Personally, I'm not sure that Perl 6 even needs to worry about any of this. I
think as long as Perl can manipulate the characters without screwing them up in
any way, that should be fine. As far as support for locale specifics, maybe it
would just be best to do what Perl does now, but allow overloading of the
locale-specific functionality. I think that's actually more than most other
implementations do.
Ok, that said, the way I see it (and I'm probably wrong), is that Perl may need
to know the following things about each character in Unicode, based upon locale
(correct me if I'm wrong):
Is this Uppercase?
Is this Lowercase (is this 'half-digits', as Hong mentioned?)
(if 'Caseless' needed, just "!Upper && !Lower"?)
Is this Punctuation?
Is this a digit?
Is this a word character?
Is this Whitespace?
Maps to opposite case letter X (only upper & lower?)
Previous character in order
Next character in order
(are there languages with character precedence rules that override sort
orders?)
---------
Does Perl also need to know (in regard to locale) print direction? Are there
'special-case' rules (like in English 'y' can be a vowel, but I'm thinking more
in the sense of the above properties)?
Grant M.
