Debian Codegraph @

ChangeLog

raw
1 libzim 9.7.0
2 ============
3
4 * Fix handling of "bad" redirections (@veloman-yunkan #1065 #1077)
5 * Fix handling of ZIM file chunks (@veloman-yunkan #1078)
6 * Made zim::writer::creator::addMetadata() exception-safe (@veloman-yunkan #1070)
7 * Fix a rare (edge-case) and old bug in ZIM reader (@veloman-yunkan #1066)
8 * Enable better memory mgmt on macOS in ZIM reader (@veloman-yunkan #1067)
9 * Further optimization of redirection memory usage for both writer and reader (@veloman-yunkan #1068)
10
11 libzim 9.6.0
12 ============
13
14 * Optimization of memory usage by dirents during ZIM creation (@veloman-yunkan #1055)
15 * Allows to open ZIM with more than ~268M entries on macOS (@jasontitus #1053)
16 * Fix ZIM metadata 'Counter' parser by properly ignoring mime-types paramters (@Pranavjeet-Naidu #1050)
17
18 libzim 9.5.1
19 ============
20
21 * Eliminated zim::Compressor trapped in an infinite loop scenario (@veloman-yunkan #1047)
22 * Use windows-2025 CI runner (@kelson42 #1041)
23
24 libzim 9.5.0
25 ============
26
27 * Handling Protection against too long words in titles (@veloman-yunkan #1004 #1034)
28 * Quick detection of corruption of 1st blob offset in cluster( @veloman-yunkan #1029 #1031)
29 * Use zim-testing-suite version 0.9.0 (@veloman-yunkan #1030)
30
31 libzim 9.4.1
32 ============
33
34 * Remove SORTPP_PASS which breaks modern Windows SDKs (@BillyONeal #1017)
35 * Add support of future Xapian 2.0 (@ojwb #1018 #1019 #1020)
36 * Fix broken CI for linux-aarch64-dyn (@veloman-yunkan #1024)
37
38 libzim 9.4.0
39 ============
40
41 * Add support of Illustrations scale (dpr), height and width (@veloman-yunkan #997)
42 * Many CI updates (@kelson42 #976 #995 #980, @veloman-yunkan #974 #987, @tusharpm #910)
43 * Improve logging primitites (@veloman-yunkan #975 #984)
44 * New ability to preload Xapian indexes (@mgautierfr #958)
45 * New user/dev defined cache sizes (@veloman-yunkan #982)
46 * Centralize cache memory usage (@veloman-yunkan #960)
47 * Stop releasing for PPA Ubuntu Focal 20.04 (@kelson42 #1009)
48 * Fix build with ICU 76+ (@cho-m #1010)
49
50 libzim 9.3.0
51 ============
52
53 * Make the cache configurable by cpp API (@mgautierfr #950)
54 * Remove old ZIM (v0) titleindex (title pointer list) (@mgautierfr #949, @veloman-yunkan #969)
55 * Fixed ZIM title indexing of items with title identical to the item's path (@veloman-yunkan #968)
56 * Fixed ZIM suggestions for a single-word-followed-by-space query (@veloman-yunkan #953)
57 * Testing based on latest zim-testing-suite 0.8.0 (@veloman-yunkan #971)
58 * Fix compilation on NetBSD because MAP_POPULATE & pread64() undefined (@leleliu008 #928 #929)
59 * Upload GitHub workflows (@kelson42 #940 #942 #970)
60
61 libzim 9.2.3
62 ============
63
64 * Correctly locate part range in which read data in case of ZIM chunks (@mgautierfr #903)
65 * Fix test compilation GCC12 #899 (#mgautierfr #899)
66 * Move Windows CI/CD to native MS Windows via GitHub Actions (@mgautierfr #899 #914)
67 * Update deb package CI/CD for Debian & Ubuntu with latest releases (@kelson42 #905 #906 #922)
68
69 libzim 9.2.2
70 ============
71
72 * Avoid crash scenario in case of invalid offset in clusters (@mgautierfr #895)
73 * Many improvements around testing (@mgautierfr #889 #897)
74 * Windows CI using GitHub Windows runner (@mgautierfr #894)
75 * Better error message when failin to open (split) ZIM file (@mgautierfr #884)
76
77 libzim 9.2.1
78 ============
79
80 * Better handling of split ZIM files (@mgautierfr #879)
81 * Fix creation of shared_ptr<char> in test (@mgautierfr #881)
82
83 libzim 9.2.0
84 ============
85
86 * Allow open Archive with a set of (positionned) file descriptors (@mgautierfr #860)
87 * Introduce new (private) method `getEntryByPathWithNamespace` (@mgautierfr #859)
88 * Internally catch xapian exception and rethrow ZimFileFormatError instead (@mgautierfr #873)
89 * Fix compilation on Haiku (@Begasus #857)
90 * Fix macos mmap (@mgautierfr #867)
91 * Optimize checksum calculation (@aryanA101a #861)
92 * Introduce `Formatter` helper (@ShaopengLin #862)
93 * Rename all `*Url*` symbols to `*Path*` (@mgautierfr #869)
94 * Build script: Allow to disable compilation of test (@kelson42 #854)
95 * [CI] Use kiwix-build github's action to download deps archive (@mgautierfr #850)
96 * [CI] Build libzim on macos-14 (@kelson42 #856)
97
98 libzim 9.1.0
99 ============
100
101 * New addAlias() method in the creator (@mgautierfr #833)
102 * Bump-up ZIM format minor version to 6.1.2 (@mgautierfr #847)
103
104 libzim 9.0.0
105 ============
106
107 * getMediaCount() does not fail anymore if M/Counter is missing (@mgautierfr #827)
108 * Reintroduce optimization of Entry::getItem() (@kelson42 @mgautierfr #836)
109 * C++17 compatible code (@mgautierfr #819)
110 * Add support to recent googletest framework (@kelson42 #830)
111 * Multiple fixes for Apple macOS/iOS compilation & CI (@mgautierfr @kelson42 @rgaudin #832 #839)
112
113 libzim 8.2.1
114 ============
115
116 * Better indexer CJK content (@mgautierfr #806)
117 * Fix export of symbol on Windows dll (@xiaoyifang @mgautierfr #796 #807)
118 * Remove accents from the search query (@mgautierfr #804)
119 * Improved performance of removing of accents when working on long string (@mgautierfr #797)
120 * Various improvement and fixes of the CI (@mgautierfr @kelson42)
121
122 libzim 8.2.0
123 ============
124
125 * Deprecate `SearchIterator::getSize()` method (@mgautierfr #774)
126 * Fix handling of search end iterator (@mgautierfr #774)
127 There were cases when we could dereference a end iterator.
128 * Fix suggestions with titles containing punctuations (@veloman-yunkan #765)
129 * Correctly publish our public API in Windows's dll (@xiaoyifang #783)
130 * Fix various warning and compilation error when compiling with last xcode version (@mgautierfr #782)
131 * Fix faulty unit-test checking for async errors (@mgautierfr #776)
132 * Update subproject wrap zstd to version 1.5.4 (and use upstream wrap file.) (@mgautierfr #749)
133 * Move main branch from `master` to `main`
134 * Add CI to build on aarch64 (@mgautierfr #784)
135 * Various CI improvement (@kelson42)
136
137 libzim 8.1.1
138 ============
139
140 * Revert a ABI breaking change introduced in 8.1.0 (Optimization of `Entry::getItem()`)
141
142 libzim 8.1.0
143 ============
144
145 * Optimization ofthe first call to `zim::Archive::iterEfficient` (@veloman-yunkan #724)
146 * Add some documentation to `zim::writer::IndexData` (@mgautierfr #727)
147 * Correctly catch and rethrow exception thrown in worker threads at creation (@mgautierfr #496 #748)
148 * Optimization of `Entry::getItem()` (@veloman-yunkan #732)
149 * Fix declaration of `zim::setICUDataDirectory()` (@MohitMaliFtechiz #733)
150 * Add `zim::Archive::getMediatCount()` (@mgautierfr #730)
151 * Make compilaton of examples optional (@mgautierfr #738)
152 * Add a CI for wasm (@mgautierfr #746)
153 * Make constructor of SuggestionItem public (@veloman-yunkan #740)
154
155 libzim 8.0.1
156 ============
157
158 * Fix debian packagin (@mgautierfr)
159
160 libzim 8.0.0
161 ============
162
163 * [API-BREAK] Remove lzma compression support in writer (@veloman-yunkan #718)
164 * Add new method `zim::Entry::getRedirectEntryIndex()` (@veloman-yunkan #716)
165 * Add new helper function `zim::setICUDataDirectory()` to help android wrapper
166 compilation (@mgautierfr #722)
167 * Fix `std::call_once` usage (alpine bug) (@veloman-yunkan #ê708)
168 * Better xapian indexation (no transaction, better compact algorithm) (@mgautierfr #719)
169 * Reserve more space (1968B instead of 944B) for mimetype list (@mgautierfr #720)
170 * [CI] Fix android compilation in the CI (@veloman-yunkan @mgautierfr #713)
171 * [CI] Add CI for Alpine (@veloman-yunkan #710)
172 * [CI] Support checkout of tag in the CI (@teeks99 #696)
173 * [CI] Remove movebot (@kelson42 704)
174 * [CI] Remove Impish and add Kinetic packages (@legoktm #715)
175 * Fix code factor report (@kelson42 #700)
176 * Fix readme (@kelson42 #701 #716)
177
178 libzim 7.2.2
179 ============
180
181 * Change the way we generate search result snippet.
182 We now ask xapian to generate "less" relevant snippet (even if in practice,
183 snippets are still good). But it know generate snippet far more quicker.
184 On cold search, no cache and low IO, search can go from 90s to 3s.
185 (@mgautier #697)
186 * [CI] Update base images (@mgautier #695)
187
188 libzim 7.2.1
189 ============
190
191 * Make suggestions diacritics insensitive (@veloman-yunkan #691)
192 * [Writer] Raise an exception when user add a invalid entry (duplicate path)
193 instead of printing a message (which can be too easily missed) and be buggy
194 (@mautierfr #690)
195 * [Writer] Do not `hasIndexData` and `getTitle` in the main thread when we add
196 an entry (@mgautier #684)
197 * [Writer] Properly clean and stop the writer even if user hasn't call
198 `finishZimCreation` (The created zim file is still invalid) (@veloman-yunkan #666)
199 * Add a default argument value for mimetype of `creator::addMetadata`
200 (@kelvinhammond #678)
201 * Use a more informative message in exception when we cannot open a file
202 (@veloman-yunkan #667 #668)
203 * Use a generic dirent lookup to search by title (@veloman-yunkan #651)
204 * Various improvements:
205 - CI, Packaging : Stop creating packages for Ubuntu Hirsute (@legoktm #664)
206 - Update Readme (@TheDuchy #660)
207 - Fix cross-compilation host machine detection (@kelson42 #665)
208 - Fix macos/ios compilation (@mgautierfr #672)
209 - Update documentation @mgautierfr #677, @veloman-yunkan #682
210
211
212 libzim 7.2.0
213 ============
214
215 * Add methods to get/print (dependences) versions (@kelson42, #452)
216 * Fix Emscripten compilation (@kelson42, @mossroy, #643)
217
218 libzim 7.1.0
219 ============
220
221 * Fix dirent test on 32 bits architectures (@mgautierfr #632)
222 * Fix compilation on Alpine - with musl (@amirouche #649)
223 * Don't crash if ZIM without illustration nor X/W namespace (@mgautierfr #641)
224 * Switch default suggestion operator to AND (@maneeshpm #644)
225 * Add a new method Archive::getMetadataItem (@mgautierfr #639)
226 * Better indexion criterias (@mgautierfr #642)
227 * Avoid duplicated archives in the searcher (@veloman-yunkan #648)
228 * Fix random entry (@veloman-yunkan #650)
229 * Various improvements.
230 - CI @mgautierfr #640, @kelson42 #638, @legoktm #654
231 - Doc @rgaudin #646
232
233 libzim 7.0.0
234 ============
235
236 Version 7.0.0 is a major release.
237
238 The API has been completely rewritten.
239 Most notable change is that namespaces are now hidden.
240 The new API is described in documentation, which includes a Transition Guide from v6.
241
242 ZIM files created with it uses new ZIM minor version (6.1 - see Header section of spec.)
243 Both backward and forward compatibility is kept.
244
245 Improvements
246 ------------
247
248 * Rewrite creator and reader API
249 This removes the namespace from the API. Article are automatically put in
250 the right namespace ('A') and the retrivial of content is made using
251 specific API. (@mgautier #454)
252 * Better handling of the conditional compilation without xapian.
253 Before that, the search API was present (but returning empty result) if
254 libzim was compiled without xapian. Now the API is not present anymore.
255 User code must check if libzim is compiled with xapian or not by checking
256 if LIBZIM_WITH_XAPIAN is defined or not. (@mgautierfr #465)
257 * Add a new specific listing in zim files to list entries considered as "front
258 article". At creation, wrapper MUST pass the hint `FRONT_ARTICLE` to
259 correctly mark the entry. Search by title uses this list if present.
260 (@mgautierfr #487)
261 * Store the wellknown entries in the `W` namespace (`W/mainPage`)
262 (@mgautierfr #497)
263 * Rewrite Search API. Fix potential memory link and allow correct reusing of
264 create search. (@mgautierfr #530)
265 * New suggestion search API. The api mimics the Search API but specialized
266 for suggestion (@maneeshpm #574)
267 * Add `zim::Archive` constructors to open an archive using a existing file descriptor.
268 This API is not available on Windows. (@veloman-yunkan #449)
269 * Make zstd the default compression algorithm (@veloman-yunkan #480)
270 * The method `zim::Archive::checkIntegrity` now if the mimetypes indicated in the
271 dirents are correct (@veloman-yunkan #505)
272 * Writer doesn't add a `.zim` extension to the given path. (@maneeshpm #503)
273 * Implement random entry picking. We are choosing a entry from the "front
274 article" list if present. (@mgautierfr #476)
275 * Creator now create the `M/Counter` metadata. (@mgautierfr)
276 * Better Illustration handling. Favicon is replaced by Illustration.
277 Illustration can now have different size and scale (even if the API do
278 not use this feature) (@mgautierfr #540)
279 * Search iterator now have a method `getZimId` to know the Id of the zim
280 corresponding to the result (useful for multizim search) (@maneeshpm #557)
281
282 Bug fixes
283 ---------
284
285 * The method `zim::Archive::checkIntegrity` now check if the dirents are
286 correctly sorted. (@veloman-yunkan #448)
287 * Handle large MIME-type list. Some zim file may have a pretty large mimetype
288 list. (@veloman-yunkan #460)
289 * Fix handling of zim file containing item of size 0. (@mgautierfr #483)
290 * Better parsing of the entry paths to detect the namespace (@maneeshpm #479)
291 * Fix zim file creation on Windows (@mgautierfr #508)
292 * Better algorithm tunning for suggestion search (@maneeshpm #492)
293 * The default indexer now index html content only. (@mgautierfr #511)
294 * Better suggestion search : Don't use stopwords, use OP_PHRASE
295 (@maneeshpm #501)
296 * Remove duplicate in the suggestion search (@maneeshpm #515)
297 * Remove the termlist from the xapian database, lower memory usage
298 (@maneeshpm #528)
299 * Add a anchor in the suggestion search to search term at the beginning of
300 the title (@maneeshpm #526)
301 * Make the suggestion search working with special characters (`&`, `+`)
302 (@veloman-yunkan #534)
303 * Fix creator issue not detecting that cluster must be extended if it
304 contains only 32-bit-sized content. (@veloman-yunkan #552)
305 * Correctly generate suggestion snippet. (@maneeshpm #545)
306 * Better cluster size configuration (@mgautierfr #555)
307 * Make search iterator `getTitle` return the real title of the entry and not
308 the one stored in the xapian database (caseless) (@maneeshpm #586)
309 * Correcly close a zim creator to avoid a crash when the creator is
310 destructed without being started (@mgautierfr #613)
311 * Reduce the creator memory usage by reducing the memory size of the dirent
312 (@mgautier #616, #628)
313 * Write the cluster using a bigger chunk size for performance
314 (@mgautierfr #506)
315 * Change the default cluster size to 2MiB (@mgautierfr #585)
316 * The default mimetype for metadata now include the utf8 chardet
317 (@rgaudin #626)
318 * Improve the estimation of the number of search/suggestion results by forcing
319 Xapian to evaluate at least 10 results (@mgautier #625)
320
321 Other
322 -----
323
324 * Update xapian stopwords list. (@data-man #447)
325 * Remove direct pthread dependency (use c++11 thread library). (@mgautierfr #443)
326 We still need pthread library on linux and freebsd as C++11 is using it internally.
327 * [CI] Make the libzim CI compile libzim natively on Windows (@mgautierfr #453).
328 * [CI] Build libzim package for Ubuntu Hirsute and Impish
329 (@legoktm #459, #580)
330 * Always create zim file using the major version 6. (@mgautierfr #512)
331 * Move the test data files out of the git repository. Now test files are
332 stored in `zim-testing-suite` repository and must be downloaded.
333 (@mgautierfr #538, #535)
334 * Add search iterator unit test (@maneeshpm #547)
335 * Correctly fix search iterator method case to use camelCase everywhere
336 (@maneeshpm #563)
337 * Add a cast to string opertor on Uuid (@maneeshpm #582)
338 * Make unittest print the path of the missing zim file when something goes
339 wrong (@kelson42 #601)
340 * Delete temporary data (index) after we called `finishZimCreation` instead of
341 waiting for creator destruction. (@mgautierfr #603)
342 * Add basic user documentation (@mgautierfr #611)
343
344 Known bugs
345 ----------
346
347 Suggestion system using in current libkiwix doesn't work with new zim files
348 created with this release (and future ones).
349 New libkiwix version will be fixed and will work with new and old zim files.
350
351
352 libzim 6.3.2
353 ============
354
355 This is a hotfix of 6.3.0 :
356 * libzim now create zimfile with zstd compression 19 instead of 22.
357 So new libzim do not need to allocate 128Mb per cluster at decompression
358 time.
359 * At reading time, on 32 bits architectures, zstd cluster are not keep in
360 cache. This avoid use to also keep the decompression stream which reserve
361 128Mb of memory address.
362
363 libzim 6.3.1
364 ============
365
366 The release process of 6.3.1 was buggy. So, no 6.3.1.
367
368
369 libzim 6.3.0
370 ============
371
372 * Rewrite internal reader structure to use stream decompression.
373 This allow libzim to not decompresse the whole cluster to get an article
374 content. This is big performance improvement, it speedups random access by
375 2, with a very small cost when doing "full" incremental reading
376 (zim-check/zim-dump). (@veloman-yunkan)
377 * Better dirent lookup.
378 Dirent lookup is the process of locating article data starting from the url
379 or title. This improves reading of zim file up to 10% (@veloman-yunkan)
380 * Add basic, first version of `validate` function to check internal structure
381 of a zim file. (@veloman-yunkan, @MiguelRocha)
382 * Fix compilation of libzim without xapian (@mgautierfr)
383 * Remove zlib dependency (and support of very old files created using zlib
384 compression) (@mgautierfr)
385 * New unit tests and various small fixes.
386
387
388 libzim 6.2.2
389 ============
390
391 * Check blob index before access it in the cluster.
392 * Refactoring of the cluster reading.
393
394 libzim 6.2.1 (release process broken)
395 =====================================
396
397 * Update readme and add link to repology.org packages list.
398 * Fix compilation on windows.
399
400 libzim 6.2.0
401 ============
402
403 * Fix compilation of libzim on freebsd.
404 * Rewrite unit tests to remove python based test and use gtest all the time.
405 * Make libzstd mandatory.
406 * Support for meson 0.45.
407 * Fix multipart support on macos.
408 * Add a documentation system.
409 * Better cache system implementation (huge speed up).
410 * Various (and numerous) small refactoring.
411
412
413 libzim 6.1.8
414 ============
415
416 * Increase default timeout for test to 120 seconds/test
417 * Compression algorithm to use can be passed to `zim::writer::Creator`
418 * Add automatic debian packaging of libzim.
419 * Fix using of tmpdir (and now use env var TMPDIR) during tests.
420
421
422 libzim 6.1.7
423 ============
424
425 * Do not assume urlPtrPos is just after the mimetype list.
426 * Fix compilation of compression test.
427 * Do not exit but throw an exception if an ASSERT is not fulfill.
428
429 libzim 6.1.6
430 ============
431
432 * Better (faster) implementation of the ordering of article by cluster.
433 * Fix compression algorithm.
434
435 libzim 6.1.5
436 ============
437
438 * [Writer] Remove unused declaration of classes.
439 Those classes were not implemented nor used at all.
440
441 libzim 6.1.4
442 ============
443
444 * [Writer] Fix excessive memory usage. Data of the cluster were clean at the
445 end of the process, not once we don't need it.
446
447 libzim 6.1.3
448 ============
449
450 * [Writer] Use a `.tmp` suffix and rename to `.zim` at the end of the write
451 proces.
452 * Add unit tests
453 * Do not include uncessary `windows.h` headers in public zim's headers.
454
455 libzim 6.1.2
456 ============
457
458 * [CI] Fix codecov configuration
459 * [Writer] Fix threads synchronization at end of writing process.
460
461 libzim 6.1.1
462 ============
463
464 * Fix bug around the find function
465
466 libzim 6.1.0
467 ============
468
469 * Compile now on OpenBSD
470 * [Test] Use the main function provided by gtest.
471 * [CI] Move the CI compilation to github actions.
472 * Add stopwords for 54 new languages.
473 * [Writer] Improve the way we are writing cluster at zim creation time.
474 - Clusters are directly written in the zim file instead of using temporary
475 files.
476 - mimetypes are limited to 944 bytes.
477 * Add a new type of iterator to iterate over articles in a performant way
478 reducing decompression of clusters. This is now the new default iterator.
479 * Add support for zim files compressed with zstd compression algorithm.
480 This is not possible to use zstd to create zim file for now.
481
482 libzim 6.0.2
483 ============
484
485 * Fix search suggestion parsing.
486
487 libzim 6.0.1
488 ============
489
490 * Fix crash when trying to open an empty file.
491 * Ensure that pytest tests are run on the CI.
492
493 libzim 6.0.0
494 ============
495
496 * [Writer] Index the articles in differents threads. This is a huge speed
497 improvement as the main thread in not blocked by indexing.
498 * Index the title only if `shouldIndex` return true.
499
500 libzim 5.1.0
501 ============
502
503 * Improve indexation of the title.
504 * Better pertinence of suggestions (only for new zim files)
505 * Improvement of the speed of Leveinstein distance for suggestions (for old
506 zims)
507
508 libzim 5.0.2
509 ============
510
511 * Improve README.
512 * Remove gtest as embeded subproject.
513 * Better lzma compression.
514 * Better performance of the leveinstein algorithm (better suggestions
515 performance)
516
517 libzim 5.0.1
518 ============
519
520 * Update README.
521 * [Writer] Add debug information (print progress of the clusters writing).
522 * [Writer] Correctly print the url to the user.
523 * [CI] Add code coverage.
524
525 libzim 5.0.0
526 ============
527
528 * Fix thread slipping for win32 crosscompilation.
529 * Fix a potential invalid access when reading dirent.
530 * Fix memory leak in the decompression algorithm.
531 * [Writer] Fix a memory leak (cluster cleanning)
532 * [Writer] Write article data in a temporary cluster file instead of a
533 temporary file per article.
534 * [Writer] Better algorithm to store the dirent while creating the zim
535 file. Better memory usage.
536 * [Writer] [API Change] Url/Ns are now handle using the same struct Url.
537 * [Writer] [API Change] No more aid and redirectAid. A redirectArticle
538 have to implement redirectUrl.
539 * [Writer] Use a memory pool to avoid multiple small memory allocations.
540 * [Writer] [API Change] Rename `ZimCreator` to `Creator`.
541 * [API Change] File's `search` and `suggestions` now return a unique_ptr
542 instead of a raw pointer.
543
544 libzim 4.0.7
545 ============
546
547 * Build libzim without rpath.
548
549 libzim 4.0.6
550 ============
551
552 * Support zim file created with cluster not written sequentially.
553 * Remove a meson warning.
554
555 libzim 4.0.5
556 ============
557
558 * Store the xapian database in the right url.
559 * Do not fail when reading very small zim file (<256b).
560 * Do not print message on normal behavior.
561 * [BUILDSYSTEM] Be able to build a dynamic lib (libzim.so) but using static
562 dependencies.
563 * [CI] Use last version of meson.
564 * [CI] Use the new deps archive xz
565
566 libzim 4.0.4
567 ============
568
569 * Fix opening of multi-part zim.
570 * Fix convertion of path to wpath on Windows.
571
572 libzim 4.0.3
573 ============
574
575 * Implement low level file manipilation using different backends
576
577 libzim 4.0.2
578 ============
579
580 * [Windows] Fix opening of zim file bigger than 4GiB
581
582 libzim 4.0.1
583 ============
584
585 * [Writer] Fix wrong redirectyon log message
586 * Make libzim compile natively on windows using MSVC
587 * Better message when failing to read a zim file.
588 * Make libzim on windows correctly open unicode path.
589 * Add compilation option to use less memory (but more I/O).
590 Usefull on low memory devices (android)
591 * Small fixes
592
593 libzim 4.0.0
594 ============
595
596 * [Writer] Remove a lot of memory copy.
597 * [Writer] Add xapian indexing directly in libzim.
598 * [Writer] Better API.
599 * [Writer] Use multi-threading to write clusters.
600 * [Writer] Ensure mimetype of articles article is not null.
601 * Extend test timeout for cluster's test.
602 * Less memory copy for cluster's test.
603 * Allow skipping test using a lot memory using env variable
604 `SKIP_BIG_MEMORY_TEST=1`
605 * Explicitly use the icu namespace to allow using of packaged icu lib.
606 * Use a temporary file name as long as the ZIM writting process is
607 not finished (#163)
608 * [Travis] Do no compile using gcc-5 (but the default trusty's one 4.8)
609
610 libzim 3.3.0
611 ============
612
613 * Fix handling of big cluster (>4GiB) on 32 bits architecture. This is mainly
614 done by :
615 * Do not mmap the whole cluster by default.
616 * MMap only the memory asociated to an article.
617 * If an article is > 4GiB, the blob associated to it is invalid
618 (data==size==0).
619 * Other information are still valid (directAccessInformation, ...)
620 * Fix writing of extended cluster in writer.
621 * Compile libzim on macos.
622 * Build libzim setting RPATH.
623 * Search result urls are now what is stored in the zim file. They should not
624 start with a `/`. This is a revert of the change made in last release.
625 (See kiwix/kiwix-lib#123)
626 * Spelling corrections in README.
627
628 libzim 3.2.0
629 ============
630
631 * Support geo query if the xapian database has indexed localisation.
632 * Handle articles bigger than 4Go in the zim file (#110).
633 * Use AND operator between search term.
634 * Fix compilation with recent clang (#95).
635 * Add method to get article's data localisation in the zim file.
636 * Be able to get only a part of article (#77).
637 * Do not crash if we cannot open the xapian Database for some reasons.
638 (kiwix/kiwix-tools#153)
639 * Do not assumen there is always a checksum in the zim file.
640 (kiwix/kiwix-tools#150)
641 * Try to do some sanity checks when opening a zim file.
642 * Use pytest to do some tests (when cython is available).
643 * Use levenshtein distance to sort and have better suggestion results.
644 * Search result urls are now always absolute (starts with a '/').
645 (kiwix/kiwix-lib#110)
646 * Open the file readonly when checking the zim file (and so be able to check
647 read only file).
648 * Accept absolute url starting with '/' when searching for article.
649 * Fix various bugs
650
651 libzim 3.1.0
652 ============
653
654 * Lzma is not a optional dependency anymore.
655 * Better handle (report and not crash) invalid zim file.
656 * Embed source of gtest (used only if gtest is not available on the system)
657 * Move zimDump tools out of libzim repository to zim-tools
658 * ZimCreator tools doesn't not read command line to set options.
659
660 libzim 3.0.0
661 ============
662
663 This is a major change of the libzim.
664 Expect a lot new improvement and API changes.
665
666 * Add a suggestion mode to the search
667 * Fix licensing issues
668 * Fix wrong stemming of the query when searching
669 * Deactivate searching (and so crash) in the embedded database if the zim is
670 splitted
671 * Rewrite the low level memory management of libzim when reading a zim file:
672 * We use a buffer base entity to handle memory and reading file instead of
673 reading file using stream.
674 * MMap the memory when posible to avoid memory copy.
675 * Use const when posible (API break)
676 * Move to googletest instead of cxxtools for unit-tests.
677 * Fix endiannes bug on arm.
678 * Do not install private headers. Those headers declare private structure and
679 should not be visible (API break)
680 * Compile libzim with `-Werror` and `-Wall` options.
681 * Make libzim thread safe for reading article.
682 The search part is not thread safe, and all search operation must be
683 protected by a lock.
684 * Add method to get only a part of a article.
685 * Move some tools to zim-tools repository.
686
687
688 libzim 2.0.0
689 ============
690
691 * Move to meson build system
692 `libzim` now use `meson` as build system instead of `autotools`
693 * Move to C++11 standard.
694 * Fulltext search in zim file.
695 We have integrated the xapian fulltext search in libzim.
696 So now, libzim provide an API to search in a zim containing embeded fulltext
697 index. This means that :
698 *libzim need xapian as (optional) dependencies (if you want compile with
699 xapian support).
700 * The old and unused search API has been removed.
701 * Remove bzip2 support.
702 * Remove Symbian support.
703 * Few API hanges
704 * Make some header files private (not installed);
705 * A `Blob` can now be cast to a `string` directly;
706 * Change a lot of `File` methods to const methods.