Since we now take language into account when formatting times the one remaining part of the old formatter now has a bug. Specifically the two regexes in MwTimeIsoFormatter::getDateFormat().
As we not use more languages we also have to take into account more tokens to remove for months and days!
Version: unspecified
Severity: major
Whiteboard: u=dev c=backend p=20 s=2014-04-23
| Status | Subtype | Assigned | Task | ||
|---|---|---|---|---|---|
| Resolved | thiemowmde | T65723 Time formatting incorrect for some languages | |||
| Resolved | None | T65733 Dates formatted as timestamp when lang set to Arabic, Farsi, etc. | |||
| Resolved | aude | T66347 Language::parseFormattedNumber does not work for lzh and zh-classical |
Change 124823 had a related patch set uploaded by Addshore:
Correctly format dates for lang=nl
Affected Langs:
$brokenLanguages = array( 'ab', 'als', 'anp', 'ar', 'arq', 'arz', 'as', 'av', 'azb', 'ba', 'bar', 'bcc', 'be', 'be-tarask', 'be-x-old', 'bh', 'bho', 'bn', 'bo', 'bpy', 'bqi', 'bxr', 'cdo', 'ce', 'ckb', 'crh', 'crh-latn', 'crh-cyrl', 'cs', 'cu', 'cv', 'da', 'de', 'de-at', 'de-ch', 'de-formal', 'dsb', 'dz', 'el', 'eo', 'et', 'fa', 'fi', 'fit', 'fiu-vro', 'fo', 'frr', 'fur', 'gan', 'gan-hans', 'gan-hant', 'gl', 'glk', 'grc', 'gsw', 'gu', 'he', ' hi', ' hr', ' hrx', ' hsb', ' hu', ' hy', 'ii', ' inh', ' is', 'ja', ' jut', 'kaa', 'kk', 'kk-arab', 'kk-cyrl', 'kk-latn', 'kk-cn', 'kk-kz', 'kk-tr', 'kl', 'km', 'kn', 'ko', 'ko-kp', 'koi', 'krc', 'ks', 'ks-arab', 'ks-deva', 'ksh', 'ku-arab', 'kv', 'la', 'lb', 'lbe', 'lez', 'liv', 'lo', 'lzh', 'mai', 'mhr', 'mr', 'mrj', 'mwl', 'my', 'myv', 'mzn', 'nan', 'nb', 'nds', 'ne', 'new', 'nn', 'no', 'oc', 'or', 'os', 'pa', 'pdc', 'pdt', 'pi', 'pnt', 'pt', 'pt-br', 'ru', 'rue', 'sa', 'sah', 'sk', 'sl', 'sli', 'sr', 'sr-ec', 'sr-el', 'stq', 'tcy', 'tg', 'tg-cyrl', 'th', 'tyv', 'udm', 'uk', 'uz', 'vep', 'vi', 'vmf', 'vo', 'vot', 'vro', 'wuu', 'xal', 'yi', 'yue', 'za', 'za', 'zh-classical', 'zh-cn', 'zh-hans', 'zh-hant', 'zh-hk', 'zh-min-nan', 'zh-mo', 'zh-my', 'zh-sg', 'zh-tw', 'zh-yue' );
Change 124824 had a related patch set uploaded by Addshore:
Add tests for formatting dates in different langs
Change 124833 had a related patch set uploaded by Addshore:
Stop ALWAYS usign regex to fix dateFormats
Change 124836 had a related patch set uploaded by Addshore:
DNM WIP Fix formatting in various languages
Change 125177 had a related patch set uploaded by Addshore:
When precision is a <= year use 'Y' dateformat
Change 124824 merged by jenkins-bot:
Add tests for formatting dates in different langs
Change 124833 abandoned by Addshore:
Stop ALWAYS using regex to fix dateFormats
Reason:
See https://gerrit.wikimedia.org/r/124836
Looking at the last patch-set it seems only the languages 'lzh' and 'zh-classical' are left.
Change 129370 had a related patch set uploaded by Aude:
Remove skip tests in MwTimeIsoFormatterTest
Date format for Hungarian is broken.
Instead of "1970. január 1." it is "1 január 1970" This is also valid for the {{#property}} parser function on huwiki. It was fine in November 2013. Why it has changed?
Hungarian [...] "1970. január 1."
The issue described in this task was fixed in April. The issue you are describing is unrelated. Don't get me wrong, it's a valid request, it's just that we don't support styles other than day-month-year (dmy) at the moment for various reasons. You are right, we had partial support for that in the past. But this was client-side only, incomplete and broken and we had to drop it when we switched to server-side parsing and formatting. Please open a new task, e.g "Add support for date formats other than dmy". Thanks.