On 5/26/06, Matthew Flatt <xxxxxx@cs.utah.edu> wrote:
>
> Straightforward additions
> -------------------------
>
>     * `char-general-category', which accepts a character and returns one
>        of 'lu, 'li, ...
>     * `string-normalize-nfd', `string-normalize-nfkd,
>       `string-normalize-nfc', and `string-normalize-nfkc', which each
>       accept a string and produce its normalization according to normal
>       form D, KD, C, or KC, respectively.

I wouldn't consider these straightforward because they remove the
option of a Scheme implementation to keep all strings internally in
the same normalization form.  For the extra work of making all
string primitives construct and retain a single normalization form,
you relieve the user from the burden of ever having to worry about
any normalization issues (although you then introduce round-trip
issues with external string sources).

What about a not-necessarily-Unicode-specific STRING-NORMALIZE that
simply converts to the implementation's preferred normal form,
possibly returning the original string?

Regardless, since all of these procedures require large tables can
we assume they are part of a library and not part of the core
language?

--
Alex