harec

commit 223e8235a93e86b6047a4a3d1f1bf253b2f534f8
parent f674e4cac551546cfee5dc1521f6f013e6b44522
Author: Jose Lombera <jose@lombera.dev>
Date:   Sun, 11 Sep 2022 20:03:04 -0500

utf8: utf8_decode: validate multi-byte codepoints

Validate byte mask in multi-byte codepoints.

Signed-off-by: Jose Lombera <jose@lombera.dev>

Diffstat:
M
src/utf8.c
 | 
10
++++++++--

1 file changed, 8 insertions(+), 2 deletions(-)
diff --git a/src/utf8.c b/src/utf8.c
@@ -59,9 +59,15 @@ utf8_decode(const char **char_str)
 	cp = **s & mask;
 	++*s;
 	while (--size) {
-		cp <<= 6;
-		cp |= **s & 0x3f;
+		uint8_t c = **s;
+
 		++*s;
+
+		if ((c >> 6) != 0x02)
+			return UTF8_INVALID;
+
+		cp <<= 6;
+		cp |= c & 0x3f;
 	}
 	return cp;
 }