On 7/21/05, John.Cowan <xxxxxx@reutershealth.com> wrote:
>
> I think this SRFI should have char-titlecase and string-titlecase
> functions.  The former maps a lowercase or uppercase character into
> its titlecase equivalent, and leaves all other characters alone.
> The latter titlecases its first character and lowercases all the rest;
> it's trivial to define string-titlecase given char-titlecase, but it's
> IMHO worth having for symmetry.

The point several people have been trying to make is that it's *not*
trivial to define string-titlecase given char-titlecase.

I'm not arguing against providing procedures that people might use
incorrectly, there's no point in that.  I'm debating the utility of
CHAR-*CASE providing limited Unicode mappings at all.  Some
people have suggested this is useful, but no one has given any
examples of how they would use it.

My Unicode library does have procedures that behave like this
called CHAR-*CASE-SINGLE, but these are not exported, because
so far the only use I've found for them is to implement STRING-*CASE.

What do you consider these useful for?

--
Alex