Simon Josefsson wrote: > > unicase case folding > > Is this NFKC?
I didn't consider normalization functions so far, on the premise that text being exchanged between applications is assumed to be precomposed. But their presence in Java APIs and the IDN case shows that APIs for composing and decomposing are probably needed. > One feature request: make it possible to use a particular version of > Unicode. The IDN stuff requires that Unicode 3.2 is used, and will > return incorrect results if used with a newer Unicode version. Would > that be problematic to support? You're right; this is a requirement that is good to know in advance. Thanks for mentioning it. Bruno
