>>>>> "Henry" == Henry Baker <[email protected]> writes:
Henry> UTF8 isn't 'critical', but not having it means that
Henry> interfacing with file systems & documents that utilize UTF8
Henry> is quite difficult.
I think it would be nice too if gcl supported some unicode. I think
all that is really needed would be to increase the size of a base-char
from 8 bits to 16 (or more)[1]. Then a combination of cl-unicode[2] and
babel[3] will give all of the necessary features. Don't know if gcl
can compile those libraries or not, though.
A utf-8 encoder/decoder is pretty simple so you don't even need
cl-unicode or babel to get what most people want: a way to read and
write utf-8 strings.
Footnotes:
[1] It seems most free lisps have base-strings (8 bits) and strings
(32 bits). Many commercial lisps use just 16 bits, I think;
cmucl uses 16 bits. 16 bits isn't enough to hold all unicode
characters, but all of the important languages fit in 16 bits.
[2] http://weitz.de/cl-unicode/
[3] http://common-lisp.net/project/babel/
_______________________________________________
Gcl-devel mailing list
[email protected]
https://lists.gnu.org/mailman/listinfo/gcl-devel
