Re: Data.ByteString candidate 3

Ketil Malde <ketil+haskell@xxxxxxxxx> writes:

> I've now added a Latin1 module, that works like Char8, but where

And now there's an ASCII module, which instead of storing bytes > 127 in
the latin1 range, puts them in the "private" area of 0xF000..0xF07F.
This way, they won't be affected by other Char functions depending on
case etc.  Packing Chars outside of this area is still an error
(i.e. no 8 bit truncation)

IMHO this is the correct way to provide a Char interface to ASCII
(albeit at a performance penalty), but I simply can't wait to hear
what other people have to say about the matter. :-) 

-k
-- 
If I haven't seen further, it is by standing in the footprints of giants