[C/C++]_[utf8字符串转换为unicode字符串]

场景:

1.windows下需要unicode来处理非ascii的字符,如中文文件路径.

2.但是做字符串处理时又需要转换成中间的utf8处理,这就涉及到了互转.

3.参考unicode和utf8关系:

http://baike.baidu.com/view/40801.htm

文件:test_utf82unicode.cpp(只做了支持双字节的unicode转码.)

#include <stdio.h>
#include <assert.h>
#include <string.h>
#include <stdlib.h>

void OneUTF82Unicode(const char* utf_char, char* unicode_char);

char* utf82unicode(const char* utf,size_t *unicode_number)
{
char* utf8 = strdup(utf);
size_t utf8_length = strlen(utf8);
//这里可以自己考虑实现更精确的大小.
//+2是留两个00.
char* unicode = (char*)malloc(utf8_length*2+2);
memset(unicode,0,utf8_length*2+2);
size_t index  = 0;
size_t start = 0;
unsigned char temp;
//10000000=0x80 1110=0xE 110=0x6
unsigned char flag = 0;
size_t unicode_index = 0;
bool is_finded = false;
while((temp = utf8[index]))
{
start = index;

temp = temp >> 4 ;
if(temp > 0xE)
{
assert(0);
printf("utf8 bigger than 4 byte is not supported.\n");
break;
}

if(temp == 0xE)
{
index+=3;
is_finded = true;
}

temp = temp >> 1;
if(!is_finded && temp == 0x6)
{
index+=2;
is_finded = true;
}

temp = temp >> 2;
if(!is_finded && temp == 0x00)
{
index+=1;
is_finded = true;
}

if(index > utf8_length)
{
break;
}
flag = utf8[index];
utf8[index] = 0;
OneUTF82Unicode(utf8+start,unicode+unicode_index);
utf8[index] = flag;
unicode_index+=2;
is_finded = false;
}
free(utf8);
*unicode_number = unicode_index/2;
return unicode;
}

void OneUTF82Unicode(const char* utf_char, char* unicode_char)
{
//unicode: 0x192->110010010 ,utf8:0xC692->11000110|10010010
//小端序
int utf_length = strlen(utf_char);
//0x3F->00111111
switch(utf_length)
{
case 1:
unicode_char[0] = utf_char[0];
unicode_char[1] = 0;
break;
case 2:
unicode_char[0] = (utf_char[1] & 0x3F) | ((utf_char[0] & 0x3) << 6);
unicode_char[1] = (utf_char[0] & 0x3C) >> 2;
break;
case 3:
unicode_char[0] = (utf_char[2] & 0x3F) | ((utf_char[1] & 0x3) << 6);
unicode_char[1] = ((utf_char[1] & 0x3C) >> 2) | ((utf_char[0] & 0xF) << 4);
break;
default:
assert(0);
printf("utf_char length is bigger than 4 unsupported.\n");
break;
}
}

int main(int argc, char *argv[])
{
printf("Hello, world\n");
char utf_char[12] = {0xC3,0xBE};
memset(utf_char+2,0,10);

char unicode_char[3] = {0,0,0};
OneUTF82Unicode(utf_char,unicode_char);
printf("%x,%x\n",(unsigned char)unicode_char[0],unicode_char[1]);

utf_char[0] = 0xE6;
utf_char[1] = 0x88;
utf_char[2] = 0x91;

OneUTF82Unicode(utf_char,unicode_char);
printf("%x,%x\n",(unsigned char)unicode_char[0],unicode_char[1]);

size_t unicode_number = 0;
char utf_str[]={0xC3,0xBE,0xE6,0x88,0x91,0xE4,0xBB,0xAC,0xE7,0x9A,0x84,0xE5,0x9B,
0xBD,0xE5,0xAE,0xB6,0x64,0x61,0xE7,0x9A,0x84,0xE5,0x93,0xA6};
char* unicode = utf82unicode(utf_str,&unicode_number);

printf("%d\n",unicode_number);
size_t index = 0;
for(size_t i=0;i<unicode_number;++i,index+=2)
{
printf("%x,%x\n",(unsigned char)unicode[index],(unsigned char)unicode[index+1]);
}
free(unicode);
return 0;
}

输出:

Hello, world
fe,0
11,62
10
fe,0
11,62
ec,4e
84,76
fd,56
b6,5b
64,0
61,0
84,76
e6,54