Yeah, but I believe that at least Unicode has one of the four that they suggest be used for non-locale specific comparisons (canonical decomposition form). So pick that one for the core and provide the others (if necessary) as library functions.
-- Mark Biggar [EMAIL PROTECTED] [EMAIL PROTECTED] > At 4:30 PM +0000 6/16/04, [EMAIL PROTECTED] wrote: > >Do we want a Normalization function here as well. If you have that > >you can use a binary compare (at least for eq/ne). > > Yeah, we probably do. The question is always "Which normalization" > since there are at least four for Unicode and two for ISO-2022. (Or > something like that--I don't think I remembered the ISO number right) > > > > >> The charset vtable needs to handle get/set grapheme, get/set > >> substring, up/down/titlecase, and (possibly) comparison. Charsets > >> also have a separate grapheme classification requirement (for > >> regexes) but we'll put that off for now. > > > -- > Dan > > --------------------------------------it's like this------------------- > Dan Sugalski even samurai > [EMAIL PROTECTED] have teddy bears and even > teddy bears get drunk
