Re: Language code to language name mapping

[Behdad Esfahbod <behdad behdad org>]

Bastien Nocera wrote:
> On Wed, 2008-11-26 at 16:52 -0500, Behdad Esfahbod wrote:
>> Vincent Untz wrote:
>>> Hey,
>>>
>>> Le mercredi 26 novembre 2008, à 16:27 -0500, Behdad Esfahbod a écrit :
>>>> My question right now is, how do applications (intlclock, etc...) do that
>>>> currently?  And how do people think it should be done.
>>> intlclock doesn't use iso-codes :-)
>>> Looking at totem: it parses the iso-codes xml. See
>>> totem_lang_table_init() in src/totem-menu.c. In the end, it all ends in
>>> hash tables.
>>>
>>> Abusing gettext could be nice, but the totem code is quite easy too.
>> Thanks for the quick response.
>>
>> Having to parse XML sucks.  It's not a lot of code, but doesn't have to be
>> that hard.  A new library for it is definitely overkill.  That's why I like
>> the gettext abuse.
> 
> How often do you actually need to do it? What's actually used for right
> now?

This was in the context of a new feature Richard and I are adding to
packagekit that prompts the user to install fonts for a language.  Not often
at all.

behdad