注册 登录  
 加关注
查看详情
   显示下一条  |  关闭
温馨提示!由于新浪微博认证机制调整,您的新浪微博帐号绑定已过期,请重新绑定!立即重新绑定新浪微博》  |  关闭

Simon

 
 
 

日志

 
 

Java 和 JavaScript 之间真正通用的Base64编码  

2008-12-08 18:46:01|  分类: 默认分类 |  标签: |举报 |字号 订阅

  下载LOFTER 我的照片书  |

在开发Java  Web应用的时候,可能会在服务器端用Java做Base64编码,而在客户端用JavaScript进行解码。这样就要求两边的Base64编码机制保持一致。

使用Base64编码,可能会碰到各种奇怪情况,甚至怀疑编码有bug。但实际上不是这样的。Base64理论上操作的对象不是字符串而是字节数组。它的原理就是把ASCII码的255个字符缩小到用64个来表示。具体就是原来三个字节用四个字节表示,编码后长度有一定的增长。

1) 最好一次编码,避免分段编码,确实要分段编码,每一段字节数应该是3的倍数。

长字节流,如果要边读取边编码,每一段必须是3的倍数,否则就可能在还原的时候出乱。一般人喜欢用2的乘方来定义数组,例如 byte[1024],因为不是3的倍数,可能还原时出错。正确的例子是:

byte[] bs=new byte[3*100] ....inputStream.read(bs)......encode(bs )....

对于字符串,一般要整个一次编码,以避免分段编码出错。

当然,如果你分段编码,还原的时候也是一段一段地还原,那是没有问题的。

2)确保字符串还原的时候按照原来的编码还原。

因为它操作的是字节数组,所以对于GBK编码的汉字和UTF-8编码汉字,经过 Base64编码后结果是不一样的。例如“我们”这两个字如果是GBK编码,转成Base64后就是ztLDxw== ;如果是UTF-8编码,转成Base64后就是5oiR5Lus。

也就是 “我们” ==》  getBytes("GBK") ==> Base64

所以Java这边用什么编码转换,在JavaScript那边就要用什么编码还原。要保证Java和JavaScript通用,我们采用Unicode的编码(JavaScript转成UTF-8、GBK不方便,所以就采用了其本身的Unicode编码),具体如下:

服务器端:

1)用getBytes("Unicode")转成Unicode字节数组。

2) 编码成Base64字符串

3)传送到客户端

客户端:

1)Base64 解码成字节数组

2)按Unicode还原

代码如下(相关的函数看附件):

Base64.encode(data,"Unicode"); //java 端编码

decode64(data);   //javascript解码

 

 

附一:Java中Base64编码

package websharp.util;

public class Base64 {

    private static final byte[] encodingTable = {

            (byte) 'A', (byte) 'B', (byte) 'C', (byte) 'D', (byte) 'E',

            (byte) 'F', (byte) 'G', (byte) 'H', (byte) 'I', (byte) 'J',

            (byte) 'K', (byte) 'L', (byte) 'M', (byte) 'N', (byte) 'O',

            (byte) 'P', (byte) 'Q', (byte) 'R', (byte) 'S', (byte) 'T',

            (byte) 'U', (byte) 'V', (byte) 'W', (byte) 'X', (byte) 'Y',

            (byte) 'Z', (byte) 'a', (byte) 'b', (byte) 'c', (byte) 'd',

            (byte) 'e', (byte) 'f', (byte) 'g', (byte) 'h', (byte) 'i',

            (byte) 'j', (byte) 'k', (byte) 'l', (byte) 'm', (byte) 'n',

            (byte) 'o', (byte) 'p', (byte) 'q', (byte) 'r', (byte) 's',

            (byte) 't', (byte) 'u', (byte) 'v', (byte) 'w', (byte) 'x',

            (byte) 'y', (byte) 'z', (byte) '0', (byte) '1', (byte) '2',

            (byte) '3', (byte) '4', (byte) '5', (byte) '6', (byte) '7',

            (byte) '8', (byte) '9', (byte) '+', (byte) '/'

        };

    private static final byte[] decodingTable;

    static {

        decodingTable = new byte[128];

        for (int i = 0; i < 128; i++) {

            decodingTable[i] = (byte) -1;

        }

        for (int i = 'A'; i <= 'Z'; i++) {

            decodingTable[i] = (byte) (i - 'A');

        }

        for (int i = 'a'; i <= 'z'; i++) {

            decodingTable[i] = (byte) (i - 'a' + 26);

        }

        for (int i = '0'; i <= '9'; i++) {

            decodingTable[i] = (byte) (i - '0' + 52);

        }

        decodingTable['+'] = 62;

        decodingTable['/'] = 63;

    }

    public static byte[] encode(byte[] data,int offset) {

        byte[] bytes;

        int realCount=data.length-offset;

        int modulus = realCount % 3;

        if (modulus == 0) {

            bytes = new byte[(4 * realCount) / 3];

        } else {

            bytes = new byte[4 * ((realCount / 3) + 1)];

        }

        int dataLength = (data.length - modulus);

        int a1;

        int a2;

        int a3;

        for (int i = offset, j = 0; i < dataLength; i += 3, j += 4) {

            a1 = data[i] & 0xff;

            a2 = data[i + 1] & 0xff;

            a3 = data[i + 2] & 0xff;

            bytes[j] = encodingTable[(a1 >>> 2) & 0x3f];

            bytes[j + 1] = encodingTable[((a1 << 4) | (a2 >>> 4)) & 0x3f];

            bytes[j + 2] = encodingTable[((a2 << 2) | (a3 >>> 6)) & 0x3f];

            bytes[j + 3] = encodingTable[a3 & 0x3f];

        }

        int b1;

        int b2;

        int b3;

        int d1;

        int d2;

        switch (modulus) {

        case 0: /* nothing left to do */

            break;

        case 1:

            d1 = data[data.length - 1] & 0xff;

            b1 = (d1 >>> 2) & 0x3f;

            b2 = (d1 << 4) & 0x3f;

            bytes[bytes.length - 4] = encodingTable[b1];

            bytes[bytes.length - 3] = encodingTable[b2];

            bytes[bytes.length - 2] = (byte) '=';

            bytes[bytes.length - 1] = (byte) '=';

            break;

        case 2:

            d1 = data[data.length - 2] & 0xff;

            d2 = data[data.length - 1] & 0xff;

            b1 = (d1 >>> 2) & 0x3f;

            b2 = ((d1 << 4) | (d2 >>> 4)) & 0x3f;

            b3 = (d2 << 2) & 0x3f;

            bytes[bytes.length - 4] = encodingTable[b1];

            bytes[bytes.length - 3] = encodingTable[b2];

            bytes[bytes.length - 2] = encodingTable[b3];

            bytes[bytes.length - 1] = (byte) '=';

            break;

        }

        return bytes;

    }

    public static byte[] decode(byte[] data) {

        byte[] bytes;

        byte b1;

        byte b2;

        byte b3;

        byte b4;

        data = discardNonBase64Bytes(data);

        if (data[data.length - 2] == '=') {

            bytes = new byte[(((data.length / 4) - 1) * 3) + 1];

        } else if (data[data.length - 1] == '=') {

            bytes = new byte[(((data.length / 4) - 1) * 3) + 2];

        } else {

            bytes = new byte[((data.length / 4) * 3)];

        }

        for (int i = 0, j = 0; i < (data.length - 4); i += 4, j += 3) {

            b1 = decodingTable[data[i]];

            b2 = decodingTable[data[i + 1]];

            b3 = decodingTable[data[i + 2]];

            b4 = decodingTable[data[i + 3]];

            bytes[j] = (byte) ((b1 << 2) | (b2 >> 4));

            bytes[j + 1] = (byte) ((b2 << 4) | (b3 >> 2));

            bytes[j + 2] = (byte) ((b3 << 6) | b4);

        }

        if (data[data.length - 2] == '=') {

            b1 = decodingTable[data[data.length - 4]];

            b2 = decodingTable[data[data.length - 3]];

            bytes[bytes.length - 1] = (byte) ((b1 << 2) | (b2 >> 4));

        } else if (data[data.length - 1] == '=') {

            b1 = decodingTable[data[data.length - 4]];

            b2 = decodingTable[data[data.length - 3]];

            b3 = decodingTable[data[data.length - 2]];

            bytes[bytes.length - 2] = (byte) ((b1 << 2) | (b2 >> 4));

            bytes[bytes.length - 1] = (byte) ((b2 << 4) | (b3 >> 2));

        } else {

            b1 = decodingTable[data[data.length - 4]];

            b2 = decodingTable[data[data.length - 3]];

            b3 = decodingTable[data[data.length - 2]];

            b4 = decodingTable[data[data.length - 1]];

            bytes[bytes.length - 3] = (byte) ((b1 << 2) | (b2 >> 4));

            bytes[bytes.length - 2] = (byte) ((b2 << 4) | (b3 >> 2));

            bytes[bytes.length - 1] = (byte) ((b3 << 6) | b4);

        }

        return bytes;

    }

    public static byte[] decode(String data) {

        byte[] bytes;

        byte b1;

        byte b2;

        byte b3;

        byte b4;

        data = discardNonBase64Chars(data);

        if (data.charAt(data.length() - 2) == '=') {

            bytes = new byte[(((data.length() / 4) - 1) * 3) + 1];

        } else if (data.charAt(data.length() - 1) == '=') {

            bytes = new byte[(((data.length() / 4) - 1) * 3) + 2];

        } else {

            bytes = new byte[((data.length() / 4) * 3)];

        }

        for (int i = 0, j = 0; i < (data.length() - 4); i += 4, j += 3) {

            b1 = decodingTable[data.charAt(i)];

            b2 = decodingTable[data.charAt(i + 1)];

            b3 = decodingTable[data.charAt(i + 2)];

            b4 = decodingTable[data.charAt(i + 3)];

            bytes[j] = (byte) ((b1 << 2) | (b2 >> 4));

            bytes[j + 1] = (byte) ((b2 << 4) | (b3 >> 2));

            bytes[j + 2] = (byte) ((b3 << 6) | b4);

        }

        if (data.charAt(data.length() - 2) == '=') {

            b1 = decodingTable[data.charAt(data.length() - 4)];

            b2 = decodingTable[data.charAt(data.length() - 3)];

            bytes[bytes.length - 1] = (byte) ((b1 << 2) | (b2 >> 4));

        } else if (data.charAt(data.length() - 1) == '=') {

            b1 = decodingTable[data.charAt(data.length() - 4)];

            b2 = decodingTable[data.charAt(data.length() - 3)];

            b3 = decodingTable[data.charAt(data.length() - 2)];

            bytes[bytes.length - 2] = (byte) ((b1 << 2) | (b2 >> 4));

            bytes[bytes.length - 1] = (byte) ((b2 << 4) | (b3 >> 2));

        } else {

            b1 = decodingTable[data.charAt(data.length() - 4)];

            b2 = decodingTable[data.charAt(data.length() - 3)];

            b3 = decodingTable[data.charAt(data.length() - 2)];

            b4 = decodingTable[data.charAt(data.length() - 1)];

            bytes[bytes.length - 3] = (byte) ((b1 << 2) | (b2 >> 4));

            bytes[bytes.length - 2] = (byte) ((b2 << 4) | (b3 >> 2));

            bytes[bytes.length - 1] = (byte) ((b3 << 6) | b4);

        }

        for(int i=0;i<bytes.length;i++) System.out.println(","+bytes[i]);

        return bytes;

    }

    private static byte[] discardNonBase64Bytes(byte[] data) {

        byte[] temp = new byte[data.length];

        int bytesCopied = 0;

        for (int i = 0; i < data.length; i++) {

            if (isValidBase64Byte(data[i])) {

                temp[bytesCopied++] = data[i];

            }

        }

        byte[] newData = new byte[bytesCopied];

        System.arraycopy(temp, 0, newData, 0, bytesCopied);

        return newData;

    }

    private static String discardNonBase64Chars(String data) {

        StringBuffer sb = new StringBuffer();

        int length = data.length();

        for (int i = 0; i < length; i++) {

            if (isValidBase64Byte((byte) (data.charAt(i)))) {

                sb.append(data.charAt(i));

            }

        }

        return sb.toString();

    }

    private static boolean isValidBase64Byte(byte b) {

        if (b == '=') {

            return true;

        } else if ((b < 0) || (b >= 128)) {

            return false;

        } else if (decodingTable[b] == -1) {

            return false;

        }

        return true;

    }

    public static String  encode(String data,String charset)throws Exception

    {

        // byte[] result =  (data.getBytes("Unicode"));

         if(data==null || data.length()==0) return data;

         int offset=0;

         // getBytes("unicode")转完后会在前头加上两字节”FE“

         byte[] result=encode (data.getBytes(charset),offset);

         StringBuffer sb=new StringBuffer(result.length);

         for (int i=0;i<result.length;i++)   sb.append((char)result[i]);

         return sb.toString();

    }

    public static String  decode(String data,String charset)throws Exception

    {

        if(data==null || data.length()==0) return data;

        return new String(Base64.decode(data),charset);

    }

    public static void main(String[] args) throws Exception {

        String data = "我们";

        String data1=encode(data,"Unicode");

        String data2=decode(data1,"Unicode");

        System.out.println(data);

        System.out.println(data1);

        System.out.println(data2);

    }

}

附二:JavaScript中Base64编码

<html>

<head>

<title>base64 Encoding/Decoding</title>

</head>

<script type="text/javascript"><!--

var keyStr = "ABCDEFGHIJKLMNOP" +

             "QRSTUVWXYZabcdef" +

             "ghijklmnopqrstuv" +

             "wxyz0123456789+/" +

             "=";

function encode64(input) {

   input = unicodetoBytes(input);

   var output = "";

   var chr1, chr2, chr3 = "";

   var enc1, enc2, enc3, enc4 = "";

   var i = 0;

   do {

      chr1 = input[i++];

      chr2 = input[i++];

      chr3 = input[i++];

      enc1 = chr1 >> 2;

      enc2 = ((chr1 & 3) << 4) | (chr2 >> 4);

      enc3 = ((chr2 & 15) << 2) | (chr3 >> 6);

      enc4 = chr3 & 63;

      if (isNaN(chr2)) {

         enc3 = enc4 = 64;

      } else if (isNaN(chr3)) {

         enc4 = 64;

      }

      output = output +

         keyStr.charAt(enc1) +

         keyStr.charAt(enc2) +

         keyStr.charAt(enc3) +

         keyStr.charAt(enc4);

      chr1 = chr2 = chr3 = "";

      enc1 = enc2 = enc3 = enc4 = "";

   } while (i < input.length);

   return output;

}

function decode64(input) {

   var output = "";

   var chr1, chr2, chr3 = "";

   var enc1, enc2, enc3, enc4 = "";

   var i = 0;

   // remove all characters that are not A-Z, a-z, 0-9, +, /, or =

   var base64test = /[^A-Za-z0-9\+\/\=]/g;

   if (base64test.exec(input)) {

      alert("There were invalid base64 characters in the input text.\n" +

            "Valid base64 characters are A-Z, a-z, 0-9, '+', '/', and '='\n" +

            "Expect errors in decoding.");

   }

   input = input.replace(/[^A-Za-z0-9\+\/\=]/g, "");

    output=new Array();

   do {

      enc1 = keyStr.indexOf(input.charAt(i++));

      enc2 = keyStr.indexOf(input.charAt(i++));

      enc3 = keyStr.indexOf(input.charAt(i++));

      enc4 = keyStr.indexOf(input.charAt(i++));

      chr1 = (enc1 << 2) | (enc2 >> 4);

      chr2 = ((enc2 & 15) << 4) | (enc3 >> 2);

      chr3 = ((enc3 & 3) << 6) | enc4;

      output.push(chr1);

      if (enc3 != 64) {

         output.push(chr2);

      }

      if (enc4 != 64) {

         output.push(chr3);

      }

      chr1 = chr2 = chr3 = "";

      enc1 = enc2 = enc3 = enc4 = "";

   } while (i < input.length);

   return bytesToUnicode(output);

}

  function unicodetoBytes(s)

   {

      var result=new Array();

      if(s==null || s=="") return result;

      result.push(255); // add "FE" to head

      result.push(254);

      for(var i=0;i<s.length;i++)

      {

        var c=s.charCodeAt(i).toString(16);

        if(c.length==1) i="000"+c;

        else if(c.length==2) c="00"+c;

        else if(c.length==3) c="0"+c;

        var var1=parseInt( c.substring(2),16);

        var var2=parseInt( c.substring(0,2),16);

        result.push( var1);

        result.push(var2) ;

      }

      return result;

   }

   function bytesToUnicode(bs)

   {

     var result="";

      var offset=0;

      if(bs.length>=2 && bs[0]==255 && bs[1]==254) offset=2;  // delete "FE"

      for(var i=offset;i<bs.length;i+=2)

      {

            var code=bs[i]+(bs[i+1]<<8);

            result+=String.fromCharCode(code);

      }

      return result;

   }

//-->

</script>

<body>

<form name="base64Form">

   Type in the message you want to encode in base64, or paste<br>

   base64 encoded text into the text field, select Encode or Decode, <br>

   and click the button!<br>

   <textarea name="theText" cols="40" rows="6"></textarea><br>

   <input type="button" name="encode" value="Encode to base64"

      onClick="document.base64Form.theText.value=encode64(document.base64Form.theText.value);">

   <input type="button" name="decode" value="Decode from base64"

      onClick="document.base64Form.theText.value=decode64(document.base64Form.theText.value);">

</form>

</body>

</html>

  评论这张
 
阅读(4189)| 评论(2)
推荐 转载

历史上的今天

在LOFTER的更多文章

评论

<#--最新日志,群博日志--> <#--推荐日志--> <#--引用记录--> <#--博主推荐--> <#--随机阅读--> <#--首页推荐--> <#--历史上的今天--> <#--被推荐日志--> <#--上一篇,下一篇--> <#-- 热度 --> <#-- 网易新闻广告 --> <#--右边模块结构--> <#--评论模块结构--> <#--引用模块结构--> <#--博主发起的投票-->
 
 
 
 
 
 
 
 
 
 
 
 
 
 

页脚

网易公司版权所有 ©1997-2018