Re: address@hidden: Re: X Input Methods support is not very convenient]

[Stephen J. Turnbull]

>>>>> "rms" == Richard Stallman <address@hidden> quotes someone:

    The original issue was the fact that Emacs 21.1 needs the locale set
    correctly (via $LANG or $LC_ALL) for the non-ASCII X keysyms to be
    decoded as users expect.

I'm not sure I understand this.  The keycode to keysym mapping is
global to the server; Emacs should not change this.  The keysyms which
have string mappings are unambiguous.  Keysyms used in input methods
should be interpreted by the currently active input method.  So
where's the problem of interpretation?

I can imagine many other problems of coordination (eg, deciding coding
system autodetection priorities for external I/O) if the language
environment isn't set correctly.  But X console input isn't one as far
as I know at this point.

    rms> I could work on this; it doesn't look too hard.  But that
    rms> sentence is ambiguous.  Are you describing what Emacs does
    rms> wrong, or what it ought to do to be right?  What locale do I
    rms> want to set up before the call to XOpenIM?

To address the current issue, you would just temporarily set LC_CTYPE
to the locale needed for the user's choice of input method, and
XMODIFIERS to the input method's identification string.  Then reset them.

To get kinput2 for Japanese, for example, set LC_CTYPE="ja_JP.eucJP"
and XMODIFIERS="@im=kinput2".

-- 
Institute of Policy and Planning Sciences     http://turnbull.sk.tsukuba.ac.jp
University of Tsukuba                    Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
              Don't ask how you can "do" free software business;
              ask what your business can "do for" free software.