The language data used by mw.language is currently only for the wgUserLanguage (exported by ResourceLoaderLanguageDataModule.php). This limits the ability to add i18n/l10n support for gadgets and user scripts that are used for editing, since edits should be made in the wgContentLanguage of the wiki.
The language data (digitTransformTable, grammarTransformations, pluralRules, etc) are not accessible even using the API. An API endpoint should be created to expose this data.
| Subject | Repo | Branch | Lines +/- | |
|---|---|---|---|---|
| Expose additional language data through ApiQueryLanguageinfo | mediawiki/core | master | +62 -2 |
ApiQueryLanguageInfo is related. Would it be acceptable to expand this API to also provide the i18n rules data?
Along with that, i18n tools such as mw.language and jquery.18n should give a choice between using wgUserLanguage or wgContentLanguage, rather than always using wgUserLanguage.
Seems like a seperate request
Change 693214 had a related patch set uploaded (by SD0001; author: SD0001):
[mediawiki/core@master] Expose additional language data through ApiQueryLanguageinfo
Currently the banana-i18n library does provide a way to do i18n in non-wgUserLanguage. However, the limitation is that it bundles a large amount of data (like fallbacks.json) for all possible languages. A more efficient approach that reduces the bundle size would be to use a library that makes optimally cached requests to fetch the language data only for the language(s) needed. Exposing the language data through the API would be the first step towards writing such an i18n library.