Function utf8CountCodepoints [src]

Returns the length of a supplied UTF-8 string literal in terms of unicode codepoints.

Prototype

pub fn utf8CountCodepoints(s: []const u8) !usize

Parameters

s: []const u8

Source

  pub fn utf8CountCodepoints(s: []const u8) !usize {
    var len: usize = 0;

    const N = @sizeOf(usize);
    const MASK = 0x80 * (std.math.maxInt(usize) / 0xff);

    var i: usize = 0;
    while (i < s.len) {
        // Fast path for ASCII sequences
        while (i + N <= s.len) : (i += N) {
            const v = mem.readInt(usize, s[i..][0..N], native_endian);
            if (v & MASK != 0) break;
            len += N;
        }

        if (i < s.len) {
            const n = try utf8ByteSequenceLength(s[i]);
            if (i + n > s.len) return error.TruncatedInput;

            switch (n) {
                1 => {}, // ASCII, no validation needed
                else => _ = try utf8Decode(s[i..][0..n]),
            }

            i += n;
            len += 1;
        }
    }

    return len;
}  