🐛🌐 add missing language codes from i18n.js, regenerate strings. #21144
Conversation
| 'bg', // Bulgarian | ||
| 'bs', // Bosnian | ||
| 'ca', // Catalan | ||
| 'cs', // Czech | ||
| 'da', // Danish | ||
| 'de', // German | ||
| 'de-CH', // Swiss German |
There was a problem hiding this comment.
not 100% sure - but do these need to be all lowercase? (also ln46)
There was a problem hiding this comment.
Looks to me like we're following https://en.wikipedia.org/wiki/IETF_language_tag. Whether that was an explicit decision between multiple conventions or not, I don't know.
The codes that are new to the file are consistent with pt-BR (Portuguese as spoken in Brazil), which has been there for quite a while. While we could decide to change conventions, that would require changing names of directories, and some sort of migration strategy for sites with pt-BR, which has been there in that form for quite some time.
So, no, they don't, and let's not make work for ourselves. :) Re sr-Cyrl (and zh-Hant, not new), that's correct. Script designators are Titlecase, while country designators are UPPERCASE, and language codes are lowercase.
Apparently several PRs for new languages were merged without the corresponding additions to i18n.js, which has caused new strings not to be generated for those languages, and missing translations in the frontend, even though we had them in the .json file.
I've added language codes for all languages that had at least some strings translated in i18n/locales, and have run yarn translate to create missing strings and files for these languages.
New language codes for: