[DRE-maint] Bug#799050: Help needed with a locale-related issue

[Hleb Valoshka]

On 9/16/15, Francesco Poli <invernomuto at paranoici.org> wrote:

> I see that the forwarded report has been closed. I have a doubt,
> though: would the patch fix the issue when LANGUAGE='it' LC_CTYPE=C ?

LC_CTYPE should not affect message language, it affects sorting order.
So the issue you've found, while it looks similar, is a different one.

> I mean: I also get a crash with
>
>   $ LANGUAGE='it' LC_CTYPE=C locale
>   LANG=en_US.UTF-8
>   LANGUAGE=it
>   LC_CTYPE=C
>   LC_NUMERIC="en_US.UTF-8"
>   LC_TIME="en_US.UTF-8"
>   LC_COLLATE="en_US.UTF-8"
>   LC_MONETARY="en_US.UTF-8"
>   LC_MESSAGES="en_US.UTF-8"
>   LC_PAPER="en_US.UTF-8"
>   LC_NAME="en_US.UTF-8"
>   LC_ADDRESS="en_US.UTF-8"
>   LC_TELEPHONE="en_US.UTF-8"
>   LC_MEASUREMENT="en_US.UTF-8"
>   LC_IDENTIFICATION="en_US.UTF-8"
>   LC_ALL=
>   $ LANGUAGE='it' LC_CTYPE=C apt-listbugs -v
>   /usr/lib/ruby/vendor_ruby/aptlistbugs/logic.rb:390:in `width': "\xC3" from
> ASCII-8BIT to UTF-8 (Encoding::UndefinedConversionError)
...
> But the patch seems to only take LC_ALL, LC_MESSAGES, and LANG into
> account...

it seems this particular issue needs additional care, maybe in
ruby-gettext, because GNU gettext with similar variables returns:
LANGUAGE='be_BY:be' LC_CTYPE='C' gettext 'celestia' "About Celestia"
??? Celestia
while ruby-gettext:
[3] pry(main)> str = gettext("About Celestia")
=> "\xD0\x9F\xD1\x80\xD0\xB0 Celestia"

I have no idea how to fix it so far (just don't do that :) ).