[Haskell-cafe] Writing binary files?

[David Menendez]

Udo Stenzel writes:

> I'd like to see the following:
> 
> - Duplicate the IO library.  The duplicate should work with [Byte]
>   everywhere where the old library uses String.  Byte is some suitable
>   unsigned integer, on most (all?) platforms this will be Word8
> 
> - Provide an explicit conversion between encodings.  A simple
>   conversion of type [Word8] -> String would suit me, iconv would
>   provide all that is needed.

I like this idea, but I say there should be a bit-oriented layer beneath
everything. Reading a text file, for example, can be interpreted at
three levels:

    1. bit string
    2. octet string
    3. character string

Most of the work involves moving between 2 and 3 but there's no reason
to move from 1 to 2 automatically, either. In any case, if the functions
compose well, people who only care about octets don't have to see the
bit-string level.

For example, this progression makes more sense for UTF-16 files:

    1. bit string
    2. Word16 string
    3. character string
-- 
David Menendez <zednenem at psualum.com> | "In this house, we obey the laws
<http://www.eyrie.org/~zednenem>      |        of thermodynamics!"