Font locking of some Unicode characters fails

[Roel van Dijk]

I couldn't fully register at
http://trac.haskell.org/haskellmode-emacs/ because it failed to
send a verification email. So I'll post here.

Of the following 2 lines only the first gets the colour
associated with types and constructors

  type Z = Integer
  type ℤ = Integer

Both are perfectly legal Haskell. The only difference I could
find is in the Unicode properties of those characters (found via
Emacs's describe-char):

Z
category: .:Base, a:ASCII, l:Latin, r:Roman
general-category: Lu (Letter, Uppercase)

ℤ
category: .:Base
general-category: Lu (Letter, Uppercase)

Does the font locking code inspect a character's category?
Perhaps it should use the general-category instead. Or maybe it's
a (simple?) regex problem.

I tried to look into the code myself but I'm not very well versed
in lisp let alone Emacs lisp. And the regexes scare me :-)

Regards,
Roel