Re: Proposal: _ctype_ table bitwidth change

[KAMADA Ken'ichi <kamada%nanohz.org@localhost>]

At Wed, 23 Mar 2011 13:32:40 -0400 (EDT), der Mouse wrote:
> 
> > NO-BREAK SPACE, which is 0xC2 0xA0 in en_US.UTF-8, obviously falls
> > into the some-codepoints-that-can't-fit-in-unsigned-char category.
> 
> It's certainly not obvious to me.  That is not the codepoint but the
> encoding of the codepoint.  The codepoint is the abstract integer 160,
> which _does_ fit into unsigned char.

Hm, I carelessly quoted "codepoint" from your previous message,
sorry about that.  I should've said
"NO-BREAK SPACE, which is 0xC2 0xA0 in en_US.UTF-8, obviously falls into
the some-characters-that-can't-fit-in-unsigned-char category".

> I maintain that is*() must be passed the codepoint - the "character".

You might forget that some character encoding scheme can encode
multiple coded character sets.  It is not always possible to convert
characters into unoverwrapping sets of "codepoints".

Ken