3 ms·

If you changed UTF-16 to UTF-32 or UCS-4 I'd support it. I think there are already implementations that use the replacement character for all "impossible" code

If you changed UTF-16 to UTF-32 or UCS-4 I'd support it. I think there are already implementations that use the replacement character for all "impossible" codes.

hsivonen 6y ago
What’s your use case for UTF-32?
- DougGwyn 6y ago
  There are several multibyte character manipulations that are easier if there is a uniform-sized encoding (wchar_t).