2 * Copyright (c) 2009, 2010 Petri Lehtinen <petri@digip.org>
4 * Jansson is free software; you can redistribute it and/or modify
5 * it under the terms of the MIT license. See LICENSE for details.
11 int utf8_encode(int32_t codepoint, char *buffer, int *size)
15 else if(codepoint < 0x80)
17 buffer[0] = (char)codepoint;
20 else if(codepoint < 0x800)
22 buffer[0] = 0xC0 + ((codepoint & 0x7C0) >> 6);
23 buffer[1] = 0x80 + ((codepoint & 0x03F));
26 else if(codepoint < 0x10000)
28 buffer[0] = 0xE0 + ((codepoint & 0xF000) >> 12);
29 buffer[1] = 0x80 + ((codepoint & 0x0FC0) >> 6);
30 buffer[2] = 0x80 + ((codepoint & 0x003F));
33 else if(codepoint <= 0x10FFFF)
35 buffer[0] = 0xF0 + ((codepoint & 0x1C0000) >> 18);
36 buffer[1] = 0x80 + ((codepoint & 0x03F000) >> 12);
37 buffer[2] = 0x80 + ((codepoint & 0x000FC0) >> 6);
38 buffer[3] = 0x80 + ((codepoint & 0x00003F));
47 int utf8_check_first(char byte)
49 unsigned char u = (unsigned char)byte;
54 if(0x80 <= u && u <= 0xBF) {
55 /* second, third or fourth byte of a multi-byte
56 sequence, i.e. a "continuation byte" */
59 else if(u == 0xC0 || u == 0xC1) {
60 /* overlong encoding of an ASCII byte */
63 else if(0xC2 <= u && u <= 0xDF) {
68 else if(0xE0 <= u && u <= 0xEF) {
72 else if(0xF0 <= u && u <= 0xF4) {
76 else { /* u >= 0xF5 */
77 /* Restricted (start of 4-, 5- or 6-byte sequence) or invalid
83 int utf8_check_full(const char *buffer, int size, int32_t *codepoint)
87 unsigned char u = (unsigned char)buffer[0];
104 for(i = 1; i < size; i++)
106 u = (unsigned char)buffer[i];
108 if(u < 0x80 || u > 0xBF) {
109 /* not a continuation byte */
113 value = (value << 6) + (u & 0x3F);
116 if(value > 0x10FFFF) {
117 /* not in Unicode range */
121 else if(0xD800 <= value && value <= 0xDFFF) {
122 /* invalid code point (UTF-16 surrogate halves) */
126 else if((size == 2 && value < 0x80) ||
127 (size == 3 && value < 0x800) ||
128 (size == 4 && value < 0x10000)) {
129 /* overlong encoding */
139 const char *utf8_iterate(const char *buffer, int32_t *codepoint)
147 count = utf8_check_first(buffer[0]);
152 value = (unsigned char)buffer[0];
155 if(!utf8_check_full(buffer, count, &value))
162 return buffer + count;
165 int utf8_check_string(const char *string, int length)
170 length = strlen(string);
172 for(i = 0; i < length; i++)
174 int count = utf8_check_first(string[i]);
179 if(i + count > length)
182 if(!utf8_check_full(&string[i], count, NULL))