aboutsummaryrefslogtreecommitdiffstats
path: root/includes/collation
Commit message (Collapse)AuthorAgeFilesLines
* CollationFa: Third time's the charmBartosz Dziewoński2017-01-051-6/+12
| | | | | | | | | | | | | We have to use a tertiary sortkey for everything with the primary sortkey of 2627. Otherwise, the "Remove duplicate prefixes" logic in IcuCollation would remove them. The following characters will now be considered separate letters in the 'xx-uca-fa' collation for the purpose of displaying the headings on category pages: ء ئ ا و ٲ ٳ Bug: T139110 Change-Id: Ibbea5d76348e4cdc38b74cba44286910b2ed592f
* IcuCollation: Add the current class name to 'first-letters' cache keyBartosz Dziewoński2016-12-151-0/+1
| | | | | | | | Instances of subclasses of IcuCollation with customizations for specific languages probably shouldn't share this cache with instances of IcuCollation with the same language. Change-Id: I06d66d199c99448a3375381baef0366c4d99c8c4
* Merge "Add CollationFa"jenkins-bot2016-12-152-0/+46
|\
| * Add CollationFaAmir Sarabadani2016-12-152-0/+46
| | | | | | | | | | Bug: T139110 Change-Id: Ie15a2ee1c22ff4a1d2b721ed137227fe83dd12ea
* | Merge "Make NumericUppercaseCollation use localized digit transforms"jenkins-bot2016-11-162-6/+55
|\ \
| * | Make NumericUppercaseCollation use localized digit transformsBrian Wolff2016-10-292-6/+55
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This will cause the numeric collation to sort localized digits for the current content language the same as how 0-9 are. This only deals with the localized digit numbers, commas and other number formatting are still not handled. Weird "numerical" unicode characters are also not handled. I was unsure if to make a "family" of numeric collations where you specify numeric-<lang code>, or if it should just use $wgContLang. Given that $wgContLang effectively never changes, and also affects all other digit handling, I opted to just use $wgContLang. Any wikis currently using the 'numeric' collation will have to have updateCollation.php --force run after this change is deployed. At the moment that includes: bnwiki, bnwikisource and hewiki Bug: T148873 Change-Id: I9eda52a8a9752a91134d1118546b0a80d3980ccf
* | | Add first letter data for bn collation (Standard and Traditional)Brian Wolff2016-11-151-0/+6
| | | | | | | | | | | | | | | | | | | | | | | | This is based solely on looking at the bn.txt collation data file. It has not been tested by native speakers. Bug: T148885 Change-Id: Ide926bc5ee8752269ef6a1bfe972e19b7188d193
* | | IcuCollation: Do not split $tailoringFirstLetters into verified/not verifiedBartosz Dziewoński2016-10-311-51/+48
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | At this point I think it's safe to assume that these mostly work well, and the split makes maintenance of the alphabetical list more difficult (some entries were already in wrong order). We've been enabling these collations for more and more Wikimedia wikis and not hearing about any problems. Mistakes, if any are present, should be treated like any other bug. Also made some comments consistent. Change-Id: I4b5fbcf4dbbdd4dc194ed821341296171fa64bb0
* | | Add firstLetter data for ~50 additional languagesBrian Wolff2016-10-291-15/+73
|/ / | | | | | | | | | | | | | | | | Based on CLDR 29 data files. This did the relatively easy languages in CLDR 29 (Which is most of them). I skipped languages with complicated tailoring files. Change-Id: I8367604f7d3a1cdef9cb4e15813893c8cbfff1ff
* | Fixing numeric sorting for numbers with leading zerosKaldari2016-10-201-2/+4
| | | | | | | | | | Bug: T148774 Change-Id: I34aa330645d9d82b6c4e57542e891dd2b36e42ad
* | IcuCollation: Update comments on $tailoringFirstLettersBartosz Dziewoński2016-09-221-15/+15
| | | | | | | | | | | | | | | | | | | | | | A few more languages marked as "Verified by native speakers", based on which collations we've been using in production on Wikimedia wikis. (I'm not sure if this makes sense now that we're fairly confident that these are good in general, but since it's already here...) Change-Id: I8e1f31fa61509eca8c76a2df4e18638005e68b77
* | Implement NumericUppercaseCollationBartosz Dziewoński2016-08-232-0/+60
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This collation orders text with numbers "naturally", so that 'Foo 1' < 'Foo 2' < 'Foo 12'. Note that this only works in terms of sequences of digits, and the behavior for decimal fractions or pretty-formatted numbers may be unexpected. This is only expected to work mostly correctly for English-language text. Consider it a proof of concept. You probably want to use an UCA collation with '-u-kn' suffix rather than this. Bug: T8948 Change-Id: Ie268f2d92c5c75d0aaecf54ede2bdda1af3b309d
* | Updating $tailoringFirstLetters for MacedonianKaldari2016-08-081-1/+1
|/ | | | | | | Per https://ssl.icu-project.org/trac/browser/icu/trunk/source/data/coll/mk.txt Bug: T26953 Change-Id: I45938402923a109cfc80f59555af5cede584fc3b
* Adding support for numeric collation when using UCA collationsKaldari2016-07-262-2/+30
| | | | | | | | To use, add '-u-kn' to the end of a collation name and set it as the value for $wgCategoryCollation. Bug: T8948 Change-Id: Ica7908daf80624fa2648127114d01665e96234c0
* Merge "Add Unicode to ICU mappings for versions 51-57"jenkins-bot2016-07-211-0/+7
|\
| * Add Unicode to ICU mappings for versions 51-57Reedy2016-07-201-0/+7
| | | | | | | | Change-Id: I35c2cdd2c56b491229f1f6d8b69b1de21af23aab
* | Do not collate "ا" and "و" in Persian language for category sortAmir Sarabadani2016-07-201-1/+8
|/ | | | | Bug: T139110 Change-Id: If174e02160c954500233e3a57945e267f2b4ae29
* Merge "Add support for icu-ta collation"jenkins-bot2016-05-301-0/+5
|\
| * Add support for icu-ta collationBrian Wolff2016-05-251-0/+5
| | | | | | | | | | | | | | | | | | | | First letters are based on https://ssl.icu-project.org/trac/browser/icu/trunk/source/data/coll/ta.txt This commit has not been verified by a native speaker yet, but is probably right. Bug: T75453 Change-Id: Ic9bb3658868917790aa770c99f8f280f2dd3eace
* | collation: Remove suppressWarnings() for getSortKey()Ori Livneh2016-05-261-11/+2
|/ | | | | | | | | | | Small optimization to IcuCollation::fetchFirstLetterData(). This used to suppress / restore warnings once per every letter of every alphabet. The workaround for string casting and error suppression is no longer needed as of PHP 5.3, in which the bug was fixed. Change-Id: Idd41a509858c0887df4f632b480b387bd74027b2
* collation: Refactor getFirstLetterData() cache handlingTimo Tijhof2016-05-091-36/+24
| | | | | | | | | * Factor out fetchFirstLetterData() as a separate method. * Move 'version' into the key instead of checking afterwards. * Use getWithSetCallback() for the cache handling. (Depends on version being in the key). Change-Id: I15bddf5d1dabcdcef47a938447ba59436bd8a294
* Cache first-letter data in APC, if availableOri Livneh2016-04-111-2/+2
| | | | | | | | | | | I noticed that `frwiki:first-letters:fr:fr:4.8.1.1` was at the very top of keys sorted by bandwidth (that is, reqs/sec * size) on one of the memcache servers on WMF prod. The data takes ~60 - 80ms to compute, in case of a cache miss. That's not enough to justify using a tiered cache abstraction here, IMO. Change-Id: If81ce8f86f2c378565f1f6a0dd2c04dee825c4e9
* Add @since tags to Collation stuffaddshore2016-04-036-7/+46
| | | | Change-Id: Iec56ac4d1418737d171f8faa9c8f498fba5383ee
* Split Collation.phpaddshore2016-04-036-0/+746
Change-Id: I6abfecf91cdce83dd34b1e8aa8e0b35315f62742
generated by cgit v1.2.3 (git 2.25.1) at 2025-04-24 11:13:29 +0000