Lucas_Werkmeister_WMDE added subscribers: Lydia_Pintscher, Addshore, 
Lucas_Werkmeister_WMDE.
Lucas_Werkmeister_WMDE added a comment.


  Result of conversation with @Addshore (I hope I’m not getting anything too 
wrong): as long as we only use `ParserOutput::getUserLang()`, we should be 
fine. All parts of the interface will use the same language (that one, 
regardless of whether it’s the same as `$wgLang` or not), and the parser cache 
is split automatically.
  
  However, T215387 <https://phabricator.wikimedia.org/T215387> says that not 
only the current language, but all the languages which a user speaks (according 
to Babel) should be prioritized on the list of languages where information is 
displayed. This complicates the situation, because the output now also depends 
on the current user. The termbox for Wikibase has the same problem, with 
complicated solutions; however, since we expect schemas to be viewed much less 
frequently, we could go for a simpler approach and, for instance, completely 
ignore the parser cache for logged in users (or users with custom Babel 
information).
  
  Alternatively, we could drop the Babel requirement. That would be the 
simplest approach – @Lydia_Pintscher, would that be acceptable?

TASK DETAIL
  https://phabricator.wikimedia.org/T217773

EMAIL PREFERENCES
  https://phabricator.wikimedia.org/settings/panel/emailpreferences/

To: Lucas_Werkmeister_WMDE
Cc: Lucas_Werkmeister_WMDE, Addshore, Lydia_Pintscher, Aklapper, Michael, 
WMDE-leszek, pdehaye, alaa_wmde, Nandana, Lahi, Gq86, GoranSMilovanovic, 
QZanden, YULdigitalpreservation, LawExplorer, Salgo60, _jensen, rosalieper, 
abian, Wikidata-bugs, aude, Mbch331
_______________________________________________
Wikidata-bugs mailing list
[email protected]
https://lists.wikimedia.org/mailman/listinfo/wikidata-bugs

Reply via email to