Okay, now that we've got the bytecode-visible stuff specified, I want
to spec the internals some, and start getting things migrated over to
it. (This should allow us to make ICU optional as well, for folks
that only want ASCII/Latin-x/EBCDIC enabled)

Once again, we're going with vtables, like the strings originally
had. Each string has two vtable pointers, one for encoding and one
for charset. Encodings and charsets'll be loadable libraries, in the
encodings/ and charset/ directories.

The encoding vtable needs to handle get/set codepoint, get/set byte,
and transform to another encoding. I don't think there's anything
else, but I could be wrong there.

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.

I think those are it, but before we nail them down I'd like to have
folks squint at this a bit so we can make sure it's right. When it is
we can define the API directly and start implementing it.
--
                                Dan

--------------------------------------it's like this-------------------
Dan Sugalski                          even samurai
d...@sidhe.org                         have teddy bears and even
                                       teddy bears get drunk