------- Comment #8 from pinskia at gcc dot gnu dot org 2006-05-28 20:03 -------
(In reply to comment #7)
> (In reply to comment #6)
> > Have you read 4.3.2.1.1? This defines the requirements on the
> > collating of the default character type. It also defines the
> > requirements on LGT, LGE, LLE, and LLT, which is why we use the
> > ascii_table in simplify.c.
>
> OK, I see. I think we don't have users around here using non ASCII-based
> charsets, and I also think other parts of the front-end and library would fall
> apart on such systems, but anyway, that's not the point.
GCC does support other charatersets besides ASCII ( the IBM one is an example):
the middle-end does something like:
target_newline = lang_hooks.to_target_charset ('\n');
target_percent = lang_hooks.to_target_charset ('%');
target_c = lang_hooks.to_target_charset ('c');
target_s = lang_hooks.to_target_charset ('s');
To find %s for folding of printf, etc.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27715
