xterm 8 bit meta characters & unicode

["Sheshadrivasan B" <sheshv@xxxxxxxxxxx>]

Can some one comment xterm generating 8-bit meta characters ? (see email 
below)

We use libtecla(API for processing of terminal input by Martin) for 
implementation of a custom shell.
and  plan to internationationalize it. This implies libtecla needs to be 
able to support
UTF-8.
Thanks,
Shesh.


From: Martin Shepherd [mcs@xxxxxxxxxxxxxxxxx]
Sent: Monday, March 28, 2005 6:31 AM
To: Sheshadrivasan B
Subject: RE: Libtecla & internationalization

Hi Shesh,

Sorry for the delay, I didn't want to respond until I had had time to take a 
better look at the issues surrounding UTF-8, and I can only work on this 
during little bits of free time. My previous claim that libtecla already 
supported UTF-8 characters that only needed one terminal column for display, 
turns out to be incorrect. Sorry about that. I've spend quite a lot of time 
this weekend reading up on UTF-8, and thinking about how to go about 
modifying libtecla. Unfortunately it isn't going to be easy, especially 
since I need to continue to support older character encodings. I'm trying to 
find out right now what to do about 8-bit meta-key values that terminals 
could output into an otherwise valid UTF-8 stream. Most terminal emulators 
appear to avoid this problem, by outputting two-character escape sequences 
instead of 8-bit meta characters, but xterm appears to default to generating 
8-bit meta characters, so I'm puzzled about how to detect these within a 
UTF-8 stream.

....
message truncated...



--
Linux-UTF8:   i18n of Linux on all levels
Archive:      http://mail.nl.linux.org/linux-utf8/