A clarification. This only supports the ISO 639-1 codes right? Not the three-letter ISO 639-2 codes which exist for a wider array of languages. Or is that a limitation in how {{#language}} is implemented?
Topic on Extension talk:CLDR
Thanks. That list seems to be using a mixture between ISO 639-1 codes and ISO 639-3 codes. Looks specifically like it only uses the 639-3 code if there is no 639-1 code.
Meanwhile our implementation through {{#language:}} only seems to support the 639-1 codes.
Example:
| Language | 639-1 | 639-3 | CLDR | {{#language:<639-1>}} | {{#language:<639-3>}} |
|---|---|---|---|---|---|
| Afar | aa | aar | aa | Qafár af | aar |
| Acoli | - | ach | ach | - | ach |
That's an unwarranted conclusion; we simply don't have that language name yet. There's plenty of ISO 639-3 codes.
Ok. Just found a 639-3 example which works (粵語). Still would a clarification for Usage and parameters saying something which boils down to "ISO 639 = ISO 639-1 if exists otherwise ISO 639-3" be correct?
Should be in general, as it's the policy described in and we follow the same (as most software in the world, I believe). You may just link Manual:Language#Language code.
That was teh one I was looking for =). Added a comment now. Thanks
I created a converter iso 639-3 to iso 639-2 and another iso 639-2 to iso 639-3, see Template:To_iso_639-3 and Template:To_iso_639-2 sur la base de en:List_of_ISO_639-1_codes. This pose several issues since there is more 639-3 than 639-2. It's not perfect, but it helps us to convert iso 639-3 "mar" into "mr" thanks to {To_iso_639-2} then into "Bangla" thanks to {{#language:mr|en}}.