utf8.rs

fn utf8_len(start: u8) -> usize {
    match start {
        0b0000_0000..=0b0111_1111 => 1,
        0b1100_0000..=0b1101_1111 => 2,
        0b1110_0000..=0b1110_1111 => 3,
        0b1111_0000..=0b1111_0111 => 4,
        _ => 0,
    }
}
enum Utf8Error {
    InvalidStart,
    InvalidContinuation,
    InvalidCodepoint,
    TruncatedCodepoint,
}
fn utf8_split(bytes: &[u8]) -> Option<Result<(char, &[u8]), Utf8Error>> {
    if bytes.is_empty() {
        return None;
    }
    let head = bytes[0];
    let len = utf8_len(head);
    if len == 0 {
        return Some(Err(Utf8Error::InvalidStart));
    }
    if bytes.len() < len {
        return Some(Err(Utf8Error::TruncatedCodepoint));
    }
    let mut chr = (head << len >> len) as u32;
    for &byte in &bytes[1..len] {
        if byte & 0b1100_0000 != 0b1000_0000 {
            return Some(Err(Utf8Error::InvalidContinuation));
        }
        chr = (chr << 6) | (byte & 0b0011_1111) as u32;
    }
    match char::from_u32(chr) {
        Some(chr) => Some(Ok((chr, &bytes[len..]))),
        None => Some(Err(Utf8Error::InvalidCodepoint)),
    }
}

fn main() {
    let mut bytes = "привіт, світ! ❤️🧡💛💚💙💜".as_bytes();
    while let Some(Ok((chr, tail))) = utf8_split(bytes) {
        bytes = tail;
        print!("'{}', ", chr);
    }
}