tlug: XEmacs 20.0 XIM? not! [was: XEmacs 20.0 Wnn Input]

["Stephen J. Turnbull" <turnbull@example.com>]

--------------------------------------------------------
tlug note from "Stephen J. Turnbull" <turnbull@example.com>
--------------------------------------------------------
XEmacs is up and running, but no XIM I can see.  I have to say that
XEmacs _does_ handle improperly sized fonts extremely well; the
default latin font is courier-infinitesimal, and the JIS font is
gothic-kyodai, but everything is readable, there are no obnoxious
overwriting or missing pixels, and the interlinear space is much
better than in GNU Emacs.

I do appreciate the separate directories for packages under ./lisp,
too.

HATE that toolbar!  What's next, MS Emacs?

>>>>> "Steve" == Steve Dunham <dunham@example.com> writes:

    Steve> Put this in your .emacs:

    Steve> (when (featurep 'wnn) (require 'egg-wnn) (setq
    Steve> egg-default-startup-file "eggrc-wnn"))

>>>>> "me" == Stephen J. Turnbull <turnbull@example.com>

    me> I take it that this means that XIM is _still_ unsupported?

    me> But you promised you promised youpromised YOUPROMISED!!

    me> Gripe gripe gripe gripe.

    Steve> As far as I know, XIM _is_ supported; I just haven't tried

I quote from xemacs-20.0/src/intl.c:

	      /* This stuff is far, far from working. */

    Steve> it yet.  There are two different XIM options when you
    Steve> compile, depending on whether you're linking with Motif or

Yeah, saw that.  Couldn't find any way to _invoke_ an X input method;
the stuff in xemacs-20.0/mule-doc/ is pre-Cambrian (well, I
exaggerate; it's at worst Cretaceous, I'm _sure_ I saw a Triceratops
in that directory).  Seriously, the change log says "Based on Emacs
18."  This worries me, since current versions of Mule (SUETSUMUHANA)
are clearly aimed at Emacs 19.

Couldn't find any place where the X input methods get interned, and
the variables that allege that i18n is implemented ('i18n3' and 'i18n4')
are not interned on my copy of XEmacs.  That doesn't mean anything,
I'm not an Emacs source hacker.  I may be looking in wildly wrong
places.

    Steve> not, so I assume it's supported.  When I get around to

Not a good assumption when you're dealing with half-documented beta
software.  If you'll weaken that "assume it's supported" to "take that
as enough evidence that it's supported for me to spend an hour trying
to get it working", douki desu.  Couldn't get it though (couldn't find 
an interface to open an input method in the lisp code), so I'm giving 
up unless you know how to get it started.

    Steve> compiling kinput2, I may try it.

Um, as far as I know there are no clients that work with kinput2 in
XIM mode, except for my hacked version of O'Reilly's i18n-input.c.  I
couldn't get kterm 6.2.0 to work with it, anyway.  kterm does, of
course, support the kinput2 protocol, so if you're going to do
Japanese input in a terminal window (jgrep or the like), you should
definitely get kinput2 running.  But don't bet on it working via the
XIM interface.

I suspect bugs in the Xlib implementation of XIM, myself, but it's
hard to say.  *Sigh* I wish I had time to stay up with the Mule/
XEmacs/X newsgroups....

    Steve> In addition to XIM, there are two lisp-based input methods,
    Steve> one of which is egg-WNN, which I described above.  I
    Steve> believe the other uses the quail stuff and canna, but I
    Steve> don't know much about it.

canna uses 'canna-egg' or something like that.  Looks just like egg's
fence-mode, anyway.  Quail is a completely separate evolutionary
branch, which is descended from skk.el.  Both the IM and the
dictionary are implemented internally to Mule in Lisp (the big
disadvantage to Quail for non-native speakers, IMHO; although even
that disappears if you are willing to use XEmacs as your text widget
in applications---gasp! the mind boggles---use XEmacs as the text
widget for XPostIt!!!  3k of XPostIt code supported by 4MB of Emacs
code plus a cast of thousands of .el files).

Anyway, I'm trashing xemacs-20.0 and going on to 20.2-beta.

Yikes, that's fast!  (53.77 kB/s!)  I'm not going to wait for the
compile, but I'll check intl.c for the "far from working" comment ... 
Ah well, it's still there....
-- 
                            Stephen J. Turnbull
Institute of Policy and Planning Sciences                    Yaseppochi-Gumi
University of Tsukuba                      http://turnbull.sk.tsukuba.ac.jp/
Tel: +81 (298) 53-5091;  Fax: 55-3849              turnbull@example.com
-----------------------------------------------------------------
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.   info@example.com
Tel: 03-3351-5977   Fax: 03-3353-6096