MULE rant...

[Aidan Kehoe]

 Ar an t-ochtú lá is fiche de mí Lúnasa, scríobh David Kastrup: 

 > [... much complaining about Mule-UCS.]
 >
 > I have a Unicode character number and want the corresponding XEmacs
 > string.  That is all for now.  I am prepared that the menu display
 > will probably be wrong.  But users can configure it on or off, and
 > they can complain when something wrong appears and perhaps you care
 > more when they complain instead of myself.  But if I don't even get
 > told how to convert the Unicode code point into a simple 1-byte
 > string, I can't even get this process started.

This should work in 21.4 and 21.5. 21.4 Mule-UCS _is_ broken, hackish, and
unsupported, though--it will probably save you a lot of headaches if you
just state that you don't support it for this.

;; More verbose than it needs to be, because it's checking its first
;; argument, which won't be anything other than 'ucs for current code in the
;; wild. 
(unless (fboundp 'decode-char)
  (cond ((fboundp 'unicode-to-char)
	 (defmacro decode-char (quote-ucs code)
	   `(prog2 (assert (eq (quote ucs) ,quote-ucs))
	     (unicode-to-char ,code))))
	((and (locate-library "un-define") 
	      (require 'un-define))
	 (defmacro decode-char (quote-ucs code)
	   `(prog2 (assert (eq (quote ucs) ,quote-ucs))
	     (unicode-to-char ,code))))))

XEmacs uses unicode-to-char and/or ucs-to-char there. (Myself, I'm a
little unclear as to the utility of generalising that function beyond
Unicode; make-char would seem to do the job for any other large character
sets that actually exist in the world.)

-- 
Like the early Christians, Marx expected the millennium very soon; like
their successors, his have been disappointed--once more, the world has shown
itself recalcitrant to a tidy formula embodying the hopes of some section of
mankind. (Russell)