On Thu, Sep 14, 2017 at 4:33 AM, Shiro Kawai <xxxxxx@gmail.com> wrote:

I'm ok with making unicode-related procedures optional, but would it be better to state how to detect the support (or the lack) of those optional procedures portably, using feature conditionals?

An excellent point, but I don't know exactly how they should be divided up. Here are the optional procedures: can you suggest a many-to-one mapping to features?

string->utf8 string->utf16 string->utf16le string->utf16be

utf8->string utf16->string utf16le->string utf16be->string

string-upcase string-downcase string-foldcase

string-normalize-nfc string-normalize-nfkc string-normalize-nfd string-normalize-nfkd

Note that the utf8 procedures are in R7RS but the utf16 (all variants) are not.

John Cowan http://vrici.lojban.org/~cowan xxxxxx@ccil.org

We call nothing profound that is not wittily expressed.

--Northrop Frye (improved)