[prev in list] [next in list] [prev in thread] [next in thread] 

List:       xfree-i18n
Subject:    [I18n]Re: non-ASCII UTF-8 keysym problems with Solaris X server
From:       Markus Kuhn <Markus.Kuhn () cl ! cam ! ac ! uk>
Date:       2002-11-26 18:08:30
[Download RAW message or body]

Markus Kuhn wrote on 2002-11-24 21:46 UTC:
> I have big problems with entering non-ASCII characters.
> Whenever I press keys to which I have assigned with xmodmap non-ASCII
> keysyms such as
> 
>   adiaeresis, leftsinglequotemark, leftdoublequotemark, ...
> 
> then Emacs inserts into the buffer a sequence of two or three characters
> that look like the UTF-8 byte sequences for the entered characters, but
> interpreted as ISO 8859-1 bytes. So pressing ä causes ä to be inserted
> into the buffer. Also, cut&paste into an xterm in UTF-8 mode is broken.
> When I select the ¤ in the buffer and paste it into UTF-8 xterm, I get
> it again expanded to  ¤.
> 
> Environment variables: LANG=en_GB.UTF-8

A few more curicial observations on this problem:

a) My X server is on a Solaris 5.8 machine (xdpyinfo Sun vendor
   release number: 6410).

   When I use an XFree86 4.1.0 server under Red Hat 7.2, then
   the above problem does *not* show up.

b) When I use an old xterm (XFree86 4.0.3(156)) and associated
   old Xlib, the problem does not show up. Only if I use a recent
   xterm and Xlib (XFree86 4.2.0(165) on SuSE 8.1 or Red Hat 8.0),
   the non-ASCII keysyms are mangled by an unnecessary Latin1->UTF-8
   conversion. Not only emacs is affected, but also xterm.

Summary: If I connect with a recent (XFree86 4.2, RH8 or SuSE81) client
that runs under a UTF-8 locale to a Solaris X server, then both for
xterm and emacs 21.2 the keystroke -> multibyte-character transform
contains somewhere an unneeded ISO 8859-1 -> UTF-8 conversion step that
mangles up the entry of non-ASCII data.

Any ideas where in the keystroke pipeline this might happen and what
might have gone wrong?

Why does it make a difference, what X server I use, whether keysyms are
translated correctly for clients running UTF-8 locales?

Very puzzled ...

Markus

-- 
Markus G. Kuhn, Computer Laboratory, University of Cambridge, UK
Email: mkuhn at acm.org,  WWW: <http://www.cl.cam.ac.uk/~mgk25/>

_______________________________________________
I18n mailing list
I18n@XFree86.Org
http://XFree86.Org/mailman/listinfo/i18n
[prev in list] [next in list] [prev in thread] [next in thread]