tried de_DE setting without @euro?

Yes! It displays the same result as de_DE.ISO-8859-15 and de_DE@euro because the .mo files are UTF-8.

Hi let's test that a bit, please. Since Archlinux is an originally English or Canadian based distro the i18n issues come only now into focus more and more. Strangely enough, some languages ship with utf8 support some doesn't . And so doesn't German. First of all please try to create a German utf8 locale on the fly: (as root)
localedef -c -f UTF-8 -i de_DE de_DE.utf8
maybe this will help. If not, try set the locale in /etc/profile to de_DE.utf8
post your resluts here, Thanks.

I don't like utf-8. Is there a way to convert the LC Messages to iso-8859-15?

Hm, UTF8 is the future probably, since it's the only way to get the whole issue standarized and displayable using an 8bit encoding. I consider the old ways of defining the higher bits of an 8 bit as language specific is an old inheritage of the computer history born out of a cultural ignorance. Also, as long as utf-8 is working for you, you don't have to mind the underlaying encoding. So far, I see it isn't working. I just had a look at some /usr/share/locale/de/*mo files and figured, that most of them are just utf8 coded others are not. This is hammered into the mo file itself. It prolly takes some time to figure out how that is working. But if Arch takes some action on i18n my idea is it'll use utf8 since qt and gtk use it as default. I'll stay tuned.

But the important question is: "How to convert the mo files int ISO-8859-15?".

Basically these files _contain_ plain text, but they are separated by binary chunks, read, these are binary files. So using a textbase converter like iconv will either fail or skrew things up. Or will work, but that lottery. You maybe need poedit, I never worked with it though, and dunno if it can save in different encodings. Internally it works in unicode now and thus I expect utf8 to be the default.

hi
when i want to start gxine 0.4.1 (it fails) i get this error:

(gxine:9921): Gtk-WARNING **: Failed to set label from markup due to error parsing markup: Fehler in Zeile 2, Zeichen 9: Ungültiger UTF-8-kodierter Text
Speicherzugriffsfehler

Most German LC_MESSAGES are in iso-8859-1 or iso-8859-15 (except wget and some other). So using UTF-8 may cause such problems.

hi,
there are a lot of utf-8 files, too (on my computer: 44' utf-8, 37' iso-8859-1, 3' iso-8859-15).
The problem is that not all utf-8 messages can be converted to iso-8859-1 (e.g. german style quotations). So utf-8 seems to be the better choice.
Finally, thats not a bug, because every project has different translators, who are using the charset there prefer.

plz close this bug

The last time I tried to delve into UTF-8 issues, it involved my portable music player and my recording of Die Zauberflote and my head hurt for about 2 weeks.

So I don't have any real input, except I guess it makes sense for Matthias-Christian to weigh in on whether or not he still feels this is a bug and what he would consider a reasonable resolution.

This "bug" should be closed before 24 Oct 2005, 1:26pm. It is top of the list since 6 month now.

It is alrealdy solved ?

This is not a bug. Please close this task!