I think the language2locale script does the right thing:
$ /usr/share/language-tools/language2locale ja_JP
ja_JP.UTF-8
$ /usr/share/language-tools/language2locale ja
ja_JP.UTF-8
$
Two thoughts (haven't looked at the g-c-c code at this time):
* If the locale has not (yet) been generated, i.e. is not included in a 'locale -a' output, language2locale will just return a newline. Can it be that the script is called prematurely?
* Can it possibly be a "UTF-8" vs. "utf8" confusion?
I think the language2locale script does the right thing:
$ /usr/share/ language- tools/language2 locale ja_JP language- tools/language2 locale ja
ja_JP.UTF-8
$ /usr/share/
ja_JP.UTF-8
$
Two thoughts (haven't looked at the g-c-c code at this time):
* If the locale has not (yet) been generated, i.e. is not included in a 'locale -a' output, language2locale will just return a newline. Can it be that the script is called prematurely?
* Can it possibly be a "UTF-8" vs. "utf8" confusion?