utf-8与Unicode互转.

在vc下使用SQLite数据库时，由于SQL语句使用utf8 编码，而CString 是unicode编码。一， utf8 转 UnicodeCString UTF8ToUnicode(char* UTF8){DWORD dwUnicodeLen;        //转换后Unicode的长度TCHAR *pwText;            //保存Unicode的指针CString strUnicode;        //返回值//获得转换后的长度，并分配内存dwUnicodeLen = MultiByteToWideChar(CP_UTF8,0,UTF8,-1,NULL,0);pwText = new TCHAR[dwUnicodeLen];if (!pwText){return strUnicode;}//转为UnicodeMultiByteToWideChar(CP_UTF8,0,UTF8,-1,pwText,dwUnicodeLen);//转为CStringstrUnicode.Format(_T("%s"),pwText);//清除内存delete []pwText;//返回转换好的Unicode字串return strUnicode;}二， Unicode转utf8size_t CDGQDialog::g_f_wctou8(char * dest_str, const wchar_t src_wchar){int count_bytes = 0;wchar_t byte_one = 0, byte_other = 0x3f; // 用于位与运算以提取位值0x3f--->00111111unsigned char utf_one = 0, utf_other = 0x80; // 用于"位或"置标UTF-8编码0x80--->1000000wchar_t tmp_wchar =L'0'; // 用于宽字符位置析取和位移(右移位)unsigned char tmp_char =L'0';if (!src_wchar)//return (size_t)-1;for (;;) // 检测字节序列长度{if (src_wchar <= 0x7f){ // <=01111111count_bytes = 1; // ASCII字符: 0xxxxxxx( ~ 01111111)byte_one = 0x7f; // 用于位与运算, 提取有效位值, 下同utf_one = 0x0;break;}if ( (src_wchar > 0x7f) && (src_wchar <= 0x7ff) ){ // <=0111,11111111count_bytes = 2; // 110xxxxx 10xxxxxx[1](最多个位, 简写为*1)byte_one = 0x1f; // 00011111, 下类推(1位的数量递减)utf_one = 0xc0; // 11000000break;}if ( (src_wchar > 0x7ff) && (src_wchar <= 0xffff) ){ //0111,11111111<=11111111,11111111count_bytes = 3; // 1110xxxx 10xxxxxx[2](MaxBits: 16*1)byte_one = 0xf; // 00001111utf_one = 0xe0; // 11100000break;}if ( (src_wchar > 0xffff) && (src_wchar <= 0x1fffff) ){ //对UCS-4的支持..count_bytes = 4; // 11110xxx 10xxxxxx[3](MaxBits: 21*1)byte_one = 0x7; // 00000111utf_one = 0xf0; // 11110000break;}if ( (src_wchar > 0x1fffff) && (src_wchar <= 0x3ffffff) ){count_bytes = 5; // 111110xx 10xxxxxx[4](MaxBits: 26*1)byte_one = 0x3; // 00000011utf_one = 0xf8; // 11111000break;}if ( (src_wchar > 0x3ffffff) && (src_wchar <= 0x7fffffff) ){count_bytes = 6; // 1111110x 10xxxxxx[5](MaxBits: 31*1)byte_one = 0x1; // 00000001utf_one = 0xfc; // 11111100break;}return (size_t)-1; // 以上皆不满足则为非法序列}// 以下几行析取宽字节中的相应位, 并分组为UTF-8编码的各个字节tmp_wchar = src_wchar;for (int i = count_bytes; i > 1; i--){ // 一个宽字符的多字节降序赋值tmp_char = (unsigned char)(tmp_wchar & byte_other);///后位与byte_other 00111111dest_str[i - 1] = (tmp_char | utf_other);/// 在前面加----跟或tmp_wchar >>= 6;//右移位}//这个时候i=1//对UTF-8第一个字节位处理，//第一个字节的开头"1"的数目就是整个串中字节的数目tmp_char = (unsigned char)(tmp_wchar & byte_one);//根据上面附值得来，有效位个数dest_str[0] = (tmp_char | utf_one);//根据上面附值得来1的个数// 位值析取分组__End!return count_bytes;}int CDGQDialog::g_f_wcs_to_pchar(CString& wstr,char * p){wchar_t wc=L'1';char c[10]="1";//申请一个缓存size_t r=0; //size_t unsigned integer Result of sizeof operatorint i=0;int j=0;for(i=0;i<wstr.GetLength();i++){wc=wstr.GetAt(i);//得到一个宽字符r=g_f_wctou8(c,wc);//将一个宽字符按UTF-8格式转换到p地址if(r==-1)//出错判断AfxMessageBox(_T("wcs_to_pchar error"));p[j]=c[0];//第一个值附给pj++; if(r>1) {for(size_t x=1;x<r;x++){p[j]=c[x];j++;}}}//p[j]='0';return 1;}三．转换实例void CMytestDlg::OnBnClickedButton2(){// TODO: 在此添加控件通知处理程序代码CString ccId=L"2007071王";CString sql;char mySql[100];memset(mySql,0,sizeof(mySql));sql.Format(L"select cxrq,cxdw,dxrq,dxdw,fxrq,fxdw,cx,flx from j_clxx where trainnum_info_id ='%s'",ccId);//wchar_t sql=L'你';g_f_wcs_to_pchar(sql,mySql);CString sql1 =UTF8ToUnicode(mySql);MessageBox(sql);//g_f_wctou8(mySql,sql);//   CString str_temp;//   for (int i=90;i<strlen(mySql);i++)//   {//   str_temp.Format(L"%c",mySql[i]);//   MessageBox(str_temp);//   }

转载：http://zhanshaoji.blog.163.com/blog/static/174816245201241842845463/