Thu, 29 Aug 2002 10:15:32 +0100, Simon Marlow pisze:

I think we should start with lazy steam encoders/decoders, and later add monadic versions and ones that work with arrays. For GHC, we'll want some really low-level ones that work on ByteArray# for use in the I/O libraries, but that can come later (maybe use libiconv?).

Have a look at http://sf.net/projects/qforeign I have some recoding machinery (I don't like it but couldn't think how to to it better) including iconv, and Unicode character properties (lib/, unidata/, charsets/). It rotted a bit, I had unresolved issues with packaging for newer ghcs. -- __("< Marcin Kowalczyk \__/ qrczak@knm.org.pl ^^ http://qrnik.knm.org.pl/~qrczak/