Mailing List Archive

Support open source code!

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: tlug: kinput2 and mule

tlug note from "Stephen J. Turnbull" <>
>>>>> "C" == C Oda <> writes:

    C> I've been experimenting with different front end processors.  I
    C> like kinput2 due to the popup kanji menu that I can use to
    C> select kanji.  Paul Gampe showed me how to get kinput2 to send
    C> kanji to a mule session in a kterm. (I've added the .emacs file
    C> to my linux-nihongo doc).  I can also use kinput2 with vi,
    C> xjdic, and most things that run in a kterm.  This is good.

    C> However, since mule has some nice X capabilities, I thought it
    C> would be interesting to have kinput2 input Japanese under mule
    C> X.  Sometimes I like to use the mouse to click a window or a
    C> frame.  What I've found is that kinput2 works with these
    C> protocols:

    C> 		    kinput protocol, kinput2 protocol,

As far as I know these are the only ones useful with current Linux

    C> Matsushita jinput protocol, Sony xlc protocol, XIMP protocol,

Forget these; the first two are only implemented on proprietary
hardware AFAIK (although I imagine both protocols are contributed for
free use), and XIMP is a now-deprecated SONY implementation of the X
Input Method abstraction, and is superseded by the X Input Method

    C> X Input Method Protocol (X Consortium standard)

I have an example program due to OReilly working with this (check out
However getting it working was hairy and dependent on all sorts of
little details which are not documented.  Using XIM is not documented
at all; what is available from the X Consortium is the protocol
standard (anybody who's read Volume Zero of OReilly's X Series will
know how useless *that* is).  The implementation in XFree86 libX11 is
not robust.

Mule does not include code to _use_ XIM, only to open it :-P  I don't
know how hard it will be to fix; I will look and see what happens in
XEmacs 20, and if contrary to promises it's not there, I will have a
go at implementing in GNU Emacs on a rudimentary level.  But that will 
have to wait until April.

Dave Iannucci and I are planning to attend the Multilingual
Information Processing conference in Tsukuba at the end of the month,
maybe we'll learn more then.  (For more information, see

    C> What I want is a consistent front end processor across all
    C> applications.  It is getting a little tedious to use uum for
    C> xjdic and jvim, mule for e-mail, and kinput2 when I feel in the
    C> mood.

kinput is an input manager, not an FEP---you can change the
translation engine.  Get the source code, and configure kinput2 to use
Canna or Wnn, and use the corresponding Mule.  Then you'll be using
the same keystrokes in all applications; mule in a native X window
will not give pop-up menus, of course, to get that you'll need to use
Mule in a kterm.

I have done the same as you, but I don't use xjdic much, nor jvim at
all, so the FEP doesn't matter (I live in Canna/Mule) and I haven't
completely reconfigured my system along the above lines.  But it
should work.

However, with XEmacs promising XIM and Japanese support, you should
consider simply waiting for that as a turn-key solution.

                            Stephen J. Turnbull
Institute of Policy and Planning Sciences                    Yaseppochi-Gumi
University of Tsukuba            
Tel: +81 (298) 53-5091;  Fax: 55-3849    
a word from the sponsor will appear below
The TLUG mailing list is proudly sponsored by TWICS - Japan's First
Public-Access Internet System.  Now offering 20,000 yen/year flat
rate Internet access with no time charges.  Full line of corporate
Internet and intranet products are available.
Tel: 03-3351-5977   Fax: 03-3353-6096

Home | Main Index | Thread Index

Home Page Mailing List Linux and Japan TLUG Members Links