1 files changed, 123 insertions, 0 deletions
diff --git a/extras/volume_id/lib/util.c b/extras/volume_id/lib/util.c
index 54d9fd0d80..eaaececadb 100644
--- a/extras/volume_id/lib/util.c
+++ b/extras/volume_id/lib/util.c
@@ -28,6 +28,129 @@
 #include "libvolume_id.h"
 #include "util.h"
 
+/* count of characters used to encode one unicode char */
+static int utf8_encoded_expected_len(const char *str)
+{
+	unsigned char c = (unsigned char)str[0];
+
+	if (c < 0x80)
+		return 1;
+	if ((c & 0xe0) == 0xc0)
+		return 2;
+	if ((c & 0xf0) == 0xe0)
+		return 3;
+	if ((c & 0xf8) == 0xf0)
+		return 4;
+	if ((c & 0xfc) == 0xf8)
+		return 5;
+	if ((c & 0xfe) == 0xfc)
+		return 6;
+	return 0;
+}
+
+/* decode one unicode char */
+static int utf8_encoded_to_unichar(const char *str)
+{
+	int unichar;
+	int len;
+	int i;
+
+	len = utf8_encoded_expected_len(str);
+	switch (len) {
+	case 1:
+		return (int)str[0];
+	case 2:
+		unichar = str[0] & 0x1f;
+		break;
+	case 3:
+		unichar = (int)str[0] & 0x0f;
+		break;
+	case 4:
+		unichar = (int)str[0] & 0x07;
+		break;
+	case 5:
+		unichar = (int)str[0] & 0x03;
+		break;
+	case 6:
+		unichar = (int)str[0] & 0x01;
+		break;
+	default:
+		return -1;
+	}
+
+	for (i = 1; i < len; i++) {
+		if (((int)str[i] & 0xc0) != 0x80)
+			return -1;
+		unichar <<= 6;
+		unichar |= (int)str[i] & 0x3f;
+	}
+
+	return unichar;
+}
+
+/* expected size used to encode one unicode char */
+static int utf8_unichar_to_encoded_len(int unichar)
+{
+	if (unichar < 0x80)
+		return 1;
+	if (unichar < 0x800)
+		return 2;
+	if (unichar < 0x10000)
+		return 3;
+	if (unichar < 0x200000)
+		return 4;
+	if (unichar < 0x4000000)
+		return 5;
+	return 6;
+}
+
+/* check if unicode char has a valid numeric range */
+static int utf8_unichar_valid_range(int unichar)
+{
+	if (unichar > 0x10ffff)
+		return 0;
+	if ((unichar & 0xfffff800) == 0xd800)
+		return 0;
+	if ((unichar > 0xfdcf) && (unichar < 0xfdf0))
+		return 0;
+	if ((unichar & 0xffff) == 0xffff)
+		return 0;
+	return 1;
+}
+
+/* validate one encoded unicode char and return its length */
+int volume_id_utf8_encoded_valid_unichar(const char *str)
+{
+	int len;
+	int unichar;
+	int i;
+
+	len = utf8_encoded_expected_len(str);
+	if (len == 0)
+		return -1;
+
+	/* ascii is valid */
+	if (len == 1)
+		return 1;
+
+	/* check if expected encoded chars are available */
+	for (i = 0; i < len; i++)
+		if ((str[i] & 0x80) != 0x80)
+			return -1;
+
+	unichar = utf8_encoded_to_unichar(str);
+
+	/* check if encoded length matches encoded value */
+	if (utf8_unichar_to_encoded_len(unichar) != len)
+		return -1;
+
+	/* check if value has valid range */
+	if (!utf8_unichar_valid_range(unichar))
+		return -1;
+
+	return len;
+}
+
 void volume_id_set_unicode16(char *str, size_t len, const uint8_t *buf, enum endian endianess, size_t count)
 {
 	unsigned int i, j;