waqar added a comment.
> With no respect for your lack of being subtle
I meant no disrespect to anyone. If you felt offended I apologize
> we need to maintain this for 20 years in the future, you making a quick
hack here doesn't make this any better, the user will still see cv_RU in random
places and will be all annoyed anyway.
Can you give examples of places where an iso_code (cv_RU) could pop up?
Spellchecker is a frequently accessed tool. It can have more languages than
any other tool. The user can just place a dictionary in the directory and it is
the job of spellchecker to load those dictionaries and their filenames
correctly.
> Now go and make the fix in Qt
Please don't give orders.
Anyways, I looked deep into this to figure out what's the issue. The relevant
data is stored in qlocale_data_p.h, it does have most of the iso_codes (even
the ones that are being displayed incorrectly). What it doesn't have is native
language names.
So, if I do,
QLocale locale("cv");
QLocale::Language l = locale.language();
QString localizedLangName = locale.languageToString(l);
It shows `Chuvash` as the language name.
Now, qlocale_data_p.h is a generated file. You can't edit it manually. Qt
takes the data from CLDR <http://cldr.unicode.org/>. And CLDR doesn't have the
native names for these languages. CLDR is a huge database of locale stuff, and
it is not very easy to contribute data if you aren't a native speaker.
> if it gets rejected, we can speak about doing dirty fixes.
It's a fallback mechanism for when QLocale fails.
REPOSITORY
R246 Sonnet
REVISION DETAIL
https://phabricator.kde.org/D25530
To: waqar
Cc: aspotashev, aacid, yurchor, kde-frameworks-devel, LeGast00n, GB_2,
michaelh, ngraham, bruns
