> Another thought, may be ToNonLDH is a more accurate name. No, because the output of ToUnicode sometimes is LDH.
> I think ToNonAscii is much better than ToUnicode. No, because the output of ToUnicode sometimes is ASCII. The only thing we can say for sure about the output of ToUnicode is that it is never ACE. (Well, and that it's Unicode, but all strings in the IDNA spec are Unicode. Even ASCII strings are Unicode, because Unicode is an extension of ASCII.) The only names I can think of that might be better than ToUnicode are ToNonACE and UnACE. Since no one has expressed support for either of those, I guess we're sticking with ToUnicode. Keld J�rn Simonsen <[EMAIL PROTECTED]> wrote: > I don't think that ToUnicode is the correct name, as it is really > 10646 that it is defined for. The difference between ISO-10646 and Unicode is not relevant here. Both use the same table that maps between code points and characters. The difference (as I understand it) is that Unicode defines lots of additional information about the characters and how to handle them, while ISO-10646 is little more than a table and some encoding forms. AMC
