src/test_md5.c

   1 /*
   2 ** 2017-10-13
   3 **
   4 ** The author disclaims copyright to this source code.  In place of
   5 ** a legal notice, here is a blessing:
   6 **
   7 **    May you do good and not evil.
   8 **    May you find forgiveness for yourself and forgive others.
   9 **    May you share freely, never taking more than you give.
  10 **
  11 *************************************************************************
  12 **
  13 ** This file contains code to implement an MD5 extension to TCL.
  14 */
  15 #include "sqlite3.h"
  16 #include <stdlib.h>
  17 #include <string.h>
  18 #include "sqlite3.h"
  19 #if defined(INCLUDE_SQLITE_TCL_H)
  20 # include "sqlite_tcl.h"
  21 #else
  22 # include "tcl.h"
  23 # ifndef SQLITE_TCLAPI
  24 #  define SQLITE_TCLAPI
  25 # endif
  26 #endif
  27
  28 /*
  29  * This code implements the MD5 message-digest algorithm.
  30  * The algorithm is due to Ron Rivest.  This code was
  31  * written by Colin Plumb in 1993, no copyright is claimed.
  32  * This code is in the public domain; do with it what you wish.
  33  *
  34  * Equivalent code is available from RSA Data Security, Inc.
  35  * This code has been tested against that, and is equivalent,
  36  * except that you don't need to include two pages of legalese
  37  * with every copy.
  38  *
  39  * To compute the message digest of a chunk of bytes, declare an
  40  * MD5Context structure, pass it to MD5Init, call MD5Update as
  41  * needed on buffers full of bytes, and then call MD5Final, which
  42  * will fill a supplied 16-byte array with the digest.
  43  */
  44
  45 /*
  46  * If compiled on a machine that doesn't have a 32-bit integer,
  47  * you just set "uint32" to the appropriate datatype for an
  48  * unsigned 32-bit integer.  For example:
  49  *
  50  *       cc -Duint32='unsigned long' md5.c
  51  *
  52  */
  53 #ifndef uint32
  54 #  define uint32 unsigned int
  55 #endif
  56
  57 struct MD5Context {
  58   int isInit;
  59   uint32 buf[4];
  60   uint32 bits[2];
  61   unsigned char in[64];
  62 };
  63 typedef struct MD5Context MD5Context;
  64
  65 /*
  66  * Note: this code is harmless on little-endian machines.
  67  */
  68 static void byteReverse (unsigned char *buf, unsigned longs){
  69         uint32 t;
  70         do {
  71                 t = (uint32)((unsigned)buf[3]<<8 | buf[2]) << 16 |
  72                             ((unsigned)buf[1]<<8 | buf[0]);
  73                 *(uint32 *)buf = t;
  74                 buf += 4;
  75         } while (--longs);
  76 }
  77 /* The four core functions - F1 is optimized somewhat */
  78
  79 /* #define F1(x, y, z) (x & y | ~x & z) */
  80 #define F1(x, y, z) (z ^ (x & (y ^ z)))
  81 #define F2(x, y, z) F1(z, x, y)
  82 #define F3(x, y, z) (x ^ y ^ z)
  83 #define F4(x, y, z) (y ^ (x | ~z))
  84
  85 /* This is the central step in the MD5 algorithm. */
  86 #define MD5STEP(f, w, x, y, z, data, s) \
  87         ( w += f(x, y, z) + data,  w = w<<s | w>>(32-s),  w += x )
  88
  89 /*
  90  * The core of the MD5 algorithm, this alters an existing MD5 hash to
  91  * reflect the addition of 16 longwords of new data.  MD5Update blocks
  92  * the data and converts bytes into longwords for this routine.
  93  */
  94 static void MD5Transform(uint32 buf[4], const uint32 in[16]){
  95         register uint32 a, b, c, d;
  96
  97         a = buf[0];
  98         b = buf[1];
  99         c = buf[2];
 100         d = buf[3];
 101
 102         MD5STEP(F1, a, b, c, d, in[ 0]+0xd76aa478,  7);
 103         MD5STEP(F1, d, a, b, c, in[ 1]+0xe8c7b756, 12);
 104         MD5STEP(F1, c, d, a, b, in[ 2]+0x242070db, 17);
 105         MD5STEP(F1, b, c, d, a, in[ 3]+0xc1bdceee, 22);
 106         MD5STEP(F1, a, b, c, d, in[ 4]+0xf57c0faf,  7);
 107         MD5STEP(F1, d, a, b, c, in[ 5]+0x4787c62a, 12);
 108         MD5STEP(F1, c, d, a, b, in[ 6]+0xa8304613, 17);
 109         MD5STEP(F1, b, c, d, a, in[ 7]+0xfd469501, 22);
 110         MD5STEP(F1, a, b, c, d, in[ 8]+0x698098d8,  7);
 111         MD5STEP(F1, d, a, b, c, in[ 9]+0x8b44f7af, 12);
 112         MD5STEP(F1, c, d, a, b, in[10]+0xffff5bb1, 17);
 113         MD5STEP(F1, b, c, d, a, in[11]+0x895cd7be, 22);
 114         MD5STEP(F1, a, b, c, d, in[12]+0x6b901122,  7);
 115         MD5STEP(F1, d, a, b, c, in[13]+0xfd987193, 12);
 116         MD5STEP(F1, c, d, a, b, in[14]+0xa679438e, 17);
 117         MD5STEP(F1, b, c, d, a, in[15]+0x49b40821, 22);
 118
 119         MD5STEP(F2, a, b, c, d, in[ 1]+0xf61e2562,  5);
 120         MD5STEP(F2, d, a, b, c, in[ 6]+0xc040b340,  9);
 121         MD5STEP(F2, c, d, a, b, in[11]+0x265e5a51, 14);
 122         MD5STEP(F2, b, c, d, a, in[ 0]+0xe9b6c7aa, 20);
 123         MD5STEP(F2, a, b, c, d, in[ 5]+0xd62f105d,  5);
 124         MD5STEP(F2, d, a, b, c, in[10]+0x02441453,  9);
 125         MD5STEP(F2, c, d, a, b, in[15]+0xd8a1e681, 14);
 126         MD5STEP(F2, b, c, d, a, in[ 4]+0xe7d3fbc8, 20);
 127         MD5STEP(F2, a, b, c, d, in[ 9]+0x21e1cde6,  5);
 128         MD5STEP(F2, d, a, b, c, in[14]+0xc33707d6,  9);
 129         MD5STEP(F2, c, d, a, b, in[ 3]+0xf4d50d87, 14);
 130         MD5STEP(F2, b, c, d, a, in[ 8]+0x455a14ed, 20);
 131         MD5STEP(F2, a, b, c, d, in[13]+0xa9e3e905,  5);
 132         MD5STEP(F2, d, a, b, c, in[ 2]+0xfcefa3f8,  9);
 133         MD5STEP(F2, c, d, a, b, in[ 7]+0x676f02d9, 14);
 134         MD5STEP(F2, b, c, d, a, in[12]+0x8d2a4c8a, 20);
 135
 136         MD5STEP(F3, a, b, c, d, in[ 5]+0xfffa3942,  4);
 137         MD5STEP(F3, d, a, b, c, in[ 8]+0x8771f681, 11);
 138         MD5STEP(F3, c, d, a, b, in[11]+0x6d9d6122, 16);
 139         MD5STEP(F3, b, c, d, a, in[14]+0xfde5380c, 23);
 140         MD5STEP(F3, a, b, c, d, in[ 1]+0xa4beea44,  4);
 141         MD5STEP(F3, d, a, b, c, in[ 4]+0x4bdecfa9, 11);
 142         MD5STEP(F3, c, d, a, b, in[ 7]+0xf6bb4b60, 16);
 143         MD5STEP(F3, b, c, d, a, in[10]+0xbebfbc70, 23);
 144         MD5STEP(F3, a, b, c, d, in[13]+0x289b7ec6,  4);
 145         MD5STEP(F3, d, a, b, c, in[ 0]+0xeaa127fa, 11);
 146         MD5STEP(F3, c, d, a, b, in[ 3]+0xd4ef3085, 16);
 147         MD5STEP(F3, b, c, d, a, in[ 6]+0x04881d05, 23);
 148         MD5STEP(F3, a, b, c, d, in[ 9]+0xd9d4d039,  4);
 149         MD5STEP(F3, d, a, b, c, in[12]+0xe6db99e5, 11);
 150         MD5STEP(F3, c, d, a, b, in[15]+0x1fa27cf8, 16);
 151         MD5STEP(F3, b, c, d, a, in[ 2]+0xc4ac5665, 23);
 152
 153         MD5STEP(F4, a, b, c, d, in[ 0]+0xf4292244,  6);
 154         MD5STEP(F4, d, a, b, c, in[ 7]+0x432aff97, 10);
 155         MD5STEP(F4, c, d, a, b, in[14]+0xab9423a7, 15);
 156         MD5STEP(F4, b, c, d, a, in[ 5]+0xfc93a039, 21);
 157         MD5STEP(F4, a, b, c, d, in[12]+0x655b59c3,  6);
 158         MD5STEP(F4, d, a, b, c, in[ 3]+0x8f0ccc92, 10);
 159         MD5STEP(F4, c, d, a, b, in[10]+0xffeff47d, 15);
 160         MD5STEP(F4, b, c, d, a, in[ 1]+0x85845dd1, 21);
 161         MD5STEP(F4, a, b, c, d, in[ 8]+0x6fa87e4f,  6);
 162         MD5STEP(F4, d, a, b, c, in[15]+0xfe2ce6e0, 10);
 163         MD5STEP(F4, c, d, a, b, in[ 6]+0xa3014314, 15);
 164         MD5STEP(F4, b, c, d, a, in[13]+0x4e0811a1, 21);
 165         MD5STEP(F4, a, b, c, d, in[ 4]+0xf7537e82,  6);
 166         MD5STEP(F4, d, a, b, c, in[11]+0xbd3af235, 10);
 167         MD5STEP(F4, c, d, a, b, in[ 2]+0x2ad7d2bb, 15);
 168         MD5STEP(F4, b, c, d, a, in[ 9]+0xeb86d391, 21);
 169
 170         buf[0] += a;
 171         buf[1] += b;
 172         buf[2] += c;
 173         buf[3] += d;
 174 }
 175
 176 /*
 177  * Start MD5 accumulation.  Set bit count to 0 and buffer to mysterious
 178  * initialization constants.
 179  */
 180 static void MD5Init(MD5Context *ctx){
 181         ctx->isInit = 1;
 182         ctx->buf[0] = 0x67452301;
 183         ctx->buf[1] = 0xefcdab89;
 184         ctx->buf[2] = 0x98badcfe;
 185         ctx->buf[3] = 0x10325476;
 186         ctx->bits[0] = 0;
 187         ctx->bits[1] = 0;
 188 }
 189
 190 /*
 191  * Update context to reflect the concatenation of another buffer full
 192  * of bytes.
 193  */
 194 static
 195 void MD5Update(MD5Context *ctx, const unsigned char *buf, unsigned int len){
 196         uint32 t;
 197
 198         /* Update bitcount */
 199
 200         t = ctx->bits[0];
 201         if ((ctx->bits[0] = t + ((uint32)len << 3)) < t)
 202                 ctx->bits[1]++; /* Carry from low to high */
 203         ctx->bits[1] += len >> 29;
 204
 205         t = (t >> 3) & 0x3f;    /* Bytes already in shsInfo->data */
 206
 207         /* Handle any leading odd-sized chunks */
 208
 209         if ( t ) {
 210                 unsigned char *p = (unsigned char *)ctx->in + t;
 211
 212                 t = 64-t;
 213                 if (len < t) {
 214                         memcpy(p, buf, len);
 215                         return;
 216                 }
 217                 memcpy(p, buf, t);
 218                 byteReverse(ctx->in, 16);
 219                 MD5Transform(ctx->buf, (uint32 *)ctx->in);
 220                 buf += t;
 221                 len -= t;
 222         }
 223
 224         /* Process data in 64-byte chunks */
 225
 226         while (len >= 64) {
 227                 memcpy(ctx->in, buf, 64);
 228                 byteReverse(ctx->in, 16);
 229                 MD5Transform(ctx->buf, (uint32 *)ctx->in);
 230                 buf += 64;
 231                 len -= 64;
 232         }
 233
 234         /* Handle any remaining bytes of data. */
 235
 236         memcpy(ctx->in, buf, len);
 237 }
 238
 239 /*
 240  * Final wrapup - pad to 64-byte boundary with the bit pattern
 241  * 1 0* (64-bit count of bits processed, MSB-first)
 242  */
 243 static void MD5Final(unsigned char digest[16], MD5Context *ctx){
 244         unsigned count;
 245         unsigned char *p;
 246
 247         /* Compute number of bytes mod 64 */
 248         count = (ctx->bits[0] >> 3) & 0x3F;
 249
 250         /* Set the first char of padding to 0x80.  This is safe since there is
 251            always at least one byte free */
 252         p = ctx->in + count;
 253         *p++ = 0x80;
 254
 255         /* Bytes of padding needed to make 64 bytes */
 256         count = 64 - 1 - count;
 257
 258         /* Pad out to 56 mod 64 */
 259         if (count < 8) {
 260                 /* Two lots of padding:  Pad the first block to 64 bytes */
 261                 memset(p, 0, count);
 262                 byteReverse(ctx->in, 16);
 263                 MD5Transform(ctx->buf, (uint32 *)ctx->in);
 264
 265                 /* Now fill the next block with 56 bytes */
 266                 memset(ctx->in, 0, 56);
 267         } else {
 268                 /* Pad block to 56 bytes */
 269                 memset(p, 0, count-8);
 270         }
 271         byteReverse(ctx->in, 14);
 272
 273         /* Append length in bits and transform */
 274         memcpy(ctx->in + 14*4, ctx->bits, 8);
 275
 276         MD5Transform(ctx->buf, (uint32 *)ctx->in);
 277         byteReverse((unsigned char *)ctx->buf, 4);
 278         memcpy(digest, ctx->buf, 16);
 279 }
 280
 281 /*
 282 ** Convert a 128-bit MD5 digest into a 32-digit base-16 number.
 283 */
 284 static void MD5DigestToBase16(unsigned char *digest, char *zBuf){
 285   static char const zEncode[] = "0123456789abcdef";
 286   int i, j;
 287
 288   for(j=i=0; i<16; i++){
 289     int a = digest[i];
 290     zBuf[j++] = zEncode[(a>>4)&0xf];
 291     zBuf[j++] = zEncode[a & 0xf];
 292   }
 293   zBuf[j] = 0;
 294 }
 295
 296
 297 /*
 298 ** Convert a 128-bit MD5 digest into sequences of eight 5-digit integers
 299 ** each representing 16 bits of the digest and separated from each
 300 ** other by a "-" character.
 301 */
 302 static void MD5DigestToBase10x8(unsigned char digest[16], char zDigest[50]){
 303   int i, j;
 304   unsigned int x;
 305   for(i=j=0; i<16; i+=2){
 306     x = digest[i]*256 + digest[i+1];
 307     if( i>0 ) zDigest[j++] = '-';
 308     sqlite3_snprintf(50-j, &zDigest[j], "%05u", x);
 309     j += 5;
 310   }
 311   zDigest[j] = 0;
 312 }
 313
 314 /*
 315 ** A TCL command for md5.  The argument is the text to be hashed.  The
 316 ** Result is the hash in base64.
 317 */
 318 static int SQLITE_TCLAPI md5_cmd(
 319   void*cd,
 320   Tcl_Interp *interp,
 321   int argc,
 322   const char **argv
 323 ){
 324   MD5Context ctx;
 325   unsigned char digest[16];
 326   char zBuf[50];
 327   void (*converter)(unsigned char*, char*);
 328
 329   if( argc!=2 ){
 330     Tcl_AppendResult(interp,"wrong # args: should be \"", argv[0],
 331         " TEXT\"", (char*)0);
 332     return TCL_ERROR;
 333   }
 334   MD5Init(&ctx);
 335   MD5Update(&ctx, (unsigned char*)argv[1], (unsigned)strlen(argv[1]));
 336   MD5Final(digest, &ctx);
 337   converter = (void(*)(unsigned char*,char*))cd;
 338   converter(digest, zBuf);
 339   Tcl_AppendResult(interp, zBuf, (char*)0);
 340   return TCL_OK;
 341 }
 342
 343 /*
 344 ** A TCL command to take the md5 hash of a file.  The argument is the
 345 ** name of the file.
 346 */
 347 static int SQLITE_TCLAPI md5file_cmd(
 348   void*cd,
 349   Tcl_Interp *interp,
 350   int argc,
 351   const char **argv
 352 ){
 353   FILE *in;
 354   int ofst;
 355   int amt;
 356   MD5Context ctx;
 357   void (*converter)(unsigned char*, char*);
 358   unsigned char digest[16];
 359   char zBuf[10240];
 360
 361   if( argc!=2 && argc!=4 ){
 362     Tcl_AppendResult(interp,"wrong # args: should be \"", argv[0],
 363         " FILENAME [OFFSET AMT]\"", (char*)0);
 364     return TCL_ERROR;
 365   }
 366   if( argc==4 ){
 367     ofst = atoi(argv[2]);
 368     amt = atoi(argv[3]);
 369   }else{
 370     ofst = 0;
 371     amt = 2147483647;
 372   }
 373   in = fopen(argv[1],"rb");
 374   if( in==0 ){
 375     Tcl_AppendResult(interp,"unable to open file \"", argv[1],
 376          "\" for reading", (char*)0);
 377     return TCL_ERROR;
 378   }
 379   fseek(in, ofst, SEEK_SET);
 380   MD5Init(&ctx);
 381   while( amt>0 ){
 382     int n;
 383     n = (int)fread(zBuf, 1, sizeof(zBuf)<=amt ? sizeof(zBuf) : amt, in);
 384     if( n<=0 ) break;
 385     MD5Update(&ctx, (unsigned char*)zBuf, (unsigned)n);
 386     amt -= n;
 387   }
 388   fclose(in);
 389   MD5Final(digest, &ctx);
 390   converter = (void(*)(unsigned char*,char*))cd;
 391   converter(digest, zBuf);
 392   Tcl_AppendResult(interp, zBuf, (char*)0);
 393   return TCL_OK;
 394 }
 395
 396 /*
 397 ** Register the four new TCL commands for generating MD5 checksums
 398 ** with the TCL interpreter.
 399 */
 400 int Md5_Init(Tcl_Interp *interp){
 401   Tcl_CreateCommand(interp, "md5", (Tcl_CmdProc*)md5_cmd,
 402                     MD5DigestToBase16, 0);
 403   Tcl_CreateCommand(interp, "md5-10x8", (Tcl_CmdProc*)md5_cmd,
 404                     MD5DigestToBase10x8, 0);
 405   Tcl_CreateCommand(interp, "md5file", (Tcl_CmdProc*)md5file_cmd,
 406                     MD5DigestToBase16, 0);
 407   Tcl_CreateCommand(interp, "md5file-10x8", (Tcl_CmdProc*)md5file_cmd,
 408                     MD5DigestToBase10x8, 0);
 409   return TCL_OK;
 410 }
 411
 412 /*
 413 ** During testing, the special md5sum() aggregate function is available.
 414 ** inside SQLite.  The following routines implement that function.
 415 */
 416 static void md5step(sqlite3_context *context, int argc, sqlite3_value **argv){
 417   MD5Context *p;
 418   int i;
 419   if( argc<1 ) return;
 420   p = sqlite3_aggregate_context(context, sizeof(*p));
 421   if( p==0 ) return;
 422   if( !p->isInit ){
 423     MD5Init(p);
 424   }
 425   for(i=0; i<argc; i++){
 426     const char *zData = (char*)sqlite3_value_text(argv[i]);
 427     if( zData ){
 428       MD5Update(p, (unsigned char*)zData, (int)strlen(zData));
 429     }
 430   }
 431 }
 432 static void md5finalize(sqlite3_context *context){
 433   MD5Context *p;
 434   unsigned char digest[16];
 435   char zBuf[33];
 436   p = sqlite3_aggregate_context(context, sizeof(*p));
 437   MD5Final(digest,p);
 438   MD5DigestToBase16(digest, zBuf);
 439   sqlite3_result_text(context, zBuf, -1, SQLITE_TRANSIENT);
 440 }
 441 int Md5_Register(
 442   sqlite3 *db,
 443   char **pzErrMsg,
 444   const sqlite3_api_routines *pThunk
 445 ){
 446   int rc = sqlite3_create_function(db, "md5sum", -1, SQLITE_UTF8, 0, 0,
 447                                  md5step, md5finalize);
 448   sqlite3_overload_function(db, "md5sum", -1);  /* To exercise this API */
 449   return rc;
 450 }