int gbk_to_unicode(uint16* out,const char* in,int n)
{
int i=0;
int j=0;
uint16 gbcode = 0;
while(n-1>=i){
if (in[i]&0x80) {
if(n-i>1){
gbcode=(in[i+1]<<8&0xff00)+(in[i]&0x00ff);
out[j]=charsets_gbk_to_ucs((uint8 *)&gbcode);
j++;
i=i+2;
}
else{
i++;
continue;
}
}
else{
gbcode = in[i]&0x00ff;
out[j] = gbcode;
i++;
j++;
}
}
return j;
}
/*
* Convert the src string to UTF8 coding dst string, and cut to length
*/
int string2utf8(unsigned char *src, unsigned char* dst, unsigned int length)
{
unsigned char *pt;
unsigned char ch;
unsigned short ucode;
unsigned int type;
unsigned int len;
len = 0;
type = 0;
pt = src;
while(*pt)
{
pt = utf8decode(pt, &ucode);
if(ucode < 0x4e00) {
if(ucode == 0 || ucode > 0x7F) {
type = 1;
break;
}
} else if(ucode > 0x9FCF) {
type = 1;
break;
}
else
len++;
if(len >= 3) break; //There is enough UTF8, so it is, to save time(>_*)
}
if(type == 0) //UTF8
{
while(*src)
{
ch = *src++;
*dst++ = ch;
if(ch < 0x80) {
if(length > 1) length -= 1;
else break;
} else if (ch < 0xe0) { /* U-00000080 - U-000007FF, 2 bytes */
if(length > 2) length -= 2;
else break;
*dst++ = *src++;
} else if (ch < 0xf0) { /* U-00000800 - U-0000FFFF, 3 bytes */
if(length > 3) length -= 3;
else break;
*dst++ = *src++;
*dst++ = *src++;
} else if (ch < 0xf5) { /* U-00010000 - U-001FFFFF, 4 bytes */
if(length > 4) length -= 4;
else break;
*dst++ = *src++;
*dst++ = *src++;
*dst++ = *src++;
} else {
break;
}
}
*dst = '\0';
}
else //assume it is GBK code
{
//GBK to UTF8
while(*src)
{
ch = *src;
if(ch < 0x80)
{
if(length > 1) length -= 1;
else break;
*dst++= ch;
src ++;
}
else
{
ucode = charsets_gbk_to_ucs(src);
if (ucode < 0x800) //2 bytes
{
if(length > 2) length -= 2;
else break;
*dst++ = 0xC0 | ((ucode >> 6) & 0x1F);
*dst++ = 0x80 | (ucode & 0x3F);
}
else //3 bytes
{
if(length > 3) length -= 3;
else break;
*dst++ = 0xE0 | (ucode >> 12);
*dst++ = 0x80 | ((ucode >>6) & 0x3F);
*dst++ = 0x80 | (ucode & 0x3F);
}
src += 2;
}
}
*dst = '\0';
}
