Would the following be a correct way to determine whether there's a problem?
* First call setlocale(LC_CTYPE, "en_US.UTF-8")
* Next feed the UTF-8 string representation of every Unicode codepoint one at a time to mbstowcs() and ensure that the output for each is a wchar_t string of length one
* If all input codepoints numerically match the output wchar_t UTF-32 code units, then the implementation is officially good, and should define __STDC_ISO_10646__?
* First call setlocale(LC_CTYPE, "en_US.UTF-8")
* Next feed the UTF-8 string representation of every Unicode codepoint one at a time to mbstowcs() and ensure that the output for each is a wchar_t string of length one
* If all input codepoints numerically match the output wchar_t UTF-32 code units, then the implementation is officially good, and should define __STDC_ISO_10646__?