Closed kelson42 closed 1 year ago
I'd advise the scraper handles this (best effort) ; keeping in mind we have the CMS.
Scraper could have a limited map of Title/Description for languages and fallback to what we have now. In any way, we'd need an Erik-provided Title and Description for English that we can translate in major languages.
As the number of languages is already important, we could imagine attempting translations using an online service but most require an API key, Libretranslate doesn't support much languages (and it's only the major ones we could get translations for) and even Google or Microsoft wouldn't support them all.
@rgaudin OK, it would be good that these two strings are besides the others. @eshellman Any comment on this? You see in the screenshot how it is today? Do you have a source of translated strings which would be better?
No, but I think the community could generate such a list in a few weeks.
so the strings to be translates would be Title: "Project Gutenberg Library - Books in
This issue has been automatically marked as stale because it has not had recent activity. It will be now be reviewed manually. Thank you for your contributions.
Best handled with Translatewiki I guess.
We just need here to have the strings in the translation files. Trivial.
@rgaudin I want to fix this ASAP now. If I implement #56, does that will work out of the box?
Currently the Gutenberg ZIM we produce share the exact same ZIM title/description metadata (in English)
This, although:
I think we should either introduce translated versions of these metadata or split the recipe in Zimfarm https://farm.openzim.org/recipes/gutenberg_mul_all.
@rgaudin Not sure what is the best approach