four-byte uncode characters confuse `'`

[steveklabnik]

This Rust program:

fn main() {
    let len = 'ஶ்ரீ'.len_utf8();
}

contains TAMIL SYLLABLE SHRII (śrī), aka U+0BB6 U+0BCD U+0BB0 U+0BC0. When trying to compile this program, I get this error:

2:20: 2:22 error: unterminated character constant: '.
2     let len = 'ஶ்ரீ'.len_utf8();

I know that it isn't a copy-paste issue, because I used vim's C-V u to type in the four code points manually.

[arielb1]

Aren't char-s Unicode scalar values?

[steveklabnik]

@arielb1 yes. am I doing something wrong here? I am bad at encodings, so this is likely.

[arielb1]

@steveklabnik

Unicode scalar value != character.

[steveklabnik]

Yes, but it's four bytes, no?

[steveklabnik]

Ahhh this isn't actually four bytes. Sigh. Thanks.

[steveklabnik]

(basically, I thought that those four things were four bytes, but they're four codepoints themselves)

[steveklabnik]

Actually, I am re-opening, because this diagnostic message is really bad. It should say that you're putting something that's larger than a single USV into a char literal.