Java 和 JavaScript 真正通用的Base64编码详解

 java 和 javascript base64编码

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>

感谢阅读，希望能帮助到大家，谢谢大家对本站的支持！