xdiff/xutils.c

   1 /*
   2  *  LibXDiff by Davide Libenzi ( File Differential Library )
   3  *  Copyright (C) 2003  Davide Libenzi
   4  *
   5  *  This library is free software; you can redistribute it and/or
   6  *  modify it under the terms of the GNU Lesser General Public
   7  *  License as published by the Free Software Foundation; either
   8  *  version 2.1 of the License, or (at your option) any later version.
   9  *
  10  *  This library is distributed in the hope that it will be useful,
  11  *  but WITHOUT ANY WARRANTY; without even the implied warranty of
  12  *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  13  *  Lesser General Public License for more details.
  14  *
  15  *  You should have received a copy of the GNU Lesser General Public
  16  *  License along with this library; if not, write to the Free Software
  17  *  Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
  18  *
  19  *  Davide Libenzi <davidel@xmailserver.org>
  20  *
  21  */
  22
  23 #include "xinclude.h"
  24
  25
  26
  27 #define XDL_GUESS_NLINES 256
  28
  29
  30
  31
  32 long xdl_bogosqrt(long n) {
  33         long i;
  34
  35         /*
  36          * Classical integer square root approximation using shifts.
  37          */
  38         for (i = 1; n > 0; n >>= 2)
  39                 i <<= 1;
  40
  41         return i;
  42 }
  43
  44
  45 int xdl_emit_diffrec(char const *rec, long size, char const *pre, long psize,
  46                      xdemitcb_t *ecb) {
  47         int i = 2;
  48         mmbuffer_t mb[3];
  49
  50         mb[0].ptr = (char *) pre;
  51         mb[0].size = psize;
  52         mb[1].ptr = (char *) rec;
  53         mb[1].size = size;
  54         if (size > 0 && rec[size - 1] != '\n') {
  55                 mb[2].ptr = (char *) "\n\\ No newline at end of file\n";
  56                 mb[2].size = strlen(mb[2].ptr);
  57                 i++;
  58         }
  59         if (ecb->outf(ecb->priv, mb, i) < 0) {
  60
  61                 return -1;
  62         }
  63
  64         return 0;
  65 }
  66
  67 void *xdl_mmfile_first(mmfile_t *mmf, long *size)
  68 {
  69         *size = mmf->size;
  70         return mmf->ptr;
  71 }
  72
  73
  74 void *xdl_mmfile_next(mmfile_t *mmf, long *size)
  75 {
  76         return NULL;
  77 }
  78
  79
  80 long xdl_mmfile_size(mmfile_t *mmf)
  81 {
  82         return mmf->size;
  83 }
  84
  85
  86 int xdl_cha_init(chastore_t *cha, long isize, long icount) {
  87
  88         cha->head = cha->tail = NULL;
  89         cha->isize = isize;
  90         cha->nsize = icount * isize;
  91         cha->ancur = cha->sncur = NULL;
  92         cha->scurr = 0;
  93
  94         return 0;
  95 }
  96
  97
  98 void xdl_cha_free(chastore_t *cha) {
  99         chanode_t *cur, *tmp;
 100
 101         for (cur = cha->head; (tmp = cur) != NULL;) {
 102                 cur = cur->next;
 103                 xdl_free(tmp);
 104         }
 105 }
 106
 107
 108 void *xdl_cha_alloc(chastore_t *cha) {
 109         chanode_t *ancur;
 110         void *data;
 111
 112         if (!(ancur = cha->ancur) || ancur->icurr == cha->nsize) {
 113                 if (!(ancur = (chanode_t *) xdl_malloc(sizeof(chanode_t) + cha->nsize))) {
 114
 115                         return NULL;
 116                 }
 117                 ancur->icurr = 0;
 118                 ancur->next = NULL;
 119                 if (cha->tail)
 120                         cha->tail->next = ancur;
 121                 if (!cha->head)
 122                         cha->head = ancur;
 123                 cha->tail = ancur;
 124                 cha->ancur = ancur;
 125         }
 126
 127         data = (char *) ancur + sizeof(chanode_t) + ancur->icurr;
 128         ancur->icurr += cha->isize;
 129
 130         return data;
 131 }
 132
 133
 134 void *xdl_cha_first(chastore_t *cha) {
 135         chanode_t *sncur;
 136
 137         if (!(cha->sncur = sncur = cha->head))
 138                 return NULL;
 139
 140         cha->scurr = 0;
 141
 142         return (char *) sncur + sizeof(chanode_t) + cha->scurr;
 143 }
 144
 145
 146 void *xdl_cha_next(chastore_t *cha) {
 147         chanode_t *sncur;
 148
 149         if (!(sncur = cha->sncur))
 150                 return NULL;
 151         cha->scurr += cha->isize;
 152         if (cha->scurr == sncur->icurr) {
 153                 if (!(sncur = cha->sncur = sncur->next))
 154                         return NULL;
 155                 cha->scurr = 0;
 156         }
 157
 158         return (char *) sncur + sizeof(chanode_t) + cha->scurr;
 159 }
 160
 161
 162 long xdl_guess_lines(mmfile_t *mf) {
 163         long nl = 0, size, tsize = 0;
 164         char const *data, *cur, *top;
 165
 166         if ((cur = data = xdl_mmfile_first(mf, &size)) != NULL) {
 167                 for (top = data + size; nl < XDL_GUESS_NLINES;) {
 168                         if (cur >= top) {
 169                                 tsize += (long) (cur - data);
 170                                 if (!(cur = data = xdl_mmfile_next(mf, &size)))
 171                                         break;
 172                                 top = data + size;
 173                         }
 174                         nl++;
 175                         if (!(cur = memchr(cur, '\n', top - cur)))
 176                                 cur = top;
 177                         else
 178                                 cur++;
 179                 }
 180                 tsize += (long) (cur - data);
 181         }
 182
 183         if (nl && tsize)
 184                 nl = xdl_mmfile_size(mf) / (tsize / nl);
 185
 186         return nl + 1;
 187 }
 188
 189 int xdl_recmatch(const char *l1, long s1, const char *l2, long s2, long flags)
 190 {
 191         int i1, i2;
 192
 193         if (s1 == s2 && !memcmp(l1, l2, s1))
 194                 return 1;
 195         if (!(flags & XDF_WHITESPACE_FLAGS))
 196                 return 0;
 197
 198         i1 = 0;
 199         i2 = 0;
 200
 201         /*
 202          * -w matches everything that matches with -b, and -b in turn
 203          * matches everything that matches with --ignore-space-at-eol.
 204          *
 205          * Each flavor of ignoring needs different logic to skip whitespaces
 206          * while we have both sides to compare.
 207          */
 208         if (flags & XDF_IGNORE_WHITESPACE) {
 209                 goto skip_ws;
 210                 while (i1 < s1 && i2 < s2) {
 211                         if (l1[i1++] != l2[i2++])
 212                                 return 0;
 213                 skip_ws:
 214                         while (i1 < s1 && XDL_ISSPACE(l1[i1]))
 215                                 i1++;
 216                         while (i2 < s2 && XDL_ISSPACE(l2[i2]))
 217                                 i2++;
 218                 }
 219         } else if (flags & XDF_IGNORE_WHITESPACE_CHANGE) {
 220                 while (i1 < s1 && i2 < s2) {
 221                         if (XDL_ISSPACE(l1[i1]) && XDL_ISSPACE(l2[i2])) {
 222                                 /* Skip matching spaces and try again */
 223                                 while (i1 < s1 && XDL_ISSPACE(l1[i1]))
 224                                         i1++;
 225                                 while (i2 < s2 && XDL_ISSPACE(l2[i2]))
 226                                         i2++;
 227                                 continue;
 228                         }
 229                         if (l1[i1++] != l2[i2++])
 230                                 return 0;
 231                 }
 232         } else if (flags & XDF_IGNORE_WHITESPACE_AT_EOL) {
 233                 while (i1 < s1 && i2 < s2 && l1[i1++] == l2[i2++])
 234                         ; /* keep going */
 235         }
 236
 237         /*
 238          * After running out of one side, the remaining side must have
 239          * nothing but whitespace for the lines to match.  Note that
 240          * ignore-whitespace-at-eol case may break out of the loop
 241          * while there still are characters remaining on both lines.
 242          */
 243         if (i1 < s1) {
 244                 while (i1 < s1 && XDL_ISSPACE(l1[i1]))
 245                         i1++;
 246                 if (s1 != i1)
 247                         return 0;
 248         }
 249         if (i2 < s2) {
 250                 while (i2 < s2 && XDL_ISSPACE(l2[i2]))
 251                         i2++;
 252                 return (s2 == i2);
 253         }
 254         return 1;
 255 }
 256
 257 static unsigned long xdl_hash_record_with_whitespace(char const **data,
 258                 char const *top, long flags) {
 259         unsigned long ha = 5381;
 260         char const *ptr = *data;
 261
 262         for (; ptr < top && *ptr != '\n'; ptr++) {
 263                 if (XDL_ISSPACE(*ptr)) {
 264                         const char *ptr2 = ptr;
 265                         int at_eol;
 266                         while (ptr + 1 < top && XDL_ISSPACE(ptr[1])
 267                                         && ptr[1] != '\n')
 268                                 ptr++;
 269                         at_eol = (top <= ptr + 1 || ptr[1] == '\n');
 270                         if (flags & XDF_IGNORE_WHITESPACE)
 271                                 ; /* already handled */
 272                         else if (flags & XDF_IGNORE_WHITESPACE_CHANGE
 273                                  && !at_eol) {
 274                                 ha += (ha << 5);
 275                                 ha ^= (unsigned long) ' ';
 276                         }
 277                         else if (flags & XDF_IGNORE_WHITESPACE_AT_EOL
 278                                  && !at_eol) {
 279                                 while (ptr2 != ptr + 1) {
 280                                         ha += (ha << 5);
 281                                         ha ^= (unsigned long) *ptr2;
 282                                         ptr2++;
 283                                 }
 284                         }
 285                         continue;
 286                 }
 287                 ha += (ha << 5);
 288                 ha ^= (unsigned long) *ptr;
 289         }
 290         *data = ptr < top ? ptr + 1: ptr;
 291
 292         return ha;
 293 }
 294
 295
 296 unsigned long xdl_hash_record(char const **data, char const *top, long flags) {
 297         unsigned long ha = 5381;
 298         char const *ptr = *data;
 299
 300         if (flags & XDF_WHITESPACE_FLAGS)
 301                 return xdl_hash_record_with_whitespace(data, top, flags);
 302
 303         for (; ptr < top && *ptr != '\n'; ptr++) {
 304                 ha += (ha << 5);
 305                 ha ^= (unsigned long) *ptr;
 306         }
 307         *data = ptr < top ? ptr + 1: ptr;
 308
 309         return ha;
 310 }
 311
 312
 313 unsigned int xdl_hashbits(unsigned int size) {
 314         unsigned int val = 1, bits = 0;
 315
 316         for (; val < size && bits < CHAR_BIT * sizeof(unsigned int); val <<= 1, bits++);
 317         return bits ? bits: 1;
 318 }
 319
 320
 321 int xdl_num_out(char *out, long val) {
 322         char *ptr, *str = out;
 323         char buf[32];
 324
 325         ptr = buf + sizeof(buf) - 1;
 326         *ptr = '\0';
 327         if (val < 0) {
 328                 *--ptr = '-';
 329                 val = -val;
 330         }
 331         for (; val && ptr > buf; val /= 10)
 332                 *--ptr = "0123456789"[val % 10];
 333         if (*ptr)
 334                 for (; *ptr; ptr++, str++)
 335                         *str = *ptr;
 336         else
 337                 *str++ = '0';
 338         *str = '\0';
 339
 340         return str - out;
 341 }
 342
 343
 344 long xdl_atol(char const *str, char const **next) {
 345         long val, base;
 346         char const *top;
 347
 348         for (top = str; XDL_ISDIGIT(*top); top++);
 349         if (next)
 350                 *next = top;
 351         for (val = 0, base = 1, top--; top >= str; top--, base *= 10)
 352                 val += base * (long)(*top - '0');
 353         return val;
 354 }
 355
 356
 357 int xdl_emit_hunk_hdr(long s1, long c1, long s2, long c2,
 358                       const char *func, long funclen, xdemitcb_t *ecb) {
 359         int nb = 0;
 360         mmbuffer_t mb;
 361         char buf[128];
 362
 363         memcpy(buf, "@@ -", 4);
 364         nb += 4;
 365
 366         nb += xdl_num_out(buf + nb, c1 ? s1: s1 - 1);
 367
 368         if (c1 != 1) {
 369                 memcpy(buf + nb, ",", 1);
 370                 nb += 1;
 371
 372                 nb += xdl_num_out(buf + nb, c1);
 373         }
 374
 375         memcpy(buf + nb, " +", 2);
 376         nb += 2;
 377
 378         nb += xdl_num_out(buf + nb, c2 ? s2: s2 - 1);
 379
 380         if (c2 != 1) {
 381                 memcpy(buf + nb, ",", 1);
 382                 nb += 1;
 383
 384                 nb += xdl_num_out(buf + nb, c2);
 385         }
 386
 387         memcpy(buf + nb, " @@", 3);
 388         nb += 3;
 389         if (func && funclen) {
 390                 buf[nb++] = ' ';
 391                 if (funclen > sizeof(buf) - nb - 1)
 392                         funclen = sizeof(buf) - nb - 1;
 393                 memcpy(buf + nb, func, funclen);
 394                 nb += funclen;
 395         }
 396         buf[nb++] = '\n';
 397
 398         mb.ptr = buf;
 399         mb.size = nb;
 400         if (ecb->outf(ecb->priv, &mb, 1) < 0)
 401                 return -1;
 402
 403         return 0;
 404 }
 405
 406 int xdl_fall_back_diff(xdfenv_t *diff_env, xpparam_t const *xpp,
 407                 int line1, int count1, int line2, int count2)
 408 {
 409         /*
 410          * This probably does not work outside Git, since
 411          * we have a very simple mmfile structure.
 412          *
 413          * Note: ideally, we would reuse the prepared environment, but
 414          * the libxdiff interface does not (yet) allow for diffing only
 415          * ranges of lines instead of the whole files.
 416          */
 417         mmfile_t subfile1, subfile2;
 418         xdfenv_t env;
 419
 420         subfile1.ptr = (char *)diff_env->xdf1.recs[line1 - 1]->ptr;
 421         subfile1.size = diff_env->xdf1.recs[line1 + count1 - 2]->ptr +
 422                 diff_env->xdf1.recs[line1 + count1 - 2]->size - subfile1.ptr;
 423         subfile2.ptr = (char *)diff_env->xdf2.recs[line2 - 1]->ptr;
 424         subfile2.size = diff_env->xdf2.recs[line2 + count2 - 2]->ptr +
 425                 diff_env->xdf2.recs[line2 + count2 - 2]->size - subfile2.ptr;
 426         if (xdl_do_diff(&subfile1, &subfile2, xpp, &env) < 0)
 427                 return -1;
 428
 429         memcpy(diff_env->xdf1.rchg + line1 - 1, env.xdf1.rchg, count1);
 430         memcpy(diff_env->xdf2.rchg + line2 - 1, env.xdf2.rchg, count2);
 431
 432         xdl_free_env(&env);
 433
 434         return 0;
 435 }