You are not logged in.
I have recently noticed that the name of languages in Xinha does not follow ISO recommendations. Languages should named in the locale format: LANGUAGE_COUNTRY, where:
LANGUAGE = ISO 639-1 language code
http://www.loc.gov/standards/iso639-2/php/code_list.php
COUNTRY = ISO 3166-1 country code
http://www.iso.org/iso/country_codes/is … ements.htm
Using this format, English in the United States would be named as "en_US". English in Great Britain would be "en_GB". This allows correct language support since in actual usage languages may be named the same but actually be quite different from country to country (such as German in Germany versus German in Switzerland, or Portugese in Portugal versus Portugese in Brazil, or Chinese in Hong Kong versus Chinese in Taiwan).
There is no logic to the current naming schema. In fact, some languages are named completely incorrectly. The language code "si" is reserved for Sinhala, but in Xinha is the language file for Slovenian.
Let me know if you would like a map of the current language names to proper locale names.
It would be a plus if Xinha would gracefully degrade language choosing, such that if the locale were en_GB but only en_US were available than it would gracefully choose en_US.
It would be an extra plus if all the English strings existed in an en_US file, so that there existed language file that contained all translatable strings. Right now the French file (fr.js) looks to be the most complete, but there is no way to be certain.
Thanks.
Offline
Good ideas both, patches welcome :-)
From memory there is a tool in the contrib of the distribution which can scan through the source and generate a "pristine" language strings file ready for translation.
James Sleeman
Offline
James Sleeman
Offline