Unicode 与 UTF-8 编码互转

Unicode 转 UTF-8

char* UnicodeToUtf8(const wchar_t* unicode)
{
    int len;
    len = WideCharToMultiByte(CP_UTF8, 0, unicode, -1, NULL, 0, NULL, NULL);
    char *szUtf8 = (char*)malloc(len + 1);
    memset(szUtf8, 0, len + 1);
    WideCharToMultiByte(CP_UTF8, 0, unicode, -1, szUtf8, len, NULL, NULL);
    return szUtf8;
}

int main(int argc, char *argv[])
{
    wchar_t* wCharUnicode = L"中国";
    char* cCharUtf = UnicodeToUtf8(wCharUnicode);

    return 0;
}           

UTF-8 转 Unicode

CString UTF82WCS(const char* szU8)
{
    //预转换，得到所需空间的大小;
    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';

    CString unicodeString(wszString);

    delete[] wszString;
    wszString = NULL;

    return unicodeString;
}

int main(int argc, char *argv[])
{
    wchar_t* wCharUnicode = L"中国";
    char* cCharUtf = UnicodeToUtf8(wCharUnicode);
    CString strUnicode = UTF82WCS(cCharUtf);

    return 0;
}           

测试效果：

int length;
wchar_t* wCharUnicode = L"中国你好";
length = wcslen(wCharUnicode);                      // length = 4;

char* cCharUtf = UnicodeToUtf8(wCharUnicode);
length = strlen(cCharUtf);                          // length = 12;

// 将UTF格式的char*转为CString
CString strUtf(cCharUtf);
length = strUtf.GetLength();                        // length = 6;

CString strUnicode = UTF82WCS(cCharUtf);
length = strUnicode.GetLength();                    // length = 4;           

int length;
wchar_t* wCharUnicode = L"中国,你好abc";
length = wcslen(wCharUnicode);                      // length = 8;

char* cCharUtf = UnicodeToUtf8(wCharUnicode);
length = strlen(cCharUtf);                          // length = 16;

// 将UTF格式的char*转为CString
CString strUtf(cCharUtf);
length = strUtf.GetLength();                        // length = 10;

CString strUnicode = UTF82WCS(cCharUtf);
length = strUnicode.GetLength();                    // length = 8;           

int length;
wchar_t* wCharUnicode = L"中国，你好abc";
length = wcslen(wCharUnicode);                      // length = 8;

char* cCharUtf = UnicodeToUtf8(wCharUnicode);
length = strlen(cCharUtf);                          // length = 18;

// 将UTF格式的char*转为CString
CString strUtf(cCharUtf);
length = strUtf.GetLength();                        // length = 10;

CString strUnicode = UTF82WCS(cCharUtf);
length = strUnicode.GetLength();                    // length = 8;