Bug 15887 – invalid char/wchar/dchar values are inconsistently accepted/rejected

---- void main() { char c = 0xFF; /* accepted */ wchar wc = 0xFF_FF; /* accepted */ dchar dc = 0xFF_FF_FF_FF; /* Error: cannot implicitly convert expression (4294967295u) of type uint to dchar */ string s = x"FF"; /* accepted */ wstring ws = x"FF FF"w; /* Error: Outside Unicode code space */ dstring ds = x"FF FF FF FF"d; /* Error: Outside Unicode code space */ } ---- Should all be accepted. The spec says about hex strings: "The hex data need not form valid UTF characters." - https://dlang.org/spec/lex.html

hex strings are about to be removed, so this is a WONTFIX.

Unicode code points are limited to 16 planes, so it's unclear if 0xFF_FF_FF_FF should be accepted or not. Some think that it should be seen as a possible value, because it is possible. As to whether it should be allowed as a literal... Should invalid utf8 text in string literals be accepted too?