javascript統計計算中文字串長度

如何讓JS統計的字串長度與php一致呢？這裡的函數代碼只是針對GBK下編碼的字元，一個漢字等於二個字元。代碼如下：

代碼如下	複製代碼
function strlen(str) {     var s = 0;     for(var i = 0; i < str.length; i++) {         if(str.charAt(i).match(/[u0391-uFFE5]/)) {             s += 2;            } else {             s++;         }     }     return s; }

抓取出每個字元，匹配全形字元和漢字的，則計2個字元，其他的則計1個字元。

 

代碼如下	複製代碼
<script>    alert (fucCheckLength("中國a"));    function fucCheckLength(strTemp)    {    var i,sum;    sum=0;    for(i=0;i<strTemp.length;i++)    {      if ((strTemp.charCodeAt(i)>=0) && (strTemp.charCodeAt(i)<=255))        sum=sum+1;      else       sum=sum+2;    }    return sum;    }    </script>

 
會得到結果是：5

要得到的位元組長度吧？請注意位元組和字元的差異。而位元組長度是和編碼有關係的，比如"中國a"，gbk/gb2312編碼是5個位元組，可是如果是utf-8，則是7個位元組（utf-8下通常一個漢字3個位元組）。

我們可以把所有字元轉換在gbk再操作，執行個體

 

代碼如下

複製代碼

function Utf8ToUnicode(strUtf8) {         var bstr = "";         var nTotalChars = strUtf8.length;        // total chars to be processed.         var nOffset = 0;                                        // processing point on strUtf8         var nRemainingBytes = nTotalChars;        // how many bytes left to be converted         var nOutputPosition = 0;         var iCode, iCode1, iCode2;                        // the value of the unicode.         while (nOffset < nTotalChars)         {                 iCode = strUtf8.charCodeAt(nOffset);                 if ((iCode & 0x80) == 0)                        // 1 byte.                 {                         if ( nRemainingBytes < 1 )                // not enough data                                 break;                         bstr += String.fromCharCode(iCode & 0x7F);                         nOffset ++;                         nRemainingBytes -= 1;                 }                 else if ((iCode & 0xE0) == 0xC0)        // 2 bytes                 {                         iCode1 =  strUtf8.charCodeAt(nOffset + 1);                         if ( nRemainingBytes < 2 ||                        // not enough data                                  (iCode1 & 0xC0) != 0x80 )                // invalid pattern                         {                                 break;                         }                         bstr += String.fromCharCode(((iCode & 0x3F) << 6) | (         iCode1 & 0x3F));                         nOffset += 2;                         nRemainingBytes -= 2;                 }                 else if ((iCode & 0xF0) == 0xE0)        // 3 bytes                 {                         iCode1 =  strUtf8.charCodeAt(nOffset + 1);                         iCode2 =  strUtf8.charCodeAt(nOffset + 2);                         if ( nRemainingBytes < 3 ||                        // not enough data                                  (iCode1 & 0xC0) != 0x80 ||                // invalid pattern                                  (iCode2 & 0xC0) != 0x80 )                         {                                 break;                         }                         bstr += String.fromCharCode(((iCode & 0x0F) << 12) |                                         ((iCode1 & 0x3F) <<  6) |                                         (iCode2 & 0x3F));                         nOffset += 3;                         nRemainingBytes -= 3;                 }                 else                                                                // 4 or more bytes -- unsupported                         break;         }         if (nRemainingBytes != 0)         {                 // bad UTF8 string.                 return "";         }         return bstr; }