Re: does XIM work in ja_JP.UTF-8 locale?

Mike> I found that including Mike> (set-coding-priority-list '(utf-8)) Mike> (set-coding-category-system 'utf-8 'utf-8)) Mike> in my ~/.xemacs/init.el or evaluating these two lines Mike> manually in the scratch buffer makes it work. Why these two Mike> lines? Because they allow autodetection of UTF-8 to work. Why we're doing autodetection I don't know. Except maybe ... Mike> And why don't Mike> (set-keyboard-coding-system 'utf-8) Mike> (set-terminal-coding-system 'utf-8) Mike> (set-default-coding-systems 'utf-8) Mike> have any influence? ... because the XIM input model bypasses all the normal input processing. Evidently Martin just attached an autodetecting lstream to the XIM stream. Yep (from event-Xt.c, ca l. 1060): /* #### Use Fget_coding_system (Vcomposed_input_coding_system) */ instream = make_decoding_input_stream (XLSTREAM (fb_instream), Fget_coding_system (Qundecided)); Nobody has ever really taken XIM seriously; all the Japanese I know use native Canna or Wnn, or the new all elisp versions (which got a _huge_ boost from the FSF's decision to remove native IM server support from GNU Emacs, boy were people upset about that). It wouldn't be hard to add the composed-input-coding-system variable, I bet. Mike> Why does it make a difference whether I evaluate these Mike> functions on the command line or in my ~/.xemacs/init.el? I don't know. This LC_CTYPE=ja_JP.UTF-8 XMODIFIERS="＠im=kinput2" \ xemacs -q -eval "(require 'un-define) \ (set-coding-priority-list '(utf-8)) \ (set-coding-category-system 'utf-8 'utf-8)" works for me. Have you tried -vanilla? Is it possible you have some relevant stuff in the site init file? -- Institute of Policy and Planning Sciences http://turnbull.sk.tsukuba.ac.jp University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN My nostalgia for Icon makes me forget about any of the bad things. I don't have much nostalgia for Perl, so its faults I remember. Scott Gilbert c.l.py