C++中GBK、Unicode、UTF8字符串之间的转换方法

wchar_t* UTF8ToUnicode(const char szU8)
{
int wcsLen = ::MultiByteToWideChar(CP_UTF8, NULL, szU8, strlen(szU8), NULL, 0);
//分配空间要给’\0’留个空间，MultiByteToWideChar不会给’\0’空间
wchar_t wszString = new wchar_t[wcsLen + 1];
//转换
::MultiByteToWideChar(CP_UTF8, NULL, szU8, strlen(szU8), wszString, wcsLen);
//最后加上’\0’
wszString[wcsLen] = ‘\0’;
return wszString;
}
char* UnicodeToUTF8(wchar_t wszString)
{
//预转换，得到所需空间的大小，这次用的函数和上面名字相反
int u8Len = ::WideCharToMultiByte(CP_UTF8, NULL, wszString, wcslen(wszString), NULL, 0, NULL, NULL);
//同上，分配空间要给’\0’留个空间
//UTF8虽然是Unicode的压缩形式，但也是多字节字符串，所以可以以char的形式保存
char szU8 = new char[u8Len + 1];
//转换
//unicode版对应的strlen是wcslen
::WideCharToMultiByte(CP_UTF8, NULL, wszString, wcslen(wszString), szU8, u8Len, NULL, NULL);
//最后加上’\0’
szU8[u8Len] = ‘\0’;
return szU8;
}
char* UnicodeToGB2312(wchar_t* szUnicodeString)
{
UINT nCodePage = 936; //GB2312
int nLength=WideCharToMultiByte(nCodePage,0,szUnicodeString,-1,NULL,0,NULL,NULL);
char* pBuffer=new char[nLength+1];
WideCharToMultiByte(nCodePage,0,szUnicodeString,-1,pBuffer,nLength,NULL,NULL);
pBuffer[nLength]=0;
return pBuffer;
}
wchar_t* GB2312ToUnicode(const char* szGBString)
{
UINT nCodePage = 936; //GB2312
int nLength=MultiByteToWideChar(nCodePage,0,szGBString,-1,NULL,0);
wchar_t* pBuffer = new wchar_t[nLength+1];
MultiByteToWideChar(nCodePage,0,szGBString,-1,pBuffer,nLength);
pBuffer[nLength]=0;
return pBuffer;
}
char* UTF8ToGBK2312(const char *szU8)
{
wchar_t *wszString = UTF8ToUnicode(szU8);
char szString = UnicodeToGB2312(wszString);
delete[] wszString;
return szString;
}
char GBK2312ToUTF8(const char *szGBK)
{
wchar_t *wszString = GB2312ToUnicode(szGBK);
char *szString = UnicodeToUTF8(wszString);
delete[] wszString;
return szString;
}

使用实例：
char *pGBKString1=“一二三四五六七八九十一二三四五六七八九十一二三四五六七八九十”;
char *pUTF8String1 = GBK2312ToUTF8(pGBKString1 );
char *pGBKString3=“A373129EB97E05AF5161BCC0CEC2C60D92D9C3EFAF90D53A70”;
char *pUTF8String2 = GBK2312ToUTF8(pGBKString2 );
char *pGBKString2=“A373129EB97E1BF90536F7DE9D88DD1A82DECFB5A9D1DB”;
char *pUTF8String2 = GBK2312ToUTF8(pGBKString2 );
char *pUTF8String3 = GBK2312ToUTF8(pGBKString3 );
::MessageBoxA(NULL,pUTF8String1,“String1”,MB_OK);
::MessageBoxA(NULL,pUTF8String2,“String2”,MB_OK);
::MessageBoxA(NULL,pUTF8String3,“String1”,MB_OK);
delete [] pUTF8String1;
delete [] pUTF8String2;
delete [] pUTF8String3;