MD5code

  1 public class MD5Code {
  2  /*
  3   * 下面这些S11-S44实际上是一个4*4的矩阵，在原始的C实现中是用#define 实现的， 这里把它们实现成为static
  4   * final是表示了只读，切能在同一个进程空间内的多个 Instance间共享
  5   */
  6  static final int S11 = 7;
  7 
  8  static final int S12 = 12;
  9 
 10  static final int S13 = 17;
 11 
 12  static final int S14 = 22;
 13 
 14  static final int S21 = 5;
 15 
 16  static final int S22 = 9;
 17 
 18  static final int S23 = 14;
 19 
 20  static final int S24 = 20;
 21 
 22  static final int S31 = 4;
 23 
 24  static final int S32 = 11;
 25 
 26  static final int S33 = 16;
 27 
 28  static final int S34 = 23;
 29 
 30  static final int S41 = 6;
 31 
 32  static final int S42 = 10;
 33 
 34  static final int S43 = 15;
 35 
 36  static final int S44 = 21;
 37 
 38  static final byte[] PADDING = { -128, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
 39    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
 40    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
 41    0, 0, 0, 0, 0, 0, 0 };
 42 
 43  /*
 44   * 下面的三个成员是MD5计算过程中用到的3个核心数据，在原始的C实现中 被定义到MD5_CTX结构中
 45   */
 46  private long[] state = new long[4];// state (ABCD)
 47 
 48  private long[] count = new long[2];// number of bits, modulo 2^64 (lsb
 49 
 50  // first)
 51 
 52  private byte[] buffer = new byte[64]; // input buffer
 53 
 54  /*
 55   * digestHexStr是MD5的唯一一个公共成员，是最新一次计算结果的 16进制ASCII表示.
 56   */
 57 
 58  public String digestHexStr;
 59 
 60  /*
 61   * digest,是最新一次计算结果的2进制内部表示，表示128bit的MD5值.
 62   */
 63  private byte[] digest = new byte[16];
 64 
 65  /*
 66   * getMD5ofStr是类MD5最主要的公共方法，入口参数是你想要进行MD5变换的字符串
 67   * 返回的是变换完的结果，这个结果是从公共成员digestHexStr取得的．
 68   */
 69  public String getMD5ofStr(String inbuf) {
 70   md5Init();
 71   md5Update(inbuf.getBytes(), inbuf.length());
 72   md5Final();
 73   digestHexStr = "";
 74   for (int i = 0; i < 16; i++) {
 75    digestHexStr += byteHEX(digest[i]);
 76   }
 77   return digestHexStr;
 78  }
 79 
 80  // 这是MD5这个类的标准构造函数，JavaBean要求有一个public的并且没有参数的构造函数
 81  public MD5Code() {
 82   md5Init();
 83   return;
 84  }
 85 
 86  /* md5Init是一个初始化函数，初始化核心变量，装入标准的幻数 */
 87  private void md5Init() {
 88   count[0] = 0L;
 89   count[1] = 0L;
 90   // /* Load magic initialization constants.
 91   state[0] = 0x67452301L;
 92   state[1] = 0xefcdab89L;
 93   state[2] = 0x98badcfeL;
 94   state[3] = 0x10325476L;
 95   return;
 96  }
 97 
 98  /*
 99   * F, G, H ,I 是4个基本的MD5函数，在原始的MD5的C实现中，由于它们是
100   * 简单的位运算，可能出于效率的考虑把它们实现成了宏，在java中，我们把它们 实现成了private方法，名字保持了原来C中的。
101   */
102  private long F(long x, long y, long z) {
103   return (x & y) | ((~x) & z);
104  }
105 
106  private long G(long x, long y, long z) {
107   return (x & z) | (y & (~z));
108  }
109 
110  private long H(long x, long y, long z) {
111   return x ^ y ^ z;
112  }
113 
114  private long I(long x, long y, long z) {
115   return y ^ (x | (~z));
116  }
117 
118  /*
119   * FF,GG,HH和II将调用F,G,H,I进行近一步变换 FF, GG, HH, and II transformations for
120   * rounds 1, 2, 3, and 4. Rotation is separate from addition to prevent
121   * recomputation.
122   */
123  private long FF(long a, long b, long c, long d, long x, long s, long ac) {
124   a += F(b, c, d) + x + ac;
125   a = ((int) a << s) | ((int) a >>> (32 - s));
126   a += b;
127   return a;
128  }
129 
130  private long GG(long a, long b, long c, long d, long x, long s, long ac) {
131   a += G(b, c, d) + x + ac;
132   a = ((int) a << s) | ((int) a >>> (32 - s));
133   a += b;
134   return a;
135  }
136 
137  private long HH(long a, long b, long c, long d, long x, long s, long ac) {
138   a += H(b, c, d) + x + ac;
139   a = ((int) a << s) | ((int) a >>> (32 - s));
140   a += b;
141   return a;
142  }
143 
144  private long II(long a, long b, long c, long d, long x, long s, long ac) {
145   a += I(b, c, d) + x + ac;
146   a = ((int) a << s) | ((int) a >>> (32 - s));
147   a += b;
148   return a;
149  }
150 
151  /*
152   * md5Update是MD5的主计算过程，inbuf是要变换的字节串，inputlen是长度，这个
153   * 函数由getMD5ofStr调用，调用之前需要调用md5init，因此把它设计成private的
154   */
155  private void md5Update(byte[] inbuf, int inputLen) {
156   int i, index, partLen;
157   byte[] block = new byte[64];
158   index = (int) (count[0] >>> 3) & 0x3F;
159   // /* Update number of bits */
160   if ((count[0] += (inputLen << 3)) < (inputLen << 3))
161    count[1]++;
162   count[1] += (inputLen >>> 29);
163   partLen = 64 - index;
164   // Transform as many times as possible.
165   if (inputLen >= partLen) {
166    md5Memcpy(buffer, inbuf, index, 0, partLen);
167    md5Transform(buffer);
168    for (i = partLen; i + 63 < inputLen; i += 64) {
169     md5Memcpy(block, inbuf, 0, i, 64);
170     md5Transform(block);
171    }
172    index = 0;
173   } else
174    i = 0;
175   // /* Buffer remaining input */
176   md5Memcpy(buffer, inbuf, index, i, inputLen - i);
177  }
178 
179  /*
180   * md5Final整理和填写输出结果
181   */
182  private void md5Final() {
183   byte[] bits = new byte[8];
184   int index, padLen;
185   // /* Save number of bits */
186   Encode(bits, count, 8);
187   // /* Pad out to 56 mod 64.
188   index = (int) (count[0] >>> 3) & 0x3f;
189   padLen = (index < 56) ? (56 - index) : (120 - index);
190   md5Update(PADDING, padLen);
191   // /* Append length (before padding) */
192   md5Update(bits, 8);
193   // /* Store state in digest */
194   Encode(digest, state, 16);
195  }
196 
197  /*
198   * md5Memcpy是一个内部使用的byte数组的块拷贝函数，从input的inpos开始把len长度的
199   * 字节拷贝到output的outpos位置开始
200   */
201  private void md5Memcpy(byte[] output, byte[] input, int outpos, int inpos,
202    int len) {
203   int i;
204   for (i = 0; i < len; i++)
205    output[outpos + i] = input[inpos + i];
206  }
207 
208  /*
209   * md5Transform是MD5核心变换程序，有md5Update调用，block是分块的原始字节
210   */
211  private void md5Transform(byte block[]) {
212   long a = state[0], b = state[1], c = state[2], d = state[3];
213   long[] x = new long[16];
214   Decode(x, block, 64);
215   /* Round 1 */
216   a = FF(a, b, c, d, x[0], S11, 0xd76aa478L); /* 1 */
217   d = FF(d, a, b, c, x[1], S12, 0xe8c7b756L); /* 2 */
218   c = FF(c, d, a, b, x[2], S13, 0x242070dbL); /* 3 */
219   b = FF(b, c, d, a, x[3], S14, 0xc1bdceeeL); /* 4 */
220   a = FF(a, b, c, d, x[4], S11, 0xf57c0fafL); /* 5 */
221   d = FF(d, a, b, c, x[5], S12, 0x4787c62aL); /* 6 */
222   c = FF(c, d, a, b, x[6], S13, 0xa8304613L); /* 7 */
223   b = FF(b, c, d, a, x[7], S14, 0xfd469501L); /* 8 */
224   a = FF(a, b, c, d, x[8], S11, 0x698098d8L); /* 9 */
225   d = FF(d, a, b, c, x[9], S12, 0x8b44f7afL); /* 10 */
226   c = FF(c, d, a, b, x[10], S13, 0xffff5bb1L); /* 11 */
227   b = FF(b, c, d, a, x[11], S14, 0x895cd7beL); /* 12 */
228   a = FF(a, b, c, d, x[12], S11, 0x6b901122L); /* 13 */
229   d = FF(d, a, b, c, x[13], S12, 0xfd987193L); /* 14 */
230   c = FF(c, d, a, b, x[14], S13, 0xa679438eL); /* 15 */
231   b = FF(b, c, d, a, x[15], S14, 0x49b40821L); /* 16 */
232   /* Round 2 */
233   a = GG(a, b, c, d, x[1], S21, 0xf61e2562L); /* 17 */
234   d = GG(d, a, b, c, x[6], S22, 0xc040b340L); /* 18 */
235   c = GG(c, d, a, b, x[11], S23, 0x265e5a51L); /* 19 */
236   b = GG(b, c, d, a, x[0], S24, 0xe9b6c7aaL); /* 20 */
237   a = GG(a, b, c, d, x[5], S21, 0xd62f105dL); /* 21 */
238   d = GG(d, a, b, c, x[10], S22, 0x2441453L); /* 22 */
239   c = GG(c, d, a, b, x[15], S23, 0xd8a1e681L); /* 23 */
240   b = GG(b, c, d, a, x[4], S24, 0xe7d3fbc8L); /* 24 */
241   a = GG(a, b, c, d, x[9], S21, 0x21e1cde6L); /* 25 */
242   d = GG(d, a, b, c, x[14], S22, 0xc33707d6L); /* 26 */
243   c = GG(c, d, a, b, x[3], S23, 0xf4d50d87L); /* 27 */
244   b = GG(b, c, d, a, x[8], S24, 0x455a14edL); /* 28 */
245   a = GG(a, b, c, d, x[13], S21, 0xa9e3e905L); /* 29 */
246   d = GG(d, a, b, c, x[2], S22, 0xfcefa3f8L); /* 30 */
247   c = GG(c, d, a, b, x[7], S23, 0x676f02d9L); /* 31 */
248   b = GG(b, c, d, a, x[12], S24, 0x8d2a4c8aL); /* 32 */
249   /* Round 3 */
250   a = HH(a, b, c, d, x[5], S31, 0xfffa3942L); /* 33 */
251   d = HH(d, a, b, c, x[8], S32, 0x8771f681L); /* 34 */
252   c = HH(c, d, a, b, x[11], S33, 0x6d9d6122L); /* 35 */
253   b = HH(b, c, d, a, x[14], S34, 0xfde5380cL); /* 36 */
254   a = HH(a, b, c, d, x[1], S31, 0xa4beea44L); /* 37 */
255   d = HH(d, a, b, c, x[4], S32, 0x4bdecfa9L); /* 38 */
256   c = HH(c, d, a, b, x[7], S33, 0xf6bb4b60L); /* 39 */
257   b = HH(b, c, d, a, x[10], S34, 0xbebfbc70L); /* 40 */
258   a = HH(a, b, c, d, x[13], S31, 0x289b7ec6L); /* 41 */
259   d = HH(d, a, b, c, x[0], S32, 0xeaa127faL); /* 42 */
260   c = HH(c, d, a, b, x[3], S33, 0xd4ef3085L); /* 43 */
261   b = HH(b, c, d, a, x[6], S34, 0x4881d05L); /* 44 */
262   a = HH(a, b, c, d, x[9], S31, 0xd9d4d039L); /* 45 */
263   d = HH(d, a, b, c, x[12], S32, 0xe6db99e5L); /* 46 */
264   c = HH(c, d, a, b, x[15], S33, 0x1fa27cf8L); /* 47 */
265   b = HH(b, c, d, a, x[2], S34, 0xc4ac5665L); /* 48 */
266   /* Round 4 */
267   a = II(a, b, c, d, x[0], S41, 0xf4292244L); /* 49 */
268   d = II(d, a, b, c, x[7], S42, 0x432aff97L); /* 50 */
269   c = II(c, d, a, b, x[14], S43, 0xab9423a7L); /* 51 */
270   b = II(b, c, d, a, x[5], S44, 0xfc93a039L); /* 52 */
271   a = II(a, b, c, d, x[12], S41, 0x655b59c3L); /* 53 */
272   d = II(d, a, b, c, x[3], S42, 0x8f0ccc92L); /* 54 */
273   c = II(c, d, a, b, x[10], S43, 0xffeff47dL); /* 55 */
274   b = II(b, c, d, a, x[1], S44, 0x85845dd1L); /* 56 */
275   a = II(a, b, c, d, x[8], S41, 0x6fa87e4fL); /* 57 */
276   d = II(d, a, b, c, x[15], S42, 0xfe2ce6e0L); /* 58 */
277   c = II(c, d, a, b, x[6], S43, 0xa3014314L); /* 59 */
278   b = II(b, c, d, a, x[13], S44, 0x4e0811a1L); /* 60 */
279   a = II(a, b, c, d, x[4], S41, 0xf7537e82L); /* 61 */
280   d = II(d, a, b, c, x[11], S42, 0xbd3af235L); /* 62 */
281   c = II(c, d, a, b, x[2], S43, 0x2ad7d2bbL); /* 63 */
282   b = II(b, c, d, a, x[9], S44, 0xeb86d391L); /* 64 */
283   state[0] += a;
284   state[1] += b;
285   state[2] += c;
286   state[3] += d;
287  }
288 
289  /*
290   * Encode把long数组按顺序拆成byte数组，因为java的long类型是64bit的， 只拆低32bit，以适应原始C实现的用途
291   */
292  private void Encode(byte[] output, long[] input, int len) {
293   int i, j;
294   for (i = 0, j = 0; j < len; i++, j += 4) {
295    output[j] = (byte) (input[i] & 0xffL);
296    output[j + 1] = (byte) ((input[i] >>> 8) & 0xffL);
297    output[j + 2] = (byte) ((input[i] >>> 16) & 0xffL);
298    output[j + 3] = (byte) ((input[i] >>> 24) & 0xffL);
299   }
300  }
301 
302  /*
303   * Decode把byte数组按顺序合成成long数组，因为java的long类型是64bit的，
304   * 只合成低32bit，高32bit清零，以适应原始C实现的用途
305   */
306  private void Decode(long[] output, byte[] input, int len) {
307   int i, j;
308   for (i = 0, j = 0; j < len; i++, j += 4)
309    output[i] = b2iu(input[j]) | (b2iu(input[j + 1]) << 8)
310      | (b2iu(input[j + 2]) << 16) | (b2iu(input[j + 3]) << 24);
311   return;
312  }
313 
314  /*
315   * b2iu是我写的一个把byte按照不考虑正负号的原则的＂升位＂程序，因为java没有unsigned运算
316   */
317  public static long b2iu(byte b) {
318   return b < 0 ? b & 0x7F + 128 : b;
319  }
320 
321  /*
322   * byteHEX()，用来把一个byte类型的数转换成十六进制的ASCII表示，
323   * 因为java中的byte的toString无法实现这一点，我们又没有C语言中的 sprintf(outbuf,"%02X",ib)
324   */
325  public static String byteHEX(byte ib) {
326   char[] Digit = { ‘0‘, ‘1‘, ‘2‘, ‘3‘, ‘4‘, ‘5‘, ‘6‘, ‘7‘, ‘8‘, ‘9‘, ‘A‘,
327     ‘B‘, ‘C‘, ‘D‘, ‘E‘, ‘F‘ };
328   char[] ob = new char[2];
329   ob[0] = Digit[(ib >>> 4) & 0X0F];
330   ob[1] = Digit[ib & 0X0F];
331   String s = new String(ob);
332   return s;
333  }
334 }
335