>>! In D5385#46, @jroelofs wrote:
> @joerg For the ctype<char> ones, I think that's fine, but I don't think that 
> works for ctype<wchar_t> because classic_table has only at least 256 entries. 
> That would get isascii out of include/__locale, and we can leave the ones in 
> src/locale.cpp, which solves this messiness.

`ctype<T>::classic_table()` is only defined for `ctype<char>`. We do perform 
these same checks in `ctype<wchar_t>::do_is()` though... I'm not sure what the 
rules are for classifying wide characters... Shouldn't we just be passing these 
down to the C library's `iswupper()` and friends as we do in 
`ctype_byname<wchar_t>::do_is()`?

http://reviews.llvm.org/D5385



_______________________________________________
cfe-commits mailing list
[email protected]
http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits

Reply via email to