[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: localization-config uploaded (former locale-config-skolelinux)

(I know, this was sent a long time ago, I just reviewed it)

On Mon, Aug 16, 2004 at 10:24:13PM +0300, Konstantinos Margaritis wrote:
> So, should I remove the @euro extension from the locales?

Some users might already have them, it's best to consider them but also add 
the sans '@euro' extension there. Since some users might have only 'XX' as 
the locale instead of 'XX@euro'. 

One of the issues I see with localization-config, however, is that it will 
not do anything with locales not defined when it might be quite useful to 
have a generic definition for some locales. Consider es_XX or en_XX. In 
most cases the database will contain the _same_ entries for es_ES, es_PE, 
es_DO, es_GT, es_PA, es_HN... but if someone forgets to add these, or the 
user just sets his locale as 'es', localization-config will not be able to 
cope with that and will not configure the package properly.

Localization-config should cope with generic 'LANG_VARIANT@CHARSET' 
definitions in a more generic way:

1.- Make it possible to have a 'LANG' generic definition (to be used by all 
LANG_VARIANTs if not defined)
2.- Make it possible to have a 'LANG_VARIANT' definition (superceedes 1)
3.- As 2 but with 'LANG_VARIANT@CHARSET'

Currently for all spanish definitions I would have to add database entries 
for es, es_ES, es_PE, es_PA, es_HN... even if the value that is going to be 
used is exactly the same. That is not good since:

a) variants might be forgotten (es_SV?)
b) might get inconsistent in the long run (only language is updated but not 
its variants)

> Forgive my persistense, but I can't really understand the use of a 
> @euro or even a new encoding, if noone is using it. 
> Hm, perhaps there should be a transition plan to move everything to 
> UTF-8 :-)

Probably the best in the long term is to have all the user's charsets as 
UTF-8 instead of ISO-8859-1 (or similar ones). That won't be possible 
until all applications fully support UTF-8 and it also means that the 
/etc/locale.alias file needs to be changed (for most european languages it 
will use ISO-8859-1 as the charset)



Attachment: signature.asc
Description: Digital signature

Reply to: