Changing the language of the gdm login screen

[hamburger123]

Is there a way to change the language the gdm login screen is in? Right now I can select language from a menu for the GNOME session, but not for the login screen itself. I'm using the default graphical greeter theme that comes with FC4. Thanks in advance.

[borromini]

Yes there is; I don't know which distro you're running though and it could be distro-specific. Open a terminal, become root, and edit the lang.sh and lang.csh files on your system (you only have two of them). The default language in both will be en_US. Change it to the one of your liking. That will change your system language.

On my Slackware-style system, both files are located in /etc/profile.d/. Maybe you have the same directory...

[hamburger123]

Will this affect a whole lot of other stuff?

And also which part should I edit? I see some zh_CN, but the login screen is still in english.

# /etc/profile.d/lang.sh - set i18n stuff

sourced=0
for langfile in /etc/sysconfig/i18n $HOME/.i18n ; do
    [ -f $langfile ] && . $langfile && sourced=1
done

if [ -n "$GDM_LANG" ]; then
    sourced=1
    LANG="$GDM_LANG"
    unset LANGUAGE
    if [ "$GDM_LANG" = "zh_CN.GB18030" ]; then
      export LANGUAGE="zh_CN.GB18030:zh_CN.GB2312:zh_CN"
    fi
fi

if [ "$sourced" = 1 ]; then
    [ -n "$LANG" ] && export LANG || unset LANG
    [ -n "$LC_ADDRESS" ] && export LC_ADDRESS || unset LC_ADDRESS
    [ -n "$LC_CTYPE" ] && export LC_CTYPE || unset LC_CTYPE
    [ -n "$LC_COLLATE" ] && export LC_COLLATE || unset LC_COLLATE
    [ -n "$LC_IDENTIFICATION" ] && export LC_IDENTIFICATION || unset LC_IDENTIFICATION
    [ -n "$LC_MEASUREMENT" ] && export LC_MEASUREMENT || unset LC_MEASUREMENT
    [ -n "$LC_MESSAGES" ] && export LC_MESSAGES || unset LC_MESSAGES
    [ -n "$LC_MONETARY" ] && export LC_MONETARY || unset LC_MONETARY
    [ -n "$LC_NAME" ] && export LC_NAME || unset LC_NAME
    [ -n "$LC_NUMERIC" ] && export LC_NUMERIC || unset LC_NUMERIC
    [ -n "$LC_PAPER" ] && export LC_PAPER || unset LC_PAPER
    [ -n "$LC_TELEPHONE" ] && export LC_TELEPHONE || unset LC_TELEPHONE
    [ -n "$LC_TIME" ] && export LC_TIME || unset LC_TIME
    if [ -n "$LC_ALL" ]; then
       if [ "$LC_ALL" != "$LANG" ]; then
         export LC_ALL
       else
         unset LC_ALL
       fi
    else
       unset LC_ALL
    fi
    [ -n "$LANGUAGE" ] && export LANGUAGE || unset LANGUAGE
    [ -n "$LINGUAS" ] && export LINGUAS || unset LINGUAS
    [ -n "$_XKB_CHARSET" ] && export _XKB_CHARSET || unset _XKB_CHARSET

    if [ -n "$CHARSET" ]; then
        case $CHARSET in
            8859-1|8859-2|8859-5|8859-15|koi*)
                if [ "$TERM" = "linux" -a "`/sbin/consoletype`" = "vt" ]; then
                       echo -n -e '\033(K' 2>/dev/null > /proc/$$/fd/0
                fi
                ;;
        esac
    elif [ -n "$SYSFONTACM" ]; then
        case $SYSFONTACM in
            iso01*|iso02*|iso05*|iso15*|koi*|latin2-ucw*)
                if [ "$TERM" = "linux" -a "`/sbin/consoletype`" = "vt" ]; then
                        echo -n -e '\033(K' 2>/dev/null > /proc/$$/fd/0
                fi
                ;;
        esac
    fi
    if [ -n "$LANG" ]; then
      case $LANG in
        *.utf8*|*.UTF-8*)
        if [ "$TERM" = "linux" -a "`/sbin/consoletype`" = "vt" ]; then
                [ -x /bin/unicode_start ] && /sbin/consoletype fg && unicode_start $SYSFONT $SYSFONTACM
        fi
        ;;
      esac
    fi

    unset SYSFONTACM SYSFONT
fi
unset sourced
unset langfile

And here's the /etc/sysconfig/i18n file:

LANG="en_US.UTF-8"
SYSFONT="latarcyrheb-sun16"
SUPPORTED="en_US.UTF-8:en_US:en"

[borromini]

What's your default language now? As for my lang.sh file, it's far more simple...

Here's mine (I do NOT recommend you replace it, I'll have a look at yours, this is just for comparison):

#!/bin/sh
# Set the system locale.  (no, we don't have a menu for this ;-)
# For a list of locales which are supported by this machine, type:
#   locale -a

# en_US is the Slackware default locale:
export LANG=nl_BE

# 'C' is the old Slackware (and UNIX) default, which is 127-bit
# ASCII with a charmap setting of ANSI_X3.4-1968.  These days,
# it's better to use en_US or another modern $LANG setting to
# support extended character sets.
#export LANG=C

# There is also support for UTF-8 locales, but be aware that
# some programs are not yet able to handle UTF-8 and will fail to
# run properly.  In those cases, you can set LANG=C before
# starting them.  Still, I'd avoid UTF unless you actually need it.
#export LANG=en_US.UTF-8

# Another option for en_US:
#export LANG=en_US.ISO8859-1

# One side effect of the newer locales is that the sort order
# is no longer according to ASCII values, so the sort order will
# change in many places.  Since this isn't usually expected and
# can break scripts, we'll stick with traditional ASCII sorting.
# If you'd prefer the sort algorithm that goes with your $LANG
# setting, comment this out.
export LC_COLLATE=C

# End of /etc/profile.d/lang.sh

Edit: I suggest you check the first two files mentioned in the file - the /etc/sysconfig/i18n and $HOME/.i18n (mind the '.' - it means it's a hidden file). Maybe you can find some clarifications in there. Because this looks like a rather complicated script to me.

[hamburger123]

Maybe I should just edit the /etc/sysconfig/i18n file, though I don't know what the "SUPPORTED=" line means.

[borromini]

Can you post the contents of that file also? I think supported= means the locales available on your system. Not sure though.

You should try this as root:

dpkg-reconfigure locales

You should be able to select your desired language there. Then restart, and it should be ok.

[hamburger123]

I'm using Fedora Core 4 so I don't think dpkg is an available option. I edited the i18n file into an earlier post, sorry I forgot to tell you . Here it is again:

LANG="en_US.UTF-8"
SYSFONT="latarcyrheb-sun16"
SUPPORTED="en_US.UTF-8:en_US:en"

I ran system-config-language(distro specific I think) and it popped up a list of languages to choose from and the only option available was "English (USA)

`locale -a` gives me a ton of different languages, is it okay if I changed "LANG=" to one of those?