Last modified: 2013-09-02 10:18:37 UTC
Currently, when you use a language which have digit transforms, it is enabled by default. and this is a user preference, they might want the default numerals or they want the transformed numerals. So I'm proposing to add a preference option to disable $wgTranslateNumerals when using a language with digit transforms. e.g. Arabic. --user:alnokta
Component: User preferences -> i18n
Is it likely enough that this warrants adding a new user preference?
I take in the numeric transform relates to sortable tables??? If so try putting var wgSeparatorTransformTable = null; var wgDigitTransformTable = null; in you monobook.js file (or whatever that is in Arabic)
No never mind, that'll only work for the tables...
Comment #2 -> yes. keep in mind that it should be visible only in languages with digit transform. ar fa ur etc.
It will be hard to implement, because there is no separation between content and interface number translation.
It would have been better if it could be implemented. Se discussion and poll in Assamese wikipedia: https://secure.wikimedia.org/wikipedia/as/wiki/Wikipedia:Community_Votes#Assamese_Vs_Latin_numerals_in_templates_using_parser_functions
You are asking for an user preference option, but that would not affect the result of magic words at all. I don't understand why you want to sacrifice usability of readers to get some convenience for advanced editors editing templates. It seems to me that you have not thought out this issue properly.
Thanks. @Niklas: Right, we could have solved the problem by editing some templates or even patching the code related to the magic words like {{CURRENTDAY}} etc. But, we, or at least the Assamese community are accustomed using Latin numerals and symbols in mathematics starting from primary schools and even in our daily lives. Some other Indian language communities, like the Hindi users also follow the same convention. Apart from that English numerals are very common in India and almost everyone can understand them. Basically this is the reason we have adopted this approach.
(In reply to comment #9) > Thanks. > > @Niklas: Right, we could have solved the problem by editing some templates or > even patching the code related to the magic words like {{CURRENTDAY}} etc. > But, > we, or at least the Assamese community are accustomed using Latin numerals > and > symbols in mathematics starting from primary schools and even in our daily > lives. Some other Indian language communities, like the Hindi users also > follow > the same convention. Apart from that English numerals are very common in > India > and almost everyone can understand them. > > Basically this is the reason we have adopted this approach. The numeral problem is a general one for indic languages. The problem is two-fold: inputting different numeral systems (parser function support) and viewing different numerals on a per-user basis. For users being able to see numerals of their choice, you could try the numeral converter gadget as used on hindi wikipedia [1]. Its a javascript-based gadget which uses a cookie to set preference of a user on a browser. It is also usable for IP users. Code is also availaible on mediawiki[2]. For input support for parsers, the module convert on english wikipedia somewhat solves this problem. See this user discussion on english wikipedia for a possible solution [3]. [1]: http://hi.wikipedia.org/wiki/%E0%A4%AE%E0%A5%80%E0%A4%A1%E0%A4%BF%E0%A4%AF%E0%A4%BE%E0%A4%B5%E0%A4%BF%E0%A4%95%E0%A4%BF:Gadget-Numeral_converter.js and http://hi.wikipedia.org/wiki/%E0%A4%AE%E0%A5%80%E0%A4%A1%E0%A4%BF%E0%A4%AF%E0%A4%BE%E0%A4%B5%E0%A4%BF%E0%A4%95%E0%A4%BF:Gadget-Numeral_converter.css [2]: http://www.mediawiki.org/wiki/MediaWiki:Gadget-Numerakri.js and http://www.mediawiki.org/wiki/MediaWiki:Gadget-Numerakri.css [3]: http://en.wikipedia.org/wiki/User_talk:Johnuniq#Module:Convert_and_non-arabic_digits
