Re: base-char bug with SBCL unicode

On Tue, 28 Dec 2004, Robert Strandh wrote:

> Today, I tripped over a bug in which some (probably inlined) call to
> char->card8 blew up for non-ascii characters.  Christophe Rhodes seems
> to think it is the use of base-char in this code snippet in the file
> dependent.lisp:
>
> Christophe thinks (sb-kernel:character-set ((0 . 255))) would be more
> appropriate for SBCL.

Actually, I think I've changed my mind: absent any indication to the 
contrary, I would expect char->card8 to work strictly for ASCII; certainly 
that's what the CLX manual leads me to expect.  Possibly it should work 
for the first 255 code points of Unicode (aka iso-8859-1) but it's not 
clear to me that you should be relying on this.

I would recommend instead either performing the conversion to iso-8859-1 
character codes yourself, or else using char-code directly (on the 
understanding that SBCL, and indeed most implementations around now, is 
unlikely to be eccentric enough to define character codes differently from 
Unicode).

Cheers,

Christophe

_______________________________________________
Portable-clx mailing list
[email protected]
http://lists.metacircles.com/cgi-bin/mailman/listinfo/portable-clx
cvs -d :ext:cvs.telent.net:/usr/local/src/cvs co clx  # over ssh
cvs -d :pserver:[email protected]:/cvs co clx  # anonymous