3 int utf8_check_first(char byte)
5 unsigned char u = (unsigned char)byte;
10 if(0x80 <= u && u <= 0xBF) {
11 /* second, third or fourth byte of a multi-byte
12 sequence, i.e. a "continuation byte" */
15 else if(u == 0xC0 || u == 0xC1) {
16 /* overlong encoding of an ASCII byte */
19 else if(0xC2 <= u && u <= 0xDF) {
24 else if(0xE0 <= u && u <= 0xEF) {
28 else if(0xF0 <= u && u <= 0xF4) {
32 else { /* u >= 0xF5 */
33 /* Restricted (start of 4-, 5- or 6-byte sequence) or invalid
39 int utf8_check_full(const char *buffer, int size)
42 unsigned char u = (unsigned char)buffer[0];
59 for(i = 1; i < size; i++)
61 u = (unsigned char)buffer[i];
63 if(u < 0x80 || u > 0xBF) {
64 /* not a continuation byte */
68 value = (value << 6) + (u & 0x3F);
71 if(value > 0x10FFFF) {
72 /* not in Unicode range */
76 else if(0xD800 <= value && value <= 0xDFFF) {
77 /* invalid code point (UTF-16 surrogate halves) */
81 else if((size == 2 && value < 0x80) ||
82 (size == 3 && value < 0x800) ||
83 (size == 4 && value < 0x10000)) {
84 /* overlong encoding */
91 int utf8_check_string(const char *string, int length)
96 length = strlen(string);
98 for(i = 0; i < length; i++)
100 int count = utf8_check_first(string[i]);
105 if(i + count > length)
108 if(!utf8_check_full(&string[i], count))