Re: XIM, kinput2 & Tk

[jwb@example.com (Jim Breen)]

>> From: "Stephen J. Turnbull" <turnbull@example.com>
>> >>>>> "B0Ti" == B0Ti  <9915104t@example.com> writes:
>>     B0Ti> Jim Breen wrote:
>>     >> And to follow up my previous partially rhetorical question, if
>>     >> a Swede wants to run a kterm, while still getting messages
>>     >> etc. in Swedish,
>>     >> (a) what does he/she set variables to?
>>     >> (b) how should kterm respond to which which variables?
>> 
>> However, in theory, you simply set LANG=se_SE.ctext.  Then what
>> happens is that the application spits out Japanese output in
>> ISO-2022-JP, while kterm's messages are in ISO-8859-1.  This only
>> works because of the definition of the X Compound Text encoding and
>> the assumption that the app is capable of producing ISO-2022-JP.

Won't this stop kterm calling up "kinput2 -xim", because LANG is not set
to Japanese?

Also you mention "the application". It's kterm itself in this case.

>> A more general solution is to set LANG=ja_JP.SJIS (or any Japanese)
>> and LC_MESSAGES=se_SE.iso8859-1.  The only way that should fail is if
>> there is no message catalog in Swedish.

That "LANG=ja_JP.SJIS" really worries me. One should be able to run
apps without the system trying to enforce a code-encapsulation.
LANG=ja_JP.SJIS is trying to stack three different things into the one
variable.

Jim
-- 
Jim Breen  [jwb@example.com  http://www.csse.monash.edu.au/~jwb/]
Visiting Professor, Institute for the Study of Languages and Cultures of 
Asia and Africa, Tokyo University of Foreign Studies, Japan
+81 3 5974 3880         [$B%8%`!&%V%j!<%s(B@$BEl5~30Bg(B]