← Back to team overview

ubuntu-translations-coordinators team mailing list archive

Re: Shipping emoji translated data

 

Hi Marco!

On 2021-05-06 16:31, Marco Trevisan wrote:
Hey guys!

First important things, first... :)
Let's talk about Emoji metadata!

So, gtk3 and gtk4 will support translated emoji metadata (from unicode
CLDR [0]) so that one will be able to find in their language the right
emoji when using `Ctrl+.` emoji chooser.
Now, for this to happen localized gresource files generated from the
emojibase [1] must be in /usr/share/gtk-{3,4}/emoji/ [2].

As per this: how do we ship them?
I think that ideally this would be included in each
`language-pack-gnome-<LANG>` package, but how can we manage this?

My first thought, if we want to ship them via the language packs, is that a package including all of them should be created. Next a new section in the pkgstriptranslations script (in pkgbinarymangler) would be needed. That way they would be stripped into a tarball at build time and added to the language-pack-gnome-<LANG>-base packages whenever a full language pack update is accomplished. I'm thinking that the emoji metadata would be handled in the same way as e.g. localized XML help pages.

If we want this to work at the release of 21.10, I think it's highly desirable to make the changes early in the cycle to be able to test.

Generating the files is easy and having a process that automatizes it
isn't complicated, but I would like how this can be handled. >
Cheers

[0] http://cldr.unicode.org/
[1] https://github.com/milesj/emojibase
[2] https://gitlab.gnome.org/GNOME/gtk/-/blob/master/gtk/emoji/README.md


--
Cheers,

Gunnar Hjalmarsson
https://launchpad.net/~gunnarhj


References