纯C实现unicode-utf8互转

 /************************************************************************/ /* Author: NadOo (nadoo@21cn.com) Blog: http://blog.csdn.net/nadoo Reference: http://blog.csdn.net/lovekatherine/archive/2007/08/30/1765903.aspx Testing Tool: http://www.hanzify.org/?Go=Show::List&ID=9627 */ /************************************************************************/ #include <stdio.h> #include <string.h> #include <malloc.h> #include <memory.h> #ifdef WIN32 #define uint8_t unsigned __int8 #define uint16_t unsigned __int16 #define uint32_t unsigned __int32 #define uint64_t unsigned __int64 #define int8_t __int8 #define int16_t __int16 #define int32_t __int32 #endif int unicode_to_utf8(uint16_t *in, int insize, uint8_t **out) { int i = 0; int outsize = 0; int charscount = 0; uint8_t *result = NULL; uint8_t *tmp = NULL; charscount = insize / sizeof(uint16_t); result = (uint8_t *)malloc(charscount * 3 + 1); memset(result, 0, charscount * 3 + 1); tmp = result; for (i = 0; i < charscount; i++) { uint16_t unicode = in[i]; if (unicode >= 0x0000 && unicode <= 0x007f) { *tmp = (uint8_t)unicode; tmp += 1; outsize += 1; } else if (unicode >= 0x0080 && unicode <= 0x07ff) { *tmp = 0xc0 | (unicode >> 6); tmp += 1; *tmp = 0x80 | (unicode & (0xff >> 2)); tmp += 1; outsize += 2; } else if (unicode >= 0x0800 && unicode <= 0xffff) { *tmp = 0xe0 | (unicode >> 12); tmp += 1; *tmp = 0x80 | (unicode >> 6 & 0x00ff); tmp += 1; *tmp = 0x80 | (unicode & (0xff >> 2)); tmp += 1; outsize += 3; } } *tmp = '/0'; *out = result; return 0; } int utf8_to_unicode(uint8_t *in, uint16_t **out, int *outsize) { uint8_t *p = in; uint16_t *result = NULL; int resultsize = 0; uint8_t *tmp = NULL; result = (uint16_t *)malloc(strlen(in) * 2 + 2); /* should be enough */ memset(result, 0, strlen(in) * 2 + 2); tmp = (uint8_t *)result; while(*p) { if (*p >= 0x00 && *p <= 0x7f) { *tmp = *p; tmp++; *tmp = '/0'; resultsize += 2; } else if ((*p & (0xff << 5))== 0xc0) { uint16_t t = 0; uint8_t t1 = 0; uint8_t t2 = 0; t1 = *p & (0xff >> 3); p++; t2 = *p & (0xff >> 2); *tmp = t2 | ((t1 & (0xff >> 6)) << 6);//t1 >> 2; tmp++; *tmp = t1 >> 2;//t2 | ((t1 & (0xff >> 6)) << 6); tmp++; resultsize += 2; } else if ((*p & (0xff << 4))== 0xe0) { uint16_t t = 0; uint8_t t1 = 0; uint8_t t2 = 0; uint8_t t3 = 0; t1 = *p & (0xff >> 3); p++; t2 = *p & (0xff >> 2); p++; t3 = *p & (0xff >> 2); //Little Endian *tmp = ((t2 & (0xff >> 6)) << 6) | t3;//(t1 << 4) | (t2 >> 2); tmp++; *tmp = (t1 << 4) | (t2 >> 2);//((t2 & (0xff >> 6)) << 6) | t3; tmp++; resultsize += 2; } p++; } *tmp = '/0'; tmp++; *tmp = '/0'; resultsize += 2; *out = result; *outsize = resultsize; return 0; } void dump_utf8(uint8_t *utf8) { uint8_t *p = utf8; while(*p) { printf("%02X", *p); p++; } putchar('/n'); } void dump_unicode(uint16_t *utf16, int size) { uint8_t *p = (uint8_t *)utf16; int i = 0; for (i = 0; i < size; i++) { printf("%02X", *p); p++; } putchar('/n'); } int main() { uint16_t unicode[] = L"中文"; uint8_t *utf8 = NULL; int unisize = 0; uint16_t *uni = NULL; printf("original unicode: /n"); dump_unicode(unicode, sizeof(unicode)); printf("converted to utf8: /n"); unicode_to_utf8(unicode, sizeof(unicode), &utf8); dump_utf8(utf8); printf("converted to unicode: /n"); utf8_to_unicode(utf8, &uni, &unisize); dump_unicode(uni, unisize); free(utf8); free(uni); return 0; }

 

输出结果：
original unicode:
4E2D6587
converted to utf8:
E2B58EE89DA5
converted to unicode:
4E2D65870000

测试写的代码，不够完善，也没有优化，应该有很多Bug.