[AD] unicode state

["Vincent Penquerc'h" <vincent@xxxxxxxxxx>]

Hi,

Does anybody here use unicode with either 16 bit encoding or utf8
with non ascii and non codepage characters (eg > 255) ?
Annie is trying to fix the buffer overflow risks from Allegro, and
had to deal with a slew of unicode format conversions, since a
good part of (constant size) string buffers are used in those
conversions.
Second question: I'm told that quite a few parts of the code wrongly
translate to or from the wrong format (eg call ascii->unicode on an
already unicode string, or vice versa). I'd like to know if this
is actually used, or at least sometimes tested.

Thanks

-- 
Lyrian