Talk:UTF-8
From Gentoo Linux Wiki
Contents |
[edit] -u ???
I wonder what -u in KEYMAP="-u de-latin1" means. Perhaps someone can add this...
- found the answer (see: man kbd_mode)
-u: UTF-8 mode (UNICODE)
[edit] Unicode fonts for CONSOLEFONT
Is there a list of these somewhere? Both the ones installed by standard, but also ones available in portage. -- Freso 09:04, 14 Jan 2005 (GMT)
I use the following UTF-8 font for console: CONSOLEFONT="lat9w-16" suitable (? seems to be utf, but...) for: Belgian, Brazilian, German, Norwegian ...don't know anything more about UTF-8 fonts :)
[edit] That script for setting terms to unicode
Does that still work? Why isn't all this stuff mainstream?! Sheesh. :P
Perhaps the 'emerge euse' should have been a 'emerge slang' in the mc paragraph ? /Bjerre
[edit] Sure about bash 3.0 or readline 5?
I've been using for more than a year gentoo with unicode (I type catalan, spanish, esperanto an russian), in my bash, always runt from 'xterm'. The better it worked, when I used bash 2.05, and readline 4.3. If I use bash 3.0 aThe init script doesn't work for me at all, because en_US.utf8 doesn't match at utf8... who thought I would have LANG=utf8 set in /etc/env.d/02locale? Would it work at all? nd readline 5, the backspace, and inserting new UTF-8 nonascii characters in the middle of a command, result in khaos. In the old bash 2.05 and readline 4.3, everything is fine. Maybe someone can give a reason? --Anonymous
I stuck a stub inside the main article concerning manual deletion of the file "/lib/libreadline.so.4" as it belongs to "sys-libs/readline-5.2_p12-r1" and not an older version of readline. Manually deleting files belonging to packages basically goes against good admin practices and, if readline.so.4 library really conflicts with UTF-8 and shouldn't be compiled & installed, it should be controlled by configure args (aka USE Flags) and not hacking. The reason I stuck this stub within the main article within the "Shells - bash" section instead of just here, this was too obviously hackish in my opinion and I felt it required the immediate attention of all users before people really hacked-up their system. IMO, mandating applications use UTF-8 should be something more controlled by the unicode USE Flag and /etc/rc.conf. --Roger 00:13, 25 February 2008 (UTC)
[edit] example init-script fixed!
it was "buggy". there was a newline too much after the else in this part:
# Using devfs? if [ -e /dev/.devfsd ] || [ -e /dev/.udev -a -d /dev/vc ]; then device=/dev/vc/ else device=/dev/tty fi
[edit] UTF-8 != utf8
there is still a bug in that script, because they tell wrong stuff in the "official gentoo utf8 guide". they say you should create en_GB.UTF-8 locale names, but on my system for ex i never did or configured anything with locales, and i got only en_GB.utf8 and similar! (note the missing "-".)
someone should fix this docu...
Actually, this docu works, so it's not broken. But using UTF-8 in "/etc/locales.build" still seems to give you locales with utf8 spelling in "locales -a". Weirdness... --mikah
Probably got confused with the freeBSD locales (hard to believe though), there they do use *.UTF-8 --Ruud 12:31, 20 May 2005 (GMT)
- cs_CZ.UTF-8 works for me (Doli). cs_CZ.utf8 doesn't.
- Does that mean the init script should have a '-o "${encoding}" = "utf8" added to it? (line 33)
- Yes, indeed. It seems the line should be '-o "${encoding}" = "utf8", otherwise the script fails, no (?)
- The init script doesn't work for me at all, because en_US.utf8 doesn't match at utf8... who thought I would have LANG=utf8 set in /etc/env.d/02locale? Would it work at all?
- Doli, you sure about that? For me it's the other way around. I'll do some more testing though. Sigh, we need to find an official spelling for this, X seems to prefer the "UTF-8" style (though it accepts utf8 too) --mikah
- Please refer to this post for further information regarding UTF-8/utf8 -- 88.134.96.160 22:15, 5 September 2007 (UTC)
[edit] Correct Keymap???
KEYMAP="-u de-latin1" => what is the correct keymap for US english? I tried -u en/-u en-latin1 but didn't work.
- I don't know if the -u flag is even required in there anymore, as long as you have a recent baselayout and UNICODE="yes" in your /etc/rc.conf. Although, I'm not sure -- somebody check /etc/init.d/keymaps and verify this.
[edit] Compose?
Using en_GB.UTF-8 locale, the compose key no longer works. ? — Posted 18:38, 2 March 2007 by User:129.215.37.15
- Is that a question?
- Has it stopped working in all applications, or only some (e.g. those based on Qt widgets)?
- Do you get it back simply by switching back to en_GB.ISO-8859-1? --Swift 08:55, 3 March 2007 (UTC)
[edit] toc breaks layout for me
The span toc breaks the layout of this page for me (in Firefox 2.0.0.4 in 1680x1050). I would prefer a normal traditional table of contents, even if it does waste a little space.
