>>>> "Jan" == Jan Vroonhof
<jan.vroonhof(a)ntlworld.com> writes:
> More additional info: --with=mule=yes --with-xim=xlib and
> Athena widgets/dialogs works as well. Seems like the instant
> you allow Motif to be linked in for *anything* (even XIM) it
> breaks Mule.
Motif has historically been problematic on open source platforms
(Linux, *BSD). Steve Baur had it completed deprecated for a long time,
would just spit at bug reports with Linux+Motif.
Jan> Does just "--with-xim=xlib" do the trick? There have been
Jan> patches floating around to disable motif based xim completely
Jan> because of its instability.
XIM itself is a crock. It is full of crashes (that's how I got
involved in XEmacs in the first place---it was easier to try to debug
Martin's XIM code than the code in FSF 19.34 and 20.0---and it was
internal to XF86, not Martin's code, as you might expect).
That said, I _think_ XEmacs --with-xim=xlib is pretty stable now. I
haven't had problems with it (but I normally use native Canna, not
Canna+kinput2 -xim, for Japanese input; I'll switch for the next devel
cycle). Anyway, I haven't seen any Japanese bitches about it, and I
know a lot of Japanese are using XIM because they want to use the
commercial input methods.
_However_, if you have any Motif libraries at all, you may be _forced_
to build in the Motif XIM stuff. Motif's I18N model, especially the
XIM stuff, varies markedly from Xlib's.
--
University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Institute of Policy and Planning Sciences Tel/fax: +81 (298) 53-5091
_________________ _________________ _________________ _________________
What are those straight lines for? "XEmacs rules."