Maybe something where you can pick the CharPointer type, instead of sticking to UTF-8 only?
Also, on windows, you should probably utilise Arial Unicode MS (if it exists) as the font in the first textbox to be able to display all the glyphs; it has the largest codepoint coverage that I know of on the OS.
The whole point is that ASCII-encoded UTF8 is the only way to reliably put unicode strings in source code. UTF16 and 32 are fine as encodings, but you can't embed them in source files without all kinds of endianness/text-editor/compiler-specific issues!
Good idea about the font though.
Ahhh, okay - I guess that means reading from streams/files... Thanks