cxx, suggestmgr: Warning-free in OOo builds.
928
929 * hashmgr.*, csutil.*: fix protected memory problems with
930 stored pointers on several not x86 platforms by
931 store_pointer(), get_stored_pointer().
932
933 * src/tools/hunspell.cxx: fix iconv support on Solaris platform.
934
935 * tests/IJ.good: add missing test file
936
937 * csutil.cxx: fix const char* related errors. Compiling bug
938 with Visual C++ reported by Ryan VanderMeulen and Ingo H. de Boer.
939
940 2008-01-03 Caolan McNamara <cmc at OO.o>:
941 * csutil.cxx: SF.net Bug 1863239, notrailingcomma patch and
942 optimization of get_currect_cs().
943
944 2007-11-01 Németh László <nemeth at OOo>:
945 * hunspell/*: new feature: morphological generation,
946 also fix experimental morphological analysis and stemming.
947 - new API functions and improved API:
948 - analyze(word): (instead of morph()) morphological analysis
949 - stem(word): stemming
950 - stem(list): stemming based on the result of an analysis
951 - generate(word, word2): morphological generation
952 - generate(word, list): morphological generation
953 - add(word): add word to the run-time dictionary (renamed put_word())
954 - add_with_affix(word, word2): (renamed put_word_pattern()):
955 add word to the run-time dictionary with affix flags of the
956 second parameter: all affixed forms of the user words will be
957 recognised by the spell checker. Especially useful for
958 agglutinative languages.
959 - remove(word): remove word from the run-time dictionary (not
960 implemented)
961 - see manual and hunspell/hunspell.hxx header and tests/morph.*
962 * tests/morph.*: test data, example for morphological analysis,
963 stemming and generation
964
965 * tools/analyze, tools/chmorph: extended and new demo applications:
966 - analyze (originally hunmorph): analyses and stems input words,
967 generates word forms from input word pairs.
968 - chmorph: morphological transformation filter
969
970 * configure.ac, hunspell/makefile.am: set library version number.
971 Bug reported by Rene Engelhard.
972
973 * affentry.cxx, affixmgr.cxx: new pattern matching algorithm in
974 condition checking of affix rules instead of the Dömölki-algorithm:
975 - Unlimited condition length (instead of max. 8 characters).
976 - Less memory consumption, especially useful for affix rich languages:
977 5,4 MB memory savings with hu_HU dictionary.
978 - Speed change depends from dictionaries and CPU caches: English spell
979 checking is 4% faster on Linux words with en_US dictionary, Hungarian
980 spell checking is 25% slower on most frequent words of Hungarian
981 Webcorpus.
982
983 * tests/sug.*, sugutf.*: updated test data (use "a" and "lot"
984 dictionary items instead of "a lot".)
985
986 * src/hunspell/hunspell.cxx: free(csconv) instead of delete csconv.
987 Report and patch by Sylvain Paschein in Mozilla Issue 398268.
988
989 * suggestmgr.cxx, tools/hunspell.cxx: bad spelling of "misspelled".
990 Ubuntu Bug #134792, patch by Malcolm Parsons.
991
992 * tests/base_utf.*: use Unicode apostrophe instead of 8-bit one.
993
994 * hunspell.cxx, hashmgr.cxx: add(): use HashMgr::add()
995
996 2007-10-25 Pavel Janík <pjanik at OOo>:
997 * hunspell/csutil.cxx: Fix type cast warnings on 64bit Linux in
998 printing of character positions in u8_u16(). OOo issue 82984.
999
1000 2007-09-05 Németh László <nemeth at OOo>:
1001 * win_api/Hunspell.vproj, parsers/testparser.cxx,textparser.hxx:
1002 warning fixes and removing unnecessary Windows project file.
1003 Reported by Ingo H. de Boer.
1004
1005 * hashmgr.*, {affixmgr,suggestmgr}.cxx: optimized data structure
1006 for variable-count fields (only "ph" transliteration field in
1007 this version, see next item). Also less memory consumption:
1008 -13% (0.75 MB) with en_US dictionary, -6% (1 MB) with hu_HU.
1009
1010 * suggestmgr.cxx: dictionary based phonetic suggestion for special
1011 or foreign pronounciation (see also rule-based PHONE in manual).
1012 Usage: tab separated field in dictionary lines, started with "ph:".
1013 The field contains a phonetic transliteration of the word:
1014
1015 Marseille ph:maarsayl
1016 * tests/phone.*: test data for dictionary and rule based phonetic
1017 suggestion.
1018
1019 * hunspell.cxx: fix potential bad memory access in allcap word
1020 capitalization in suggest() (bug of previous version).
1021
1022 * hunspell.cxx, atypes.hxx: set correct limit for UTF-8 encoded
1023 input words (256 byte).
1024
1025 * suggestmgr.cxx: improved REP suggestions with spaces: it works
1026 without dictionary modification.
1027 OOo issue 80147, reported by Davide Prina.
1028 * tests/rep.*: new test data: higher priority for "alot" -> "a lot",
1029 and Italian suggestion "un'alunno" -> "un alunno".
1030
1031 * affixmgr.cxx: fix Unicode ngram suggestions in expand_rootword().
1032 (Suggestions with bad affixes.)
1033 Bug reported by Vitaly Piryatinksy <piv dot v dot vitaly at gmail>.
1034 * tests/ngram_utf_fix.*: test based on Vitaly Piryatinksy's data.
1035
1036 * suggestmgr.cxx: fix twowords() for last UTF-8 multibyte character.
1037 (conditional jump or move depended on uninitialised value).
1038
1039 2007-08-29 Ingo H. de Boer <idb_winshell at SF.net>:
1040 * win_api/{hunspell,libhunspell, testparser}.vcproj: new project
1041 files for the library and the executables.
1042
1043 * Hunspell.rc, Hunspell.sln, config.h: updated versions.
1044 Version number problem also reported by András Tímár.
1045
1046 2007-08-27 Németh László <nemeth at OOo>:
1047 * suggestmgr.hxx: put fixed version. Bug report by Ingo H. de Boer.
1048
1049 * suggestmgr.cxx: remove variable-length local character array
1050 reported by Ingo H. de Boer.
1051
1052 2007-08-27 Németh László <nemeth at OOo>:
1053 * suggestmgr.hxx: change bad time_t to clock_t in header, too.
1054 Bug reports or patches by Ingo H. de Boer under SF.net
1055 Bug ID 1781951, János Mohácsi and Gábor Zahemszky, András Tímár,
1056 OMax3 at SF.net under SF.net Bug ID 1781592.
1057
1058 * phonet.*: change variable-length local character array to
1059 portable fixed size character array. Problem reported by
1060 Ingo H. de Boer under SF.net Bug ID 1781951 and
1061 Ryan VanderMeulen.
1062
1063 * suggestmgr.cxx: remove debug message (also by
1064 Ingo H. de Boer).
1065
1066 2007-08-26 Ingo H. de Boer <idb_winshell at SF.net>:
1067 * win_api/Hunspell.vcproj: updated version (with phonet.*)
1068
1069 2007-08-23 Németh László <nemeth at OOo>:
1070 * phonet.{c,h}xx, suggestmgr.cxx: PHONE parameter:
1071 pronounciation based suggestion using Björn Jacke's original Aspell
1072 phonetic transcription algorithm (http://aspell.net), relicensed
1073 under GPL/LGPL/MPL tri-license with the permission of the author.
1074 Usage: see manual.
1075
1076 * affixmgr,suggestmgr.cxx: add KEY parameter for keyboard and
1077 input method error related suggestions.
1078 Example: KEY qwertyuiop|asdfghjkl|zxcvbnm
1079
1080 * man/hunspell.4: description about PHONE and KEY suggestion parameters.
1081
1082 * suggestmgr.cxx: enhancements for better suggestions:
1083 - Set ngram suggestions for badchar-type errors
1084 and only two word and compound word suggestions, too.
1085 - Separate not compound and compound word
1086 suggestions for MAP suggestion, too.
1087 - Double swap suggestions for short words.
1088 For example: ahev -> have, hwihc -> which.
1089 - Better time limits using clock() instead of time()
1090 (tenths of a second resolution instead of second ones).
1091 - leftcommonsubstring() weigth function.
1092
1093 * htype.hxx, hashmgr.cxx: blen (byte length) and clen (character
1094 length) fields instead of wlen
1095
1096 * affixmgr.cxx: fix get_syllable() for bad Unicode inputs.
1097
1098 * tests/suggestiontest/*: test environment for suggestions
1099
1100 2007-08-07 Martijn Wargers:
1101 * csutil.cxx: fix Mingw build error associated with ToUpper() call.
1102 Report and patch in Mozilla Issue 391447.
1103
1104 2007-08-07 Robert Longson:
1105 * atypes.cxx: use empty inline function HUNSPELL_WARNING instead of
1106 variadic macros to switch of Hunspell warnings.
1107 Reported by Gavin Sharp in Mozilla Issue 391147.
1108
1109 2007-08-05 Ginn Chen:
1110 * hashmgr.cxx: Hunspell failed to compile on OpenSolaris (use stdio
1111 instead of csdio). Report and patch in Mozilla Issue 391040.
1112
1113 2007-07-25 Németh László <nemeth at OOo>:
1114 * parsers/*.cxx: Hunspell executable recognises and accepts URLs,
1115 e-mail addresses, directory paths, reported by Jeppe Bundsgaard.
1116 * src/tools/hunspell.cxx: --check-url: new option of Hunspell program.
1117 Use --check-url, if you want check URLs, e-mail addresses and paths.
1118
1119 * parsers/textparser.cxx: strip colon at end of words for Finnish
1120 and Swedish (colon may be in words in Finnish and Swedish).
1121 Problem reported by Lars Aronsson.
1122 * tests/colons_in_words.*: test data
1123
1124 * tests/digits_in_words.*: example for using digits in words
1125 (eg. 1-jährig, 112-jährig etc. in German), reported by Lars Aronsson.
1126
1127 * hashmgr.cxx: Hunspell accepts allcaps forms of mixed case
1128 words of personal dictionaries (+allcaps custom dictionary words with
1129 allcaps affixes).
1130 Sf.net Bug ID 1755272, reported by Ellis Miller.
1131
1132 * hashmgr.cxx: fix small memory leaks with alias compressed
1133 dictionaries (free flag vectors of affixed personal dictionary words
1134 and flag vectors of hidden capitalized forms of mixed case and
1135 allcaps words).
1136
1137 * affixmgr.cxx: fix COMPOUNDRULE checking with affixed compounds.
1138 Sf.net Bug ID 1706659, reported by Björn Jacke. Also fixing for
1139 OOo Issue 76067 (crash-like deceleration for hexadecimal numbers
1140 with long FFFFFF sequence using en_US dictionary).
1141
1142 * tools/hunspell.cxx: add missing return to save_privdic().
1143
1144 * man/hunspell.4: add information about affixation of personal words:
1145 "Personal dictionaries are simple word lists, but with optional
1146 word patterns for affixation, separated by a slash:
1147
1148 foo
1149 Foo/Simpson
1150
1151 In this example, "foo" and "Foo" are personal words, plus Foo
1152 will be recognised with affixes of Simpson (Foo's etc.)."
1153
1154 2007-07-18 Németh László <nemeth at OOo>:
1155 * src/win_api/: add missing resource files, reported by Ingo H. de Boer.
1156
1157 2007-07-16 Németh László <nemeth at OOo>:
1158 * hunspell.cxx: fix dot removing from UTF-8 encoded words in cleanword2()
1159 (Capitalised words with dots, as "Something." were not recognised
1160 using Unicode encoded dictionaries.)
1161 * tests/{base.*,base_utf.*}: extended and new test files for
1162 dot removing and Unicode support.
1163
1164 * tools/hunspell.cxx: fix Cygwin, OS X compatibility using platform
1165 specifics iconv() header by ICONV_CONST macro of Autoconf.
1166 Sf.net Bug ID 1746030, reported by Mike Tian-Jian Jiang.
1167 Sf.net Bug ID 1753939, reported by Jean-Christophe Helary.
1168
1169 * tools/hunspell.cxx: fix missing global path setting with -d option.
1170
1171 * tests/test.sh: fix broken Valgrind checking (missing warnings
1172 with VALGRIND=memcheck make check).
1173
1174 * csutil.cxx: fix condition in u8_u16() to avoid invalid read
1175 of not null-terminated character arrays (detected by Valgrind
1176 in Hunspell executable: associated with 8-bit character table
1177 conversion in tools/hunspell.cxx).
1178
1179 * csutil.cxx: free_utf_tbl(): use utf_tbl_count-- instead of utf_tbl--.
1180 Memory leak in Hunspell executable detected by Valgrind.
1181
1182 * hashmgr.cxx: add missing free_utf_tbl(), memory leak in Hunspell
1183 executable detected by Valgrind.
1184
1185 * hashmgr.cxx: load_tables(): fix memory error in spec. capitalization.
1186 Use sizeof(unsigned short) instead of bad sizeof(unsigned short*).
1187 Invalid memory read detected by Valgrind.
1188
1189 * hashmgr.cxx: add_word(): fix memory error in spec. capitalization.
1190 Update also affix array length of capitalized homonyms. Invalid
1191 memory read detected by Valgrind.
1192
1193 * hunspell.cxx: suggest(): fix invalid memory write and leak.
1194 Bad realloc() and missing free() detected by Valgrind associated
1195 with suggestions for "something.The" type spelling errors.
1196
1197 * {dictmgr,csutil,hashmgr,suggestmgr}.cxx: check memory allocation.
1198 Sf.net Bug ID 1747507, based on the patch by Jose da Silva.
1199
1200 2007-07-13 Ingo H. de Boer <idb_winshell at SF.net>:
1201 * atypes.cxx: fix Visual C compatibility: Using
1202 "HUNSPELL_WARNING(a,b,...} {}" macro instead of empty "X(a,b...)".
1203
1204 * hunspell.cxx: changes for Windows API.
1205 * win_api/Hunspell.*: new resource files
1206 * win_api/hunspelldll.*: set optional Hunspell and Borland spec. codes
1207 Sf.net Bug ID 1753802, patch by Ingo H. de Boer.
1208 See also Sf.net Bug ID 1751406, patch by Mike Tian-Jian Jiang.
1209
1210 2007-07-09 Caolan McNamara <cmc at OO.o>:
1211 * {hunspell,hashmgr,affentry}.cxx: fix warnings of Coverity program
1212 analyzer. Sf.net Bug ID, 1750219.
1213
1214 2007-07-06 Németh László <nemeth at OOo>:
1215 * atypes.cxx: warning-free swallowing of conditional warning messages
1216 and their parameters using empty HUNSPELL_WARNING(a,b...) macro.
1217 * {affixmgr,atypes,csutil}.cxx: fix unused variable warnings
1218 using WARNVAR macro for conditionally named variables.
1219 * hashmgr.cxx: fix unused variable warning in add_word() by cond. name
1220 * hunspell.cxx: fix shadowed declaration of captype var. in suggest()
1221
1222 2006-06-29 Caolan McNamara <cmc at OO.o>:
1223 * hunspell.cxx: patch to fix possible memory leak in analyze() of
1224 experimental morphological analyzer code. Sf.net Bug ID 1745263.
1225
1226 2007-06-29 Németh László <nemeth at OOo>:
1227 improvements:
1228 * src/hunspell/hunspell.cxx: check bad capitalisation of Dutch letter IJ.
1229 - Sf.net Feature Request ID 1640985, reported by Frank Fesevur.
1230 - Solution: FORBIDDENWORD for capitalised word forms (need
1231 an improved Dutch dictionary with forbidden words: Ijs/*, etc.).
1232 * tests/IJ.*: test data and example.
1233
1234 * hashmgr.cxx, hunspell.cxx: check capitalization of special word forms
1235 - words with mixed capitalisation: OpenOffice.org - OPENOFFICE.ORG
1236 Sf.net Bug ID 1398550, reported by Dmitri Gabinski.
1237 - allcap words and suffixes: UNICEF's - UNICEF'S
1238 - prefixes with apostrophe and proper names: Sant'Elia - SANT'ELIA
1239 For Catalan, French and Italian languages.
1240 Reported by Davide Prina in OOo Issue 68568.
1241 * tests/allcaps*: tests for OPENOFFICE.ORG, UNICEF'S capitalization.
1242 * tests/i68568*: tests for SANT'ELIA capitalization.
1243
1244 * hunspell/hunspell.cxx: suggestion for missing sentence spacing:
1245 something.The -> something. The
1246
1247 * tools/hunspell.cxx: multiple character encoding support
1248 - -i option: custom input encoding
1249 Sf.net Bug ID 1610866, reported by Thobias Schlemmer.
1250 Sf.net Bug ID 1633413, reported by Dan Kenigsberg.
1251 See also hunspell-1.1.5-encoding.patch of Fedora from Caolan Mc'Namara.
1252 * tests/*.test: add input encodings
1253
1254 * tools/hunspell.cxx: use locale data for default dictionary names.
1255 Sf.net Bug ID 1731630, report and patch from Bernhard Rosenkraenzer,
1256 See also hunspell-1.1.4-defaultdictfromlang.patch of Fedora Linux
1257 from Caolan McNamara.
1258
1259 * tools/hunspell.cxx: fix 8-bit tokenization (letters without
1260 casing, like ß or Hebrew characters now are handled well)
1261
1262 * tools/hunspell.cxx: dictionary search path
1263 - DICPATH environmental variable
1264 - -D option: show directory path of loaded dictionary
1265 - automatic detection of OpenOffice.org directories
1266
1267 fixes:
1268 * affixmgr.cxx: fault-tolerant patch for REP and other affix
1269 table data problems. Problem with Hunspell and en_GB dictionary
1270 reported by Thomas Lange in OOo Issue 76098 and
1271 Stephan Bergmann in OOo Issue 76100.
1272 Sf.net Bug ID 1698240, reported by Ingo H. de Boer.
1273
1274 * csutil.cxx: fix mkallcap_utf() for allcaps suggestion in UTF-8.
1275
1276 * suggestmgr.cxx: fix bad movechar_utf() (missing strlen()).
1277
1278 * hunspell.cxx: fix bad degree sign detection in Unicode
1279 hu_HU environment.
1280
1281 * hunspell/hunspell.cxx: free allocated memory of csconv in
1282 ported Mozilla code.
1283 - Mozilla Bugzilla Bug 383564, report and Mozilla MySpell patch
1284 by Andrew Geul. Reported by Ryan VanderMeulen for Hunspell.
1285
1286 * suggestmgr.cxx: fix minor difference in Unicode suggestion
1287 (ngram suggestion of allcaps words in Unicode).
1288
1289 * hashmgr.cxx: close file handle after errors.
1290 Sf.net Bug ID 1736286, reported by John Nisly.
1291
1292 * configure.ac: syntax error (shell variable with spaces).
1293 Sf.net Bug ID 1731625, reported by Bernhard Rosenkraenzer.
1294
1295 * hunspell.cxx: check_word(): fix bad usage of info pointer.
1296
1297 * hashmgr.cxx: fix de_DE related bug (accept words with leading dash).
1298 Sf.net Bug ID 1696134, reported by Björn Jacke.
1299
1300 * suggestmgr.cxx, tests/1695964.*: fix NEEDAFFIX homonym suggestion.
1301 Sf.net Bug ID 1695964, reported by Björn Jacke.
1302
1303 * tests/1463589*: capitalized ngram suggestion test data for
1304 Sf.net Bug ID 1463589, reported by Frederik Fouvry.
1305
1306 * csutil.cxx, affixmgr.cxx: fix possible heap error with
1307 multiple instances of utf_tbl.
1308 Sf.net Bug ID 1693875, reported by Ingo H. de Boer.
1309
1310 * affixmgr.cxx, suggestmgr.cxx, license.hunspell: convert to ASCII.
1311 Locale dependent compiling problems. Sf.net Bug ID 1694379, reported
1312 by Mike Tian-Jian Jiang. OOo Issue 78018 reported by Thomas Lange.
1313
1314 * tests/test.sh: compatibility issues
1315 - fix Valgrind support (check shared library instead of shell wrapper)
1316 - remove deprecated "tail +2" syntax
1317 - set 8-bit locale for testing (LC_ALL=C)
1318
1319 * hunspell.hxx: remove license.* and config.h dependencies.
1320 - hunspell-1.1.5-badheader.patch from Caolan McNamara <cmc at OO.o>
1321
1322 2007-03-21 Németh László <nemeth at OOo>:
1323 * tools/Makefile.am, munch.h, unmunch.h: add missing munch.h and unmunch.h
1324 Reported by Björn Jacke and Khaled Hosny (sf.net Bug ID 1684144)
1325 * hunspell/hunspell.cxx, hunspell.hxx: fix --with-ui compliling error (add get_csconv())
1326 Reported by Khaled Hosny (sf.net Bug ID 1685010)
1327
1328 2007-03-19 Németh László <nemeth at OOo>:
1329 * csutil.cxx, hunspell/hunspell.cxx: Unicode non BMP area (>65K character range) support
1330 (except conditional patterns and strip characters of affix rules)
1331 * tests/utf8_nonbmp*: test data
1332
1333 * src/hunspell/*: add Mozilla patches from David Einstein
1334 - run-time generated 8-bit character tables
1335 - other Mozilla related changes (see Mozilla Bugzilla Bug 319778)
1336
1337 * csutil.cxx, affixmgr.cxx, hashmgr.cxx: optimized version of IGNORE feature
1338 - IGNORE works with affixes (except strip characters and affix conditions)
1339 * tests/ignore*: test data with latin characters
1340 * tests/ignoreutf*: Unicode test data with Arabic diacritics (Harakat)
1341
1342 * src/hunspell/suggestmgr.cxx: new edit distance suggestion methods
1343 - capitalization: nasa -> NASA
1344 - long swap: permenant -> permanent
1345 - long mov.: Ghandi -> Gandhi
1346 - double two characters: vacacation -> vacation
1347 * tests/sug.*: test data
1348
1349 * src/hunspell/affixmgr.cxx: space in REP strings (alot -> a lot)
1350 Note: Underline character signs the space in REP strings: REP alot a_lot, and
1351 put the expression with space ("a lot") into the dic file (see tests/sug).
1352
1353 * hashmgr.cxx, affixmgr.cxx: ignore Unicode byte order mark (BOM sequence)
1354 * tests/utf8_bom*: test data
1355
1356 * hunspell/*.cxx: OOo Issue 68903 - Make lingucomponent warning-free on wntmsci10
1357 - fix Hunspell related warning messages on Windows platform (except some assignment
1358 within conditional expressions). Reported and started by Stephan Bergmann.
1359
1360 * hunspell/affixmgr.cxx: fix OOo Issue 66683 - hunspell dmake debug=x fails
1361 - Reported by Stephan Bergmann.
1362
1363 * src/hunspell/hunspell.[ch]xx: thread safe API for Hunspell executable
1364 (removing prev*() functions, new spell(word, info, root) function)
1365
1366 * configure.ac, src/hunspell/*: HUNSPELL_EXPERIMENTAL code
1367 --with-experimental configure option (conditional compiling of morphological analyser
1368 and stemmer tools)
1369
1370 * configure.ac, src/hunspell/*: conditional Hunspell warning messages
1371 --with-warnings configure option
1372
1373 * affixmgr.cxx: new, optimized parsing functions
1374
1375 * affixmgr.cxx: fix homonym handling for German dictionary project,
1376 reported by Björn Jacke (sf.net Bug ID 1592880).
1377 * tests/1592880.*: test data by Björn Jacke
1378
1379 * src/hunspell/affixmgr.cxx: fix CIRCUMFIX suggestion
1380 Bug reported by Erdal Ronahi.
1381
1382 * hunspell.cxx: reverse root word output (complex prefixes)
1383 Bug reported by Munzir Taha.
1384
1385 * tools/hunspell.cxx: fix Emacs compatibility, patch by marot at sf.net
1386 - no % command in PIPE mode (SourceForge BugTracker 1595607)
1387 - fix HUNSPELL_VERSION string
1388
1389 * suggestmgr.[hc]xx: rename check() functions to checkword() (OOo Issue 68296)
1390 adopt MySpell patch by Bryan Petty (tierra at ooo) for Hunspell source
1391
1392 * csutil.cxx, munch.c, unmunch.c: adopt relevant parts of the MinGW patch
1393 (OOo Issue 42504) by tonal at ooo
1394
1395 * affigmgr.cxx: remove double candidate_check() call, reported by Bram Moolenaar
1396
1397 * tests/test.sh: add LC_ALL="C" environment. Locale dependency of make check
1398 reported by Gentoo project.
1399
1400 * src/tools/hunspell.cxx: UTF-8 highlighting fix for console UI
1401 (not solved: breaking long UTF-8 lines)
1402
1403 * src/tools/unmunch.c: fix bad generation if strip is shorter than condition,
1404 reported by Davide Prina
1405 * src/tools/unmunch.h: increase 5000 -> 500000
1406
1407 * src/tools/hunspell.cxx: fix memory error in suggestion (uninitialized parameter),
1408 Bug also reported by Björn Jacke in SourceForge Bug 1469957
1409
1410 * csutil.cxx, affixmgr.cxx: fix Caolan McNamara's patch for non OOo environment
1411
1412 2006-11-11 Caolan McNamara <cmc at OO.o>:
1413 * csutil.cxx, affixmgr.cxx: UTF-8 table patch (OOo Issue 71449)
1414 Description: memory optimization (OOo doesn't use the large UTF-8 table).
1415
1416 * Makefile.am: shared library patch (Sourceforge ID 1610756)
1417
1418 * hunspell.h, hunspell.cxx: C API patch (Sourceforge ID 1616353)
1419
1420 * hunspell.pc: pkgconfig patch (Sourceforge ID 1639128)
1421
1422 2006-10-17 Ryan Jones <at Mozilla Bugzilla>:
1423 * affixmgr.cxx: missing fclose(affixlst) calls
1424 Reported by <gavins at ooo> in OOo Issue 70408
1425
1426 2007-07-11 Taha Zerrouki <taha at gawab>:
1427 * affixmgr.cxx, hunspell.cxx, hashmgr.cxx, csutil.cxx: IGNORE feature to remove
1428 optional Arabic and other characters from input and dictionary words.
1429 * src/hunspell/langnum.hxx: add Arabic language number, lang_ar=96
1430 * tests/ignore.*: test data
1431
1432 2006-05-28 Miha Vrhovnik <mvrhov at users.sourceforge>:
1433 * src/win_api/*: C API for Windows DLLs
1434 - also Delphi text editor example (see on Hunspell Sourceforge page)
1435
1436 2006-05-18 Kevin F. Quinn <kevquinn at gentoo>:
1437 * utf_info.cxx: struct -> static struct
1438 Shared library patch also developed by Gentoo developers (Hanno Meyer-Thurow,
1439 Diego Pettenò, Kevin F. Quinn)
1440
1441 2006-02-02 Németh László <[email protected]>:
1442 * src/hunspell/hunspell.cxx: suggest(): replace "fooBar" -> "foo bar" suggestions
1443 with "fooBar" ->"foo Bar" (missing spaces are typical OCR bugs).
1444 Bug reported by stowrob at OOo in Issue 58202.
1445 * src/hunspell/suggestmgr.cxx: twowords(): permit 1-character words.
1446 (restore MySpell's original behavior). Here: "aNew" -> "a New".
1447 * tests/i58202.*: test data
1448
1449 * src/parsers/textparser.cxx: fix Unicode tokenization in is_wordchar()
1450 (extra word characters (WORDCHARS) didn't work on big-endian platforms).
1451
1452 * src/hunspell/{csutil,affixmgr}.cxx: inline isSubset(), isRevSubset():
1453 little speed optimalization for languages with rich morphology.
1454
1455 * src/tools/hunspell.cxx: fix bad --with-ui and --with-readline compiling
1456 when (N)curses is missing. Reported by Daniel Naber.
1457
1458 2006-01-19 Tor Lillqvist <[email protected]>
1459 * src/hunspell/csutil.cxx: mystrsep(): fix locale-dependent isspace() tokenization
1460
1461 2006-01-06 András Tímár <[email protected]>
1462 * src/hunspell/{hashmgr.hxx,hunspell.cxx}: fix Visual C++ compiling errors
1463
1464 2006-01-05 Németh László <[email protected]>:
1465 * COPYING: set GPL/LGPL/MPL tri-license for Mozilla integration.
1466 Rationale: Mozilla source code contains an old MySpell version
1467 with GPL/LGPL/MPL tri-license. (MPL license is a copyleft license, similar
1468 to the LGPL, but it acts on file level.)
1469 * COPYING.LGPL: GNU Lesser General Public License 2.1 (LGPL)
1470 * COPYING.MPL: Mozilla Public License 1.1 (MPL)
1471 * license.hunspell, src/hunspell/license.hunspell: GPL/LGPL/MPL tri-license
1472
1473 * src/hunspell/{affixmgr,hashmgr}.*: AF, AM alias definitions in affix file:
1474 compression of flag sets and morphological descriptions (see manual,
1475 and tests/alias* test files).
1476 Rationale: Alias compression is also good for loading time and memory
1477 efficiency, not only smaller resources.
1478 * src/tools/makealias: alias compression utility
1479 (usage: ./makealias file.dic file.aff)
1480 * tests/alias{,2,3}: AF, AM tests
1481 * man/hunspell.4: add AF, AM documentation
1482 * src/hunspell/affentry.cxx, atypes.hxx: add new opts bits (aeALIASM, aeALIASF)
1483
1484 * tools/hunspell, src/parser/*, src/hunspell/*: Hunspell program
1485 tokenizes Unicode texts (only with UTF-8 encoded dictionaries).
1486 Missing Unicode tokenization reported by Björn Jacke, Egmont Koblinger,
1487 Jess Body and others.
1488 Note: Curses interactive interface hasn't worked perfectly yet.
1489 * tests/*.tests: remove -1 parameters of Hunspell
1490 * tests/*.{good,wrong}: remove tabulators
1491
1492 * src/hunspell/{hunspell,affixmgr}.cxx: BREAK option: break words at
1493 specified break points and checking word parts separately (see manual).
1494 Note: COMPOUNDRULE is better (or will be better) for handling dashes and
1495 other compound joining characters or character strings. Use BREAK, if you
1496 want check words with dashes or other joining characters and there is no time
1497 or possibility to describe precise compound rules with COMPOUNDRULE.
1498 * tests/break.*: BREAK example.
1499
1500 * src/hunspell/{affixmgr,hunspell}.cxx: add CHECKSHARPS declaration instead
1501 of LANG de_DE definitions to handle German sharp s in both spelling and
1502 suggestion.
1503 * src/hunspell/hunspell.cxx: With CHECKSHARPS, uppercase words are valid
1504 with both lower sharp s (it's is optional for names in German legal texts)
1505 and SS (MÜßIG, MÜSSIG). Missing lower sharp s form reported by Björn Jacke.
1506 * src/hunspell/hunspell.cxx: KEEPCASE flag on a sharp s word has a special
1507 meaning with CHECKSHARPS declaration: KEEPCASE permits capitalisation and SS upper
1508 casing of a sharp s word (Müßig and MÜSSIG), but forbids the upper cased form
1509 with lower sharp s character(s): *MÜßIG.
1510 * tests/germancompounding*: add CHECKSHARPS, remove LANG
1511 * tests/checksharps*: add CHECKSHARPS and KEEPCASE, remove LANG
1512
1513 * src/hunspell/hunspell.cxx: improved suggestions:
1514 - suggestions for pressed Caps Lock problems: macARONI -> macaroni
1515 - suggestions for long shift problems: MAcaroni -> Macaroni, macaroni
1516 - suggestions for KEEPCASE words: KG -> kg
1517 * src/hunspell/csutil.cxx: fix mystrrep() function:
1518 - suggestions for lower sharp s in uppercased words: MÜßIG -> MÜSSIG
1519 * tests/checksharps{,utf}.sug: add tests for mystrrep() fix
1520
1521 * src/hunspell/hashmgr.cxx: Now dictionary words can contain slashes
1522 with the "\/" syntax. Problem reported by Frederik Fouvry.
1523
1524 * src/hunspell/hunspell.cxx: fix bad duplicate filter in suggest().
1525 (Suggesting some capitalised compound words caused program crash
1526 with Hungarian dictionary, OOo Issue 59055).
1527
1528 * src/hunspell/affixmgr.cxx: fix bad defcpd_check() call in compound_check().
1529 (Overlapping new COMPOUNDRULE and old compounding methods caused program
1530 crash at suggestion.)
1531
1532 * src/hunspell/affixmgr.{cxx,hxx}: check affix flag duplication at affix classes.
1533 Suggested by Daniel Naber.
1534
1535 * src/hunspell/affentry.cxx: remove unused variable declarations (OOo i58338).
1536 Compiler warnings reported by András Tímár and Martin Hollmichel.
1537
1538 * src/hunspell/hunspell.cxx: morph(): not analyse bad mixed uppercased forms
1539 (fix Arabic morphological analysis with Buckwalter's Arabic transliteration)
1540
1541 * src/hunspell/affentry.{cxx,hxx}, atypes.hxx: little memory optimization
1542 in affentry:
1543 - using unsigned char fields instead of short (stripl, appndl, numconds)
1544 - rename xpflg field to opts
1545 - removing utf8 field, use aeUTF8 bit of opts field
1546
1547 * configure.ac: set tests/maputf.test to XFAILED on ARM platform.
1548 Fail reported by Rene Engelhard.
1549
1550 * configure.ac: link Ncursesw library, if exists.
1551
1552 * BUGS: add BUGS file
1553
1554 * tests/complexprefixes2.*: test for morphological analysis with COMPLEXPREFIXES
1555
1556 * src/hunspell/affixmgr.cxx: use "COMPOUNDRULE" instead of
1557 "COMPOUND". The new name suggested by Bram Moolenaar.
1558 * tests/compoundrule*: modified and renamed compound.* test files
1559
1560 * man/hunspell.4: AF, AM, BREAK, CHECKSHARPS, COMPOUNDRULE, KEEPCASE.
1561 - also new addition to the documentation:
1562 Header of the dictionary file define approximate dictionary size:
1563 ``A dictionary file (*.dic) contains a list of words, one per line.
1564 The first line of the dictionaries (except personal dictionaries)
1565 contains the _approximate_ word count (for optimal hash memory size).''
1566 Asked by Frederik Foudry.
1567
1568 One-character replacements in REP definitions: ``It's very useful to
1569 define replacements for the most typical one-character mistakes, too:
1570 with REP you can add higher priority to a subset of the TRY suggestions
1571 (suggestion list begins with the REP suggestions).''
1572
1573 2005-11-11 Németh László <[email protected]>:
1574 * src/hunspell/affixmgr.*: fix Unicode MAP errors (sorted only n-1
1575 characters instead of n ones in UTF-16 MAP character lists).
1576 Bug reported by Rene Engelhard.
1577
1578 * src/hunspell/affixmgr.*: fix infinite COMPOUND matching (default char
1579 type is unsigned on PowerPC, s390 and ARM platforms and it will never
1580 be negative). Bug reported by Rene Engelhard.
1581
1582 * src/hunspell/{affixmgr,suggestmgr}.cxx: fix bad ONLYINCOMPOUND
1583 word suggestions.
1584 * tests/onlyincompound.sug: empty test file to check this fix.
1585 Bug reported by Björn Jacke.
1586
1587 * src/hunspell/affixmgr.cxx: fix backtracking in COMPOUND pattern matching.
1588 * tests/compound6.*: test files to check this fix.
1589
1590 * csutil.cxx: set bigger range types in flag_qsort() and flag_bsearch().
1591
1592 * affixmgr.hxx: set better type for cont_classes[] Boolean data (short -> char)
1593
1594 * configure.ac, tests/automake.am: set platform specific XFAIL test
1595 (flagutf8.test on ARM platform)
1596
1597 2005-11-09 Németh László <[email protected]>:
1598 improvements:
1599 * src/hunspell/affixmgr.*: new and improved affix file parameters:
1600
1601 - COMPOUND definitions: compound patterns with regexp-like matching.
1602 See manual and test files: tests/compound*.*
1603 Suggested by Bram Moolenaar.
1604 Also useful for simple word-level lexical scanning, for example
1605 analysing numbers or words with numbers (OOo Issue #53643):
1606 http://qa.openoffice.org/issues/show_bug.cgi?id=53643
1607 Examples: tests/compound{4,5}.*.
1608
1609 - NOSUGGEST flag: words signed with NOSUGGEST flag are not suggested.
1610 Proposed flag for vulgar and obscene words (OOo Issue #55498).
1611 Example: tests/nosuggest.*.
1612 Problem reported by bobharvey at OOo:
1613 http://qa.openoffice.org/issues/show_bug.cgi?id=55498
1614
1615 - KEEPCASE flag: Forbid capitalized and uppercased forms of words
1616 signed with KEEPCASE flags. Useful for special ortographies
1617 (measurements and currency often keep their case in uppercased
1618 texts) and other writing systems (eg. keeping lower case of IPA
1619 characters).
1620
1621 - CHECKCOMPOUNDCASE: Forbid upper case characters at word bound in compounds.
1622 Examples: tests/checkcompoundcase* and tests/germancompounding.*
1623
1624 - FLAG UTF-8: New flag type: Unicode character encoded with UTF-8.
1625 Example: tests/flagutf8.*.
1626 Rationale: Unicode character type can be more readable
1627 (in a Unicode text editor) than `long' or `num' flag type.
1628
1629 bug fixes:
1630 * src/hunspell/hunspell.cxx: accept numbers and numbers with separators (i53643)
1631 Bug reported by skelet at OOo:
1632 http://qa.openoffice.org/issues/show_bug.cgi?id=53643
1633
1634 * src/hunspell/csutil.cxx: fix casing data in ISO 8859-13 character table.
1635
1636 * src/hunspell/csutil.cxx: add ISO-8859-15 character encoding (i54980)
1637 Rationale: ISO-8859-15 is the default encoding of the French OpenOffice.org
1638 dictionary. ISO-8859-15 is a modified version of ISO-8859-1
1639 (latin-1) character encoding with French œ ligatures and euro
1640 symbol. Problem reported by cbrunet at OOo in OOo Issue 54980:
1641 http://qa.openoffice.org/issues/show_bug.cgi?id=54980
1642
1643 * src/hunspell/affixmgr.cxx: fix zero-byte malloc after a bad affix header.
1644 Patch by Harri Pitkänen.
1645
1646 * src/hunspell/suggestmgr.cxx: fix bad NEEDAFFIX word suggestion
1647 in ngram suggestions. Reported by Daniel Naber and Friedel Wolff.
1648
1649 * src/hunspell/hashmgr.cxx: fix bad white space checking in affix files.
1650 src/hunspell/{csutil,affixmgr}.cxx: add other white space separators.
1651 Problems with tabulators reported by Frederik Fouvry.
1652
1653 * src/hunspell/*: replace system-dependent <license.*> #include
1654 parameters with quoted ones. Problem reported by Dafydd Jones.
1655
1656 * src/hunspell/hunspell.cxx: fix missing morphological analysis of dot(s)
1657 Reported by Trón Viktor.
1658
1659 changes:
1660 * src/hunspell/affixmgr.cxx: rename PSEUDOROOT to NEEDAFFIX.
1661 Suggested by Bram Moolenaar.
1662
1663 * src/hunspell/suggestmgr.hxx: Increase default maximum of
1664 ngram suggestions (3->5). Suggested by Kevin Hendricks.
1665
1666 * src/hunspell/htypes.hxx: Increase MAXDELEN for long affix flags.
1667
1668 * src/hunspell/suggestmgr.cxx: modify (perhaps fix) Unicode map suggestion.
1669 tests/maputf test fail on ARM platform reported by Rene Engelhard.
1670
1671 * src/hunspell/{affentry.cxx,atypes.hxx}: remove [PREFIX] and
1672 MISSING_DESCRIPTION messages from morphological analysis.
1673 Problems reported by Trón Viktor.
1674
1675 * tests/germancompounding.{aff,good}: Add "Computer-Arbeit" test word.
1676 Suggested by Daniel Naber.
1677
1678 * doc/man/hunspell.4: Proof-reading patch by Goldman Eleonóra.
1679
1680 * doc/man/hunspell.4: Fix bad affix example (replace `move' with `work').
1681 Bug reported by Frederik Fouvry.
1682
1683 * tests/*: new test files:
1684 affixes.*: simple affix compression example from Hunspell 4 manual page
1685 checkcompoundcase.*, checkcompoundcase2.*, checkcompoundcaseutf.*
1686 compound.*, compound2.*, compound3.*, compound4.*, compound5.*
1687 compoundflag.* (former compound.*)
1688 flagutf8.*: test for FLAG UTF-8
1689 germancompounding.*: simplification with CHECKCOMPOUNDCASE.
1690 germancompoundingold.* (former germancompounding.*)
1691 i53643.*: check numbers with separators
1692 i54980.*: ISO8859-15 test
1693 keepcase.*: test for KEEPCASE
1694 needaffix*.* (former pseudoroot*.* tests)
1695 nosuggest.*: test for NOSUGGEST
1696
1697 2005-09-19 Németh László <[email protected]>:
1698 * src/hunspell/suggestmgr.cxx: improved ngram suggestion:
1699 - detect not neighboring swap characters (pernament -> permanent)
1700 Rationale: ngram method has a significant error with not neighboring
1701 swap characters, especially when swap is in the middle of the word.
1702 - suggest uppercase forms (unesco -> UNESCO, siggraph's -> SIGGRAPH's)
1703 - suggest only ngram swap character and uppercase form, if they exist.
1704 Rationale: swap character and casing equivalence give mutch better
1705 suggestions as any other (weighted) ngram suggestions.
1706 - add uppercase suggestion (PERMENANT -> PERMANENT)
1707
1708 * src/hunspell/*: complete comparison with MySpell 3.2 (in OOo beta 2):
1709 - affixmgr.cxx: add missing numrep initialization
1710 - hashmgr.cxx: add_word(): don't allocate temporary records
1711 - hunspell.cxx: in suggest():
1712 - check capitalized words first (better sug. order for proper names),
1713 - check pSMgr->suggest() return value
1714 - set pSMgr->suggest() call to not optional in HUHCAP
1715 - csutil.cxx: fix bad KOI8-U -> koi8r_tbl reference in enc_entry encds
1716 - csutil.cxx: fix casing data in ISO 8859-2, Windows 1251 and KOI8-U
1717 encoding tables. Bug reported by Dmitri Gabinski.
1718
1719 * src/hunspell/affixmgr.*: improved compound word and other features
1720 - generalize hu_HU specific compound word features with new affix file
1721 parameters, suggested by Bram Moolenaar:
1722 - CHECKCOMPOUNDDUP: forbid word duplication in compounds (eg. foo|foo)
1723 - CHECKCOMPOUNDTRIPLE: forbid triple letters in compounds (eg. foo|obar)
1724 - CHECKCOMPOUNDPATTERN: forbid patterns at word bounds in compounds
1725 - CHECKCOMPOUNDREP: using REP replacement table, forbid presumably bad
1726 compounds (useful for languages with unlimited number of compounds)
1727 - ONLYINCOMPOUND flag works also with words (see tests/onlyincompound.*)
1728 Suggested by Daniel Naber, Björn Jacke, Trón Viktor & Bram Moolenaar.
1729 - PSEUDOROOT works also with prefixes and prefix + suffix combinations
1730 (see tests/pseudoroot5.*). Suggested by Trón Viktor.
1731 - man/hunspell.4: updated man page
1732
1733 * src/hunspell/affixmgr.*: fix incomplete prefix handling with twofold
1734 suffixes (delete unnecessary contclasses[] conditions in
1735 prefix_check_twosfx() and prefix_check_twosfx_morph()).
1736 Bug reported by Trón Viktor.
1737
1738 * src/hunspell/affixmgr.*: complete also *_morph() functions with
1739 conditions of new Hunspell features (circumfix, pseudoroot etc.).
1740
1741 * src/hunspell/suggestmgr.cxx:
1742 - fix missing suggestions for words with crossed prefix and suffix
1743 - fix redundant non-compound word checking
1744 - fix losing suggestions problem. Bug reported by Dmitri Gabinski.
1745
1746 * src/hunspell/dictmgr.*:
1747 - add new dictionary manager for Hunspell UNO modul
1748 Problems with eo_ANY Esperanto locale reported by Dmitri Gabinski.
1749
1750 * src/hunspell/*: use precise constant sizes for 8-bit and 16-bit character
1751 arrays with MAXWORDUTF8LEN and MAXSWUTF8L macros.
1752
1753 * src/hunspell/affixmgr.cxx: fix bad MAXNGRAMSUGS parameter handling
1754
1755 * src/hunspell/affixmgr.cxx, src/tools/{un}munch.*: fix GCC 4.0 warnings
1756 on fgets(), reported by Dvornik László
1757
1758 * po/hu.po: improved translation by Dvornik László
1759
1760 * tests/test.sh: improved test environment
1761 - add suggestion testing (see tests/*.sug)
1762 - add memory debugging environment, based on the excellent Valgrind debugger.
1763 Usage on Linux and experimental platforms of Valgrind:
1764 VALGRIND=memcheck make check
1765 - rename test_hunmorph to test.sh
1766
1767 * tests/*: new tests:
1768 - base.*: base example based on MySpell's checkme.lst.
1769 - map{,utf}.*, rep{,utf}: MAP and REP suggestion examples
1770 - tests on new CHECKCOMPOUND, ONLYINCOMPOUND and PSEUDOROOT features
1771 - i54633.*: capitalized suggestion test for Issue 54633 from OOo's Issuezilla
1772 - i35725.*: improved ngram suggestion test for Issue 35725
1773
1774 2005-08-26 Németh László <[email protected]>:
1775 improvements:
1776
1777 * src/hunspell/suggestmgr.cxx:
1778 Unicode support in related character map suggestion
1779
1780 * src/hunspell/suggestmgr.cxx: Unicode support in ngram suggestion
1781
1782 * src/hunspell/{suggestmgr,affixmgr,hunspell}.cxx: improve ngram suggestion.
1783 Fix http://qa.openoffice.org/issues/show_bug.cgi?id=35725. See release
1784 notes for examples. This problem reported by beccablain at OOo.
1785 - ngram suggestions now are case insensitive (see `Permenant' bug in Issuezilla)
1786 - weight ngram suggestions (with the longest common subsequent algorithm,
1787 also considering lengths of bad word and suggestion, identical first
1788 letters and almost completely identical character positions)
1789 - set strict affix congruency in expand_rootword(). Now ngram suggestions
1790 are good for languages with rich morphology and also better for English.
1791 Rationale: affixed forms of the first ngram suggestion
1792 very often suppress the second and subsequent root word suggestions. But
1793 faults in affixes are more uncommon, and can be fix without suggestions.
1794 We must prefer the more informative second and subsequent root word
1795 suggestions instead of the suggestions for bad affixes.
1796 - a better suggestion may not be substring of a less good suggestion
1797 Rationale: Suggesting affixed forms of a root word is
1798 unnecessary, when root word has got better weighted ngram value.
1799 (Checking substrings is a good approximation for this refinement.)
1800 - lesser ngram suggestions (default 3 maximum instead of 10)
1801 Rationale: For users need a big extra effort to check a lot of bad ngram
1802 suggestions, nine times out of ten unnecessarily. It is very
1803 distracting, because ngram suggestions could be very different.
1804 Usually Myspell and Hunspell suggest one or two suggestions with
1805 the old suggestion algorithms (maximum is 15), with ngram algorithm
1806 often gives maximum number suggestions. With strict affix congruency
1807 and other refinements, the good suggestion there is usually among the
1808 first three elements.
1809 - new affix parameter: MAXNGRAMSUG
1810
1811 * src/hunspell/*: support agglutinative languages with rich prefix
1812 morphology or with right-to-left writing system (for example, Turkic
1813 and Austronesian languages with (modified) Arabic scripts).
1814 - new affix parameter: COMPLEXPREFIXES
1815 Set twofold prefix stripping (but single suffix stripping)
1816 * src/hunspell/affixmgr.cxx:
1817 - speed up prefix loading with tree sorting algorithm.
1818 * tests/complexprefixes.*, tests/complexprefixesutf.*:
1819 Coptic example posted by Moheb Mekhaiel
1820
1821 * src/hunspell/hashmgr.cxx: check size attribute in dic file
1822 suggested by Daniel Naber
1823 Rationale: With missing size attribute Hunspell allocates too small and
1824 more slower hash memory, and Hunspell can lose first dictionary word.
1825
1826 * src/hunspell/affixmgr.cxx: check stripping characters and condition
1827 compatibility in affix rules (bugs detected in cs_CZ, es_ES, es_NEW,
1828 es_MX, lt_LT, nn_NO, pt_PT, ro_RO and sk_SK dictionaries). See release
1829 notes of Hunspell 1.0.9 in NEWS.
1830
1831 * src/hunspell/affixmgr.cxx: check unnecessary fields in affix rules
1832 (bugs detected in ro_RO and sv_SE dictionaries). See release notes.
1833
1834 * src/hunspell/affixmgr.cxx: remove redundant condition checking
1835 in affix rules with stripping characters (redundancy in OpenOffice.org
1836 dictionaries reported by Eleonóra Goldman)
1837 Rationale: this is a little optimization, but it was excellent for
1838 detect the bad ngram affixation with bad or weak affix conditions.
1839
1840 * tests/germancompounding.aff: improve compound definition
1841 - use dash prefix instead of language specific tokenizer
1842 Rationale: Using uniform approach is the right way to check and analyze
1843 compound words. Language specific word breaking is deprecated, need
1844 a sophisticated grammar checking for word-like word pairs
1845 (for example in Hungarian there is a substandard, but accepted
1846 syntax with dash for word pairs: cats, dogs -> kutyák-macskák (like
1847 cats/dogs in English).
1848
1849 * test Hunspell with 54 OpenOffice.org dictionaries: see release notes
1850
1851 bug fixes:
1852
1853 * src/hunspell/suggestmgr.*: add time limit to exponential
1854 algorithm of the related character map suggestion
1855 Rationale: a long word in agglutinative languages or a special pattern
1856 (for example a horizontal rule) made of map characters can `crash' the
1857 spell checker.
1858
1859 * src/hunspell/affentry.cxx: add() functions: fix bad word generation
1860 checking stripping characters (see similar bug in unmunch)
1861
1862 * src/hunspell/affixmgr.cxx: parse_file(): fix unconditional getNext()
1863 call for ~AffixMgr() when affix file is corrupt.
1864
1865 * src/hunspell/affixmgr.*: AffixMgr(), parse_cpdsyllable(): fix missing
1866 string duplications for ~AffixMgr() when affix file is corrupt.
1867
1868 * src/hunspell/affixmgr.*: parse_affix(): fix fprintf() call when affix
1869 file is corrupt. Bug reported by Daniel Naber.
1870
1871 * suggestmgr.cxx: replace single usage of 'strdup' with 'mystrdup'
1872 patch by Chris Halls (debian.org)
1873
1874 * src/hunspell/makefile.mk: add makefile.mk for compiling in OpenOffice.org
1875 See README in Hunspell UNO modul.
1876 Problems with separated compiling reported by Rene Engelhard
1877
1878 * src/hunspell/hunspell.cxx: fix pseudoroot support
1879 - search a not pseudoroot homonym in check()
1880 * tests/pseudoroot4.*: test this fix
1881
1882 * src/tools/unmunch.c: fix bad word generation when conditions
1883 are shorter or incompatible with stripping characters in affix rules
1884
1885 * src/tools/unmunch.c: fix mychomp() for de_AT.dic and other dic files
1886 without last new line character.
1887
1888 other changes:
1889 * src/hunspell/suggestmgr.*: erase ACCENT suggestion
1890 Rationale: ACCENT suggestion was the same as Kevin Hendrick's map
1891 suggestion algorithm, but with a less good interface in affix file.
1892
1893 * src/hunspell/suggestmgr.*: combine cycle number limit
1894 in badchar(), and forgotchar() with a time limit.
1895
1896 * src/hunspell/affixmgr.*: remove NOMAPSUGS affix parameter
1897
1898 * src/hunspell/{suggestmgr,hunspell}.*: strip periods from
1899 suggestions (restore MySpell's original behaviour)
1900 Rationale: OpenOffice.org has an automatic period handling mechanism
1901 and suggestions look better without periods.
1902 - new affix file parameter: SUGSWITHDOTS
1903 Add period(s) to suggestions, if input word terminates in period(s).
1904 (No need for OpenOffice.org dictionaries.)
1905
1906 * tests/germancompounding.aff: improve bad german affix in affix example
1907 (computeren->computern). Suggested by Daniel Naber.
1908
1909 * src/tools/example.cxx: add Myspell's example
1910
1911 * src/tools/munch.cxx: add Myspell's munch
1912
1913 * man{,/hu}/hunspell.4: refresh manual pages
1914
1915 2005-08-01 Németh László <[email protected]>:
1916 * add missing MySpell files and features:
1917 - add MySpell license.readme, README and CONTRIBUTORS ({license,README,AUTHORS}.myspell)
1918 - add MySpell unmunch program (src/tools/unmunch.c)
1919 - add licenses to source (src/hunspell/license.{myspell,hunspell})
1920 - port MAP suggestion (with imperfect UTF-8 support)
1921 - add NOSPLITSUGS affix parameter
1922 - add NOMAPSUGS affix parameter
1923
1924 * src/man/man.4: MAP, COMPOUNDPERMITFLAG, NOSPLITSUGS, NOMAPSUGS
1925
1926 * src/hunspell/aff{entry,ixmgr}.cxx:
1927 - improve compound word support
1928 - new affix parameter: COMPOUNDPERMITFLAG (see manual)
1929 * src/tests/compoundaffix{,2}.*: examples for COMPOUNDPERMITFLAG
1930 * src/tests/germancompounding.*: new solution for German compounding
1931 Problems with German compounding reported by Daniel Naber
1932
1933 * src/hunspell/hunspell.cxx: fix German uppercase word spelling
1934 with the spellsharps() recursive algorithm.
1935 Default recursive depth is 5 (MAXSHARPS).
1936 * src/tests/germansharps*: extended German sharp s tests
1937
1938 * src/tools/hunspell.cxx: fix fatal memory bug in non-interactive
1939 subshells without HOME environmental variable
1940 Bug detected with PHP by András Izsók.
1941
1942 2005-07-22 Németh László <[email protected]>:
1943 * src/hunspell/csutil.hxx: utf16_u8()
1944 - fix 3-byte UTF-8 character conversion
1945
1946 2005-07-21 Németh László <[email protected]>:
1947 * src/hunspell/csutil.hxx: hunspell_version() for OOo UNO modul
1948
1949 2005-07-19 Németh László <[email protected]>:
1950 * renaming:
1951 - src/morphbase -> src/hunspell
1952 - src/hunspell, src/hunmorph -> src/tools
1953 - src/huntokens -> src/parsers
1954
1955 * src/tools/hunstem.cxx: add stemmer example
1956
1957 2005-07-18 Németh László <[email protected]>:
1958 * configure.ac: --with-ui, --with-readline configure options
1959 * src/hunspell/hunspell.cxx: fix conditional compiling
1960
1961 * src/hunspell/hunspell.cxx: set HunSPELL.bak temporaly file
1962 in the same dictionary with the checked file.
1963
1964 * src/morphbase/morphbase.cxx:
1965
1966 - handling German sharp s (ß)
1967
1968 - fix (temporaly) analyize()
1969
1970 * tests: a lot of new tests
1971
1972 * po/, intl/, m4/: add gettext from GNU hello
1973
1974 * po/hu.po: add Hungarian translation
1975
1976 * doc/, man/: rename doc to man
1977
1978 2005-07-04 Németh László <[email protected]>:
1979 * src/morphbase/hashmgr.cxx: set FLAG attributum instead of FLAG_NUM and FLAG_LONG
1980
1981 * doc/hunspell.4: manual in English
1982
1983 2005-06-30 Németh László <[email protected]>:
1984 * src/morphbase/csutil.cxx: add character tables from csutil.cxx of OOo 1.1.4
1985
1986 * src/morphbase/affentry.cxx: fix Unicode condition checking
1987
1988 * tests/{,utf}compound.*: tests compounding
1989
1990 2005-06-27 Németh László <[email protected]>:
1991 * src/morphbase/*: fix Unicode compound handling
1992
1993 2005-06-23 Halácsy Péter:
1994 * src/hunmorph/hunmorph.cxx: delete spelling error message and suggest_auto() call
1995
1996 2005-06-21 Németh László <[email protected]>:
1997 * src/morphbase: Unicode support
1998 * tests/utf8.*: SET UTF-8 test
1999
2000 * src/morphbase: checking and fixing with Valgrind
2001 Memory handling error reported by Ferenc Szidarovszky
2002
2003 2005-05-26 Németh László <[email protected]>:
2004 * suggestmgr.cxx: fix stemming
2005 * AUTHORS, COPYING, ChangeLog: set CC-LGPL free software license
2006
2007 2004-05-25 Varga Dániel <[email protected]>
2008 * src/stemtool: new subproject
2009
2010 2005-05-25 Halácsy Péter <[email protected]>
2011 * AUTHORS, COPYING: set CC Attribution license
2012
2013 2004-05-23 Varga Dániel <[email protected]>
2014 * src: - modifications for compiling with Visual C++
2015
2016 * src/hunmorph/csutil.cxx: correcting header of flag_qsort(),
2017 * src/hunmorph/*: correct csutil include
2018
2019 2005-05-19 Németh László <[email protected]>
2020 * csutil.cxx: fix loop condition in lineuniq()
2021 bug reported by Viktor Nagy (nagyv nyelvtud hu).
2022
2023 * morphbase.cxx: handle PSEUDOROOT with zero affixes
2024 bug reported by Viktor Nagy (nagyv nyelvtud hu).
2025 * tests/zeroaffix.*: add zeroaffix tests
2026
2027 2005-04-09 Németh László <[email protected]>
2028 * config.h.in: reset with autoheader
2029
2030 * src/hunspell/hunspell.cxx: set version
2031
2032 2005-04-06 Németh László <[email protected]>
2033 * tests: tests
2034
2035 * src/morphbase:
2036 New optional parameters in affix file:
2037 - PSEUDOROOT: for forbidding root with not forbidden suffixed forms.
2038 - COMPOUNDWORDMAX: max. words in compounds (default is no limit)
2039 - COMPOUNDROOT: signs compounds in dictionary for handling special compound rules
2040 - remove COMPOUNDWORD, ONLYROOT
2041
2042 2005-03-21 Németh László <[email protected]>
2043 * src/morphbase/*:
2044 - 2-byte flags, FLAG_NUM, FLAG_LONG
2045 - CIRCUMFIX: signed suffixes and prefixes can only occur together
2046 - ONLYINCOMPOUND for fogemorpheme (Swedish, Danish) or Flute-elements (German)
2047 - COMPOUNDBEGIN: allow signed roots, and roots with signed suffix in begin of compounds
2048 - COMPOUNDMIDDLE: like before, but middle of compounds
2049 - COMPOUNDEND: like before, but end of compounds
2050 - remove COMPOUNDFIRST, COMPOUNDLAST