Unicode I/O

Hi!

Currently, the reader/writer/lexer use plain (u)chars to represent
characters. We will eventually want to change that; I guess it's ok in
the meantime to just assume that no non-ISO8859-1 chars will enter our
system and code like that:

invariant (!scm_string_length (&length, arena, str));
extracted = (t_uchar) lim_malloc (length);
invariant (!scm_extract_string8_n (extracted, &length, uni_iso8859_1, 
                                   arena, str));
/* do output here */

When we rewrite the reader and the write functions to support unicode,
it would be nice to have basic unicode-supporting functions in
hackerlab, e.g. safe_printfmt_uni(). Is something like this planned?
Good/bad idea? Would this operate on ustr or udstr layer?

Regards, Rotty
-- 
Andreas Rottmann         | Rotty@ICQ      | 118634484@ICQ | [email protected]
http://yi.org/rotty      | GnuPG Key: http://yi.org/rotty/gpg.asc
Fingerprint              | DFB4 4EB4 78A4 5EEE 6219  F228 F92F CFC5 01FD 5B62

Software Patents: Where do you want to stifle inovation today?