utf8ByteSequenceLength: reject lead bytes > 0xF4

jedisct1 · jedisct1 · commit 8c078a6e6935 · 2025-11-26T23:01:23.000+01:00
UTF-8 start bytes in the range [0xF5 .. 0xFF] are disallowed as they
would encode code points above U+10FFFF
diff --git a/lib/std/unicode.zig b/lib/std/unicode.zig
@@ -31,7 +31,7 @@ pub fn utf8ByteSequenceLength(first_byte: u8) !u3 {
         0b0000_0000...0b0111_1111 => 1,
         0b1100_0000...0b1101_1111 => 2,
         0b1110_0000...0b1110_1111 => 3,
-        0b1111_0000...0b1111_0111 => 4,
+        0b1111_0000...0b1111_0100 => 4,
         else => error.Utf8InvalidStartByte,
     };
 }

Original file line number	Diff line number	Diff line change
`@@ -31,7 +31,7 @@ pub fn utf8ByteSequenceLength(first_byte: u8) !u3 {`
`31`	`31`	`0b0000_0000...0b0111_1111 => 1,`
`32`	`32`	`0b1100_0000...0b1101_1111 => 2,`
`33`	`33`	`0b1110_0000...0b1110_1111 => 3,`
`34`		`- 0b1111_0000...0b1111_0111 => 4,`
	`34`	`+ 0b1111_0000...0b1111_0100 => 4,`
`35`	`35`	`else => error.Utf8InvalidStartByte,`
`36`	`36`	`};`
`37`	`37`	`}`