Issue #7117, continued: Remove substitution of %g-style formatting for
[python.git] / Modules / md5.c
blob642f0bdc920235dcbc79731102a772b541a38531
1 /*
2 Copyright (C) 1999, 2000, 2002 Aladdin Enterprises. All rights reserved.
4 This software is provided 'as-is', without any express or implied
5 warranty. In no event will the authors be held liable for any damages
6 arising from the use of this software.
8 Permission is granted to anyone to use this software for any purpose,
9 including commercial applications, and to alter it and redistribute it
10 freely, subject to the following restrictions:
12 1. The origin of this software must not be misrepresented; you must not
13 claim that you wrote the original software. If you use this software
14 in a product, an acknowledgment in the product documentation would be
15 appreciated but is not required.
16 2. Altered source versions must be plainly marked as such, and must not be
17 misrepresented as being the original software.
18 3. This notice may not be removed or altered from any source distribution.
20 L. Peter Deutsch
21 ghost@aladdin.com
24 /* $Id: md5.c,v 1.6 2002/04/13 19:20:28 lpd Exp $ */
26 Independent implementation of MD5 (RFC 1321).
28 This code implements the MD5 Algorithm defined in RFC 1321, whose
29 text is available at
30 http://www.ietf.org/rfc/rfc1321.txt
31 The code is derived from the text of the RFC, including the test suite
32 (section A.5) but excluding the rest of Appendix A. It does not include
33 any code or documentation that is identified in the RFC as being
34 copyrighted.
36 The original and principal author of md5.c is L. Peter Deutsch
37 <ghost@aladdin.com>. Other authors are noted in the change history
38 that follows (in reverse chronological order):
40 2002-04-13 lpd Clarified derivation from RFC 1321; now handles byte order
41 either statically or dynamically; added missing #include <string.h>
42 in library.
43 2002-03-11 lpd Corrected argument list for main(), and added int return
44 type, in test program and T value program.
45 2002-02-21 lpd Added missing #include <stdio.h> in test program.
46 2000-07-03 lpd Patched to eliminate warnings about "constant is
47 unsigned in ANSI C, signed in traditional"; made test program
48 self-checking.
49 1999-11-04 lpd Edited comments slightly for automatic TOC extraction.
50 1999-10-18 lpd Fixed typo in header comment (ansi2knr rather than md5).
51 1999-05-03 lpd Original version.
54 #include "md5.h"
55 #include <string.h>
56 #include <limits.h>
58 #undef BYTE_ORDER /* 1 = big-endian, -1 = little-endian, 0 = unknown */
59 #ifdef ARCH_IS_BIG_ENDIAN
60 # define BYTE_ORDER (ARCH_IS_BIG_ENDIAN ? 1 : -1)
61 #else
62 # define BYTE_ORDER 0
63 #endif
65 #define T_MASK ((md5_word_t)~0)
66 #define T1 /* 0xd76aa478 */ (T_MASK ^ 0x28955b87)
67 #define T2 /* 0xe8c7b756 */ (T_MASK ^ 0x173848a9)
68 #define T3 0x242070db
69 #define T4 /* 0xc1bdceee */ (T_MASK ^ 0x3e423111)
70 #define T5 /* 0xf57c0faf */ (T_MASK ^ 0x0a83f050)
71 #define T6 0x4787c62a
72 #define T7 /* 0xa8304613 */ (T_MASK ^ 0x57cfb9ec)
73 #define T8 /* 0xfd469501 */ (T_MASK ^ 0x02b96afe)
74 #define T9 0x698098d8
75 #define T10 /* 0x8b44f7af */ (T_MASK ^ 0x74bb0850)
76 #define T11 /* 0xffff5bb1 */ (T_MASK ^ 0x0000a44e)
77 #define T12 /* 0x895cd7be */ (T_MASK ^ 0x76a32841)
78 #define T13 0x6b901122
79 #define T14 /* 0xfd987193 */ (T_MASK ^ 0x02678e6c)
80 #define T15 /* 0xa679438e */ (T_MASK ^ 0x5986bc71)
81 #define T16 0x49b40821
82 #define T17 /* 0xf61e2562 */ (T_MASK ^ 0x09e1da9d)
83 #define T18 /* 0xc040b340 */ (T_MASK ^ 0x3fbf4cbf)
84 #define T19 0x265e5a51
85 #define T20 /* 0xe9b6c7aa */ (T_MASK ^ 0x16493855)
86 #define T21 /* 0xd62f105d */ (T_MASK ^ 0x29d0efa2)
87 #define T22 0x02441453
88 #define T23 /* 0xd8a1e681 */ (T_MASK ^ 0x275e197e)
89 #define T24 /* 0xe7d3fbc8 */ (T_MASK ^ 0x182c0437)
90 #define T25 0x21e1cde6
91 #define T26 /* 0xc33707d6 */ (T_MASK ^ 0x3cc8f829)
92 #define T27 /* 0xf4d50d87 */ (T_MASK ^ 0x0b2af278)
93 #define T28 0x455a14ed
94 #define T29 /* 0xa9e3e905 */ (T_MASK ^ 0x561c16fa)
95 #define T30 /* 0xfcefa3f8 */ (T_MASK ^ 0x03105c07)
96 #define T31 0x676f02d9
97 #define T32 /* 0x8d2a4c8a */ (T_MASK ^ 0x72d5b375)
98 #define T33 /* 0xfffa3942 */ (T_MASK ^ 0x0005c6bd)
99 #define T34 /* 0x8771f681 */ (T_MASK ^ 0x788e097e)
100 #define T35 0x6d9d6122
101 #define T36 /* 0xfde5380c */ (T_MASK ^ 0x021ac7f3)
102 #define T37 /* 0xa4beea44 */ (T_MASK ^ 0x5b4115bb)
103 #define T38 0x4bdecfa9
104 #define T39 /* 0xf6bb4b60 */ (T_MASK ^ 0x0944b49f)
105 #define T40 /* 0xbebfbc70 */ (T_MASK ^ 0x4140438f)
106 #define T41 0x289b7ec6
107 #define T42 /* 0xeaa127fa */ (T_MASK ^ 0x155ed805)
108 #define T43 /* 0xd4ef3085 */ (T_MASK ^ 0x2b10cf7a)
109 #define T44 0x04881d05
110 #define T45 /* 0xd9d4d039 */ (T_MASK ^ 0x262b2fc6)
111 #define T46 /* 0xe6db99e5 */ (T_MASK ^ 0x1924661a)
112 #define T47 0x1fa27cf8
113 #define T48 /* 0xc4ac5665 */ (T_MASK ^ 0x3b53a99a)
114 #define T49 /* 0xf4292244 */ (T_MASK ^ 0x0bd6ddbb)
115 #define T50 0x432aff97
116 #define T51 /* 0xab9423a7 */ (T_MASK ^ 0x546bdc58)
117 #define T52 /* 0xfc93a039 */ (T_MASK ^ 0x036c5fc6)
118 #define T53 0x655b59c3
119 #define T54 /* 0x8f0ccc92 */ (T_MASK ^ 0x70f3336d)
120 #define T55 /* 0xffeff47d */ (T_MASK ^ 0x00100b82)
121 #define T56 /* 0x85845dd1 */ (T_MASK ^ 0x7a7ba22e)
122 #define T57 0x6fa87e4f
123 #define T58 /* 0xfe2ce6e0 */ (T_MASK ^ 0x01d3191f)
124 #define T59 /* 0xa3014314 */ (T_MASK ^ 0x5cfebceb)
125 #define T60 0x4e0811a1
126 #define T61 /* 0xf7537e82 */ (T_MASK ^ 0x08ac817d)
127 #define T62 /* 0xbd3af235 */ (T_MASK ^ 0x42c50dca)
128 #define T63 0x2ad7d2bb
129 #define T64 /* 0xeb86d391 */ (T_MASK ^ 0x14792c6e)
132 static void
133 md5_process(md5_state_t *pms, const md5_byte_t *data /*[64]*/)
135 md5_word_t
136 a = pms->abcd[0], b = pms->abcd[1],
137 c = pms->abcd[2], d = pms->abcd[3];
138 md5_word_t t;
139 #if BYTE_ORDER > 0
140 /* Define storage only for big-endian CPUs. */
141 md5_word_t X[16];
142 #else
143 /* Define storage for little-endian or both types of CPUs. */
144 md5_word_t xbuf[16];
145 const md5_word_t *X;
146 #endif
149 #if BYTE_ORDER == 0
151 * Determine dynamically whether this is a big-endian or
152 * little-endian machine, since we can use a more efficient
153 * algorithm on the latter.
155 static const int w = 1;
157 if (*((const md5_byte_t *)&w)) /* dynamic little-endian */
158 #endif
159 #if BYTE_ORDER <= 0 /* little-endian */
162 * On little-endian machines, we can process properly aligned
163 * data without copying it.
165 if (!((data - (const md5_byte_t *)0) & 3)) {
166 /* data are properly aligned */
167 X = (const md5_word_t *)data;
168 } else {
169 /* not aligned */
170 memcpy(xbuf, data, 64);
171 X = xbuf;
174 #endif
175 #if BYTE_ORDER == 0
176 else /* dynamic big-endian */
177 #endif
178 #if BYTE_ORDER >= 0 /* big-endian */
181 * On big-endian machines, we must arrange the bytes in the
182 * right order.
184 const md5_byte_t *xp = data;
185 int i;
187 # if BYTE_ORDER == 0
188 X = xbuf; /* (dynamic only) */
189 # else
190 # define xbuf X /* (static only) */
191 # endif
192 for (i = 0; i < 16; ++i, xp += 4)
193 xbuf[i] = xp[0] + (xp[1] << 8) + (xp[2] << 16) + (xp[3] << 24);
195 #endif
198 #define ROTATE_LEFT(x, n) (((x) << (n)) | ((x) >> (32 - (n))))
200 /* Round 1. */
201 /* Let [abcd k s i] denote the operation
202 a = b + ((a + F(b,c,d) + X[k] + T[i]) <<< s). */
203 #define F(x, y, z) (((x) & (y)) | (~(x) & (z)))
204 #define SET(a, b, c, d, k, s, Ti)\
205 t = a + F(b,c,d) + X[k] + Ti;\
206 a = ROTATE_LEFT(t, s) + b
207 /* Do the following 16 operations. */
208 SET(a, b, c, d, 0, 7, T1);
209 SET(d, a, b, c, 1, 12, T2);
210 SET(c, d, a, b, 2, 17, T3);
211 SET(b, c, d, a, 3, 22, T4);
212 SET(a, b, c, d, 4, 7, T5);
213 SET(d, a, b, c, 5, 12, T6);
214 SET(c, d, a, b, 6, 17, T7);
215 SET(b, c, d, a, 7, 22, T8);
216 SET(a, b, c, d, 8, 7, T9);
217 SET(d, a, b, c, 9, 12, T10);
218 SET(c, d, a, b, 10, 17, T11);
219 SET(b, c, d, a, 11, 22, T12);
220 SET(a, b, c, d, 12, 7, T13);
221 SET(d, a, b, c, 13, 12, T14);
222 SET(c, d, a, b, 14, 17, T15);
223 SET(b, c, d, a, 15, 22, T16);
224 #undef SET
226 /* Round 2. */
227 /* Let [abcd k s i] denote the operation
228 a = b + ((a + G(b,c,d) + X[k] + T[i]) <<< s). */
229 #define G(x, y, z) (((x) & (z)) | ((y) & ~(z)))
230 #define SET(a, b, c, d, k, s, Ti)\
231 t = a + G(b,c,d) + X[k] + Ti;\
232 a = ROTATE_LEFT(t, s) + b
233 /* Do the following 16 operations. */
234 SET(a, b, c, d, 1, 5, T17);
235 SET(d, a, b, c, 6, 9, T18);
236 SET(c, d, a, b, 11, 14, T19);
237 SET(b, c, d, a, 0, 20, T20);
238 SET(a, b, c, d, 5, 5, T21);
239 SET(d, a, b, c, 10, 9, T22);
240 SET(c, d, a, b, 15, 14, T23);
241 SET(b, c, d, a, 4, 20, T24);
242 SET(a, b, c, d, 9, 5, T25);
243 SET(d, a, b, c, 14, 9, T26);
244 SET(c, d, a, b, 3, 14, T27);
245 SET(b, c, d, a, 8, 20, T28);
246 SET(a, b, c, d, 13, 5, T29);
247 SET(d, a, b, c, 2, 9, T30);
248 SET(c, d, a, b, 7, 14, T31);
249 SET(b, c, d, a, 12, 20, T32);
250 #undef SET
252 /* Round 3. */
253 /* Let [abcd k s t] denote the operation
254 a = b + ((a + H(b,c,d) + X[k] + T[i]) <<< s). */
255 #define H(x, y, z) ((x) ^ (y) ^ (z))
256 #define SET(a, b, c, d, k, s, Ti)\
257 t = a + H(b,c,d) + X[k] + Ti;\
258 a = ROTATE_LEFT(t, s) + b
259 /* Do the following 16 operations. */
260 SET(a, b, c, d, 5, 4, T33);
261 SET(d, a, b, c, 8, 11, T34);
262 SET(c, d, a, b, 11, 16, T35);
263 SET(b, c, d, a, 14, 23, T36);
264 SET(a, b, c, d, 1, 4, T37);
265 SET(d, a, b, c, 4, 11, T38);
266 SET(c, d, a, b, 7, 16, T39);
267 SET(b, c, d, a, 10, 23, T40);
268 SET(a, b, c, d, 13, 4, T41);
269 SET(d, a, b, c, 0, 11, T42);
270 SET(c, d, a, b, 3, 16, T43);
271 SET(b, c, d, a, 6, 23, T44);
272 SET(a, b, c, d, 9, 4, T45);
273 SET(d, a, b, c, 12, 11, T46);
274 SET(c, d, a, b, 15, 16, T47);
275 SET(b, c, d, a, 2, 23, T48);
276 #undef SET
278 /* Round 4. */
279 /* Let [abcd k s t] denote the operation
280 a = b + ((a + I(b,c,d) + X[k] + T[i]) <<< s). */
281 #define I(x, y, z) ((y) ^ ((x) | ~(z)))
282 #define SET(a, b, c, d, k, s, Ti)\
283 t = a + I(b,c,d) + X[k] + Ti;\
284 a = ROTATE_LEFT(t, s) + b
285 /* Do the following 16 operations. */
286 SET(a, b, c, d, 0, 6, T49);
287 SET(d, a, b, c, 7, 10, T50);
288 SET(c, d, a, b, 14, 15, T51);
289 SET(b, c, d, a, 5, 21, T52);
290 SET(a, b, c, d, 12, 6, T53);
291 SET(d, a, b, c, 3, 10, T54);
292 SET(c, d, a, b, 10, 15, T55);
293 SET(b, c, d, a, 1, 21, T56);
294 SET(a, b, c, d, 8, 6, T57);
295 SET(d, a, b, c, 15, 10, T58);
296 SET(c, d, a, b, 6, 15, T59);
297 SET(b, c, d, a, 13, 21, T60);
298 SET(a, b, c, d, 4, 6, T61);
299 SET(d, a, b, c, 11, 10, T62);
300 SET(c, d, a, b, 2, 15, T63);
301 SET(b, c, d, a, 9, 21, T64);
302 #undef SET
304 /* Then perform the following additions. (That is increment each
305 of the four registers by the value it had before this block
306 was started.) */
307 pms->abcd[0] += a;
308 pms->abcd[1] += b;
309 pms->abcd[2] += c;
310 pms->abcd[3] += d;
313 void
314 md5_init(md5_state_t *pms)
316 pms->count[0] = pms->count[1] = 0;
317 pms->abcd[0] = 0x67452301;
318 pms->abcd[1] = /*0xefcdab89*/ T_MASK ^ 0x10325476;
319 pms->abcd[2] = /*0x98badcfe*/ T_MASK ^ 0x67452301;
320 pms->abcd[3] = 0x10325476;
323 void
324 md5_append(md5_state_t *pms, const md5_byte_t *data, unsigned int nbytes)
326 const md5_byte_t *p = data;
327 unsigned int left = nbytes;
328 unsigned int offset = (pms->count[0] >> 3) & 63;
329 md5_word_t nbits = (md5_word_t)(nbytes << 3);
331 if (nbytes <= 0)
332 return;
334 /* this special case is handled recursively */
335 if (nbytes > INT_MAX - offset) {
336 unsigned int overlap;
338 /* handle the append in two steps to prevent overflow */
339 overlap = 64 - offset;
341 md5_append(pms, data, overlap);
342 md5_append(pms, data + overlap, nbytes - overlap);
343 return;
346 /* Update the message length. */
347 pms->count[1] += nbytes >> 29;
348 pms->count[0] += nbits;
349 if (pms->count[0] < nbits)
350 pms->count[1]++;
352 /* Process an initial partial block. */
353 if (offset) {
354 unsigned int copy = (offset + nbytes > 64 ? 64 - offset : nbytes);
356 memcpy(pms->buf + offset, p, copy);
357 if (offset + copy < 64)
358 return;
359 p += copy;
360 left -= copy;
361 md5_process(pms, pms->buf);
364 /* Process full blocks. */
365 for (; left >= 64; p += 64, left -= 64)
366 md5_process(pms, p);
368 /* Process a final partial block. */
369 if (left)
370 memcpy(pms->buf, p, left);
373 void
374 md5_finish(md5_state_t *pms, md5_byte_t digest[16])
376 static const md5_byte_t pad[64] = {
377 0x80, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
378 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
379 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
380 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0
382 md5_byte_t data[8];
383 int i;
385 /* Save the length before padding. */
386 for (i = 0; i < 8; ++i)
387 data[i] = (md5_byte_t)(pms->count[i >> 2] >> ((i & 3) << 3));
388 /* Pad to 56 bytes mod 64. */
389 md5_append(pms, pad, ((55 - (pms->count[0] >> 3)) & 63) + 1);
390 /* Append the length. */
391 md5_append(pms, data, 8);
392 for (i = 0; i < 16; ++i)
393 digest[i] = (md5_byte_t)(pms->abcd[i >> 2] >> ((i & 3) << 3));