usr.bin/grep/util.c

   1 /*      $NetBSD: util.c,v 1.5 2007/12/06 20:33:49 cjep Exp $    */
   2
   3 /*-
   4  * Copyright (c) 1999 James Howard and Dag-Erling Coïdan Smørgrav
   5  * All rights reserved.
   6  *
   7  * Redistribution and use in source and binary forms, with or without
   8  * modification, are permitted provided that the following conditions
   9  * are met:
  10  * 1. Redistributions of source code must retain the above copyright
  11  *    notice, this list of conditions and the following disclaimer.
  12  * 2. Redistributions in binary form must reproduce the above copyright
  13  *    notice, this list of conditions and the following disclaimer in the
  14  *    documentation and/or other materials provided with the distribution.
  15  *
  16  * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
  17  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
  18  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
  19  * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
  20  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
  21  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
  22  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
  23  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
  24  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
  25  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
  26  * SUCH DAMAGE.
  27  *
  28  */
  29
  30 #include <sys/cdefs.h>
  31 #ifndef lint
  32 __RCSID("$NetBSD: util.c,v 1.5 2007/12/06 20:33:49 cjep Exp $");
  33 #endif /* not lint */
  34
  35 #include <sys/types.h>
  36 #include <sys/stat.h>
  37
  38 #include <ctype.h>
  39 #include <zlib.h>
  40 #include <err.h>
  41 #include <errno.h>
  42 #include <fts.h>
  43 #include <regex.h>
  44 #include <stdio.h>
  45 #include <stdlib.h>
  46 #include <string.h>
  47 #include <unistd.h>
  48
  49 #include "grep.h"
  50
  51 /*
  52  * Process a file line by line...
  53  */
  54
  55 static int linesqueued, newfile;
  56 static int procline(str_t *l, int nottext);
  57
  58 int
  59 grep_tree(char **argv)
  60 {
  61         FTS *fts;
  62         FTSENT *p;
  63         int c, fts_flags;
  64
  65         c = fts_flags = 0;
  66
  67 /*      if (linkbehave == LINK_EXPLICIT)
  68                 fts_flags = FTS_COMFOLLOW;
  69         if (linkbehave == LINK_SKIP)
  70                 fts_flags = FTS_PHYSICAL;
  71         if (linkbehave == LINK_FOLLOW)
  72                 fts_flags = FTS_LOGICAL;*/
  73
  74         fts_flags |= FTS_NOSTAT | FTS_NOCHDIR | FTS_LOGICAL;
  75
  76         if ((fts = fts_open(argv, fts_flags, NULL)) == NULL)
  77                 err(2, NULL);
  78         while ((p = fts_read(fts)) != NULL) {
  79                 switch (p->fts_info) {
  80                 case FTS_DNR:
  81                         break;
  82                 case FTS_ERR:
  83                         errx(2, "%s: %s", p->fts_path, strerror(p->fts_errno));
  84                         break;
  85                 case FTS_DP:
  86                 case FTS_D:
  87                         break;
  88                 case FTS_DC:
  89                         warnx("warning: %s: recursive directory loop",
  90                                 p->fts_path);
  91                         break;
  92                 default:
  93                         c += procfile(p->fts_path);
  94                         break;
  95                 }
  96         }
  97
  98         return c;
  99 }
 100
 101 int
 102 procfile(const char *fn)
 103 {
 104         str_t ln;
 105         file_t *f;
 106         struct stat sb;
 107         mode_t s;
 108         int c, t, z, nottext, skip;
 109
 110         tail = 0;
 111         newfile = 1;
 112
 113         if (fn == NULL) {
 114                 fn = stdin_label;
 115                 f = grep_fdopen(STDIN_FILENO, "r");
 116         } else {
 117                 skip = 1;
 118                 if (dirbehave == GREP_SKIP || devbehave == GREP_SKIP) {
 119                         if (stat(fn, &sb)) {
 120                                 fprintf(stderr, "Cannot stat %s %d\n",
 121                                         fn, errno);
 122                                 /* XXX record error variable */
 123                         } else {
 124                                 s = sb.st_mode & S_IFMT;
 125                                 if (s == S_IFDIR && dirbehave == GREP_SKIP)
 126                                         skip = 0;
 127                                 if (   (s == S_IFIFO || s == S_IFCHR ||
 128                                         s == S_IFBLK || s == S_IFSOCK)
 129                                         && devbehave == GREP_SKIP)
 130                                                         skip = 0;
 131                         }
 132                 }
 133                 if (skip == 0)
 134                         return 0;
 135
 136                 f = grep_open(fn, "r");
 137         }
 138         if (f == NULL) {
 139                 if (!sflag)
 140                         warn("%s", fn);
 141                 return 0;
 142         }
 143
 144         nottext = grep_bin_file(f);
 145
 146         if (nottext && binbehave == BIN_FILE_SKIP) {
 147                 /* Skip this file as it is binary */
 148                 grep_close(f);
 149                 return 0;
 150         }
 151
 152         ln.file = fn;
 153         ln.line_no = 0;
 154         linesqueued = 0;
 155         ln.off = -1;
 156
 157         if (Bflag > 0)
 158                 initqueue();
 159         for (c = 0; !(lflag && c);) {
 160                 ln.off += ln.len + 1;
 161                 if ((ln.dat = grep_fgetln(f, &ln.len)) == NULL)
 162                         break;
 163                 if (ln.len > 0 && ln.dat[ln.len - 1] == line_endchar)
 164                         --ln.len;
 165                 ln.line_no++;
 166
 167                 z = tail;
 168
 169                 if ((t = procline(&ln, nottext)) == 0 && Bflag > 0 && z == 0) {
 170                         enqueue(&ln);
 171                         linesqueued++;
 172                 }
 173                 c += t;
 174
 175                 /* If we have a maximum number of matches, stop processing */
 176                 if (mflag && c >= maxcount)
 177                         break;
 178         }
 179         if (Bflag > 0)
 180                 clearqueue();
 181         grep_close(f);
 182
 183         if (cflag) {
 184                 if (output_filenames)
 185                         printf("%s%c", ln.file, fn_colonchar);
 186                 printf("%u\n", c);
 187         }
 188
 189         if (lflag && c != 0)
 190                 printf("%s%c", fn, fn_endchar);
 191         if (Lflag && c == 0)
 192                 printf("%s%c", fn, fn_endchar);
 193         if (c && !cflag && !lflag && !Lflag &&
 194                 binbehave == BIN_FILE_BIN && nottext && !qflag)
 195                         printf("Binary file %s matches\n", fn);
 196
 197         return c;
 198 }
 199
 200
 201 /*
 202  * Process an individual line in a file. Return non-zero if it matches.
 203  */
 204
 205 #define isword(x) (isalnum((unsigned char)(x)) || (x) == '_')
 206
 207 static int
 208 procline(str_t *l, int nottext)
 209 {
 210         regmatch_t pmatch;
 211         regmatch_t matches[MAX_LINE_MATCHES];
 212         int c = 0, i, r, t, m = 0;
 213         regoff_t st = 0;
 214
 215         if (matchall) {
 216                 c = !vflag;
 217                 goto print;
 218         }
 219
 220         t = vflag ? REG_NOMATCH : 0;
 221
 222         while (st >= 0 && (size_t)st <= l->len) {
 223                 pmatch.rm_so = st;
 224                 pmatch.rm_eo = l->len;
 225                 for (i = 0; i < patterns; i++) {
 226                         r = regexec(&r_pattern[i], l->dat, 1, &pmatch, eflags);
 227                         if (r == REG_NOMATCH && t == 0)
 228                                 continue;
 229                         if (r == 0) {
 230                                 if (wflag) {
 231                                         if ((pmatch.rm_so != 0 && isword((unsigned char)l->dat[pmatch.rm_so - 1]))
 232                                             || ((size_t)pmatch.rm_eo != l->len && isword((unsigned char)l->dat[pmatch.rm_eo])))
 233                                                 r = REG_NOMATCH;
 234                                 }
 235                                 if (xflag) {
 236                                         if (pmatch.rm_so != 0 || (size_t)pmatch.rm_eo != l->len)
 237                                                 r = REG_NOMATCH;
 238                                 }
 239                         }
 240                         if (r == t) {
 241                                 if (m == 0)
 242                                         c++;
 243                                 if (m < MAX_LINE_MATCHES) {
 244                                         matches[m] = pmatch;
 245                                         m++;
 246                                 }
 247                                 st = pmatch.rm_eo;
 248                                 break;
 249                         }
 250                 }
 251
 252                 /* One pass if we are not recording matches */
 253                 if (!oflag && !colours)
 254                         break;
 255
 256                 if (st == pmatch.rm_so)
 257                         break;  /* No matches */
 258
 259         }
 260
 261 print:
 262
 263         if (c && binbehave == BIN_FILE_BIN && nottext)
 264                 return c;       /* Binary file */
 265
 266         if ((tail > 0 || c) && !cflag && !qflag) {
 267                 if (c) {
 268
 269                         if ( (Aflag || Bflag) && first > 0 &&
 270                            ( (Bflag <= linesqueued && tail == 0) || newfile) )
 271                                                 printf("--\n");
 272
 273                         first = 1;
 274                         newfile = 0;
 275                         tail = Aflag;
 276                         if (Bflag > 0)
 277                                 printqueue();
 278                         linesqueued = 0;
 279                         printline(l, fn_colonchar, matches, m);
 280                 } else {
 281                         printline(l, fn_dashchar, matches, m);
 282                         tail--;
 283                 }
 284
 285         }
 286         return c;
 287 }
 288
 289 void *
 290 grep_malloc(size_t size)
 291 {
 292         void *ptr;
 293
 294         if ((ptr = malloc(size)) == NULL)
 295                 err(2, "malloc");
 296         return ptr;
 297 }
 298
 299 void *
 300 grep_realloc(void *ptr, size_t size)
 301 {
 302         if ((ptr = realloc(ptr, size)) == NULL)
 303                 err(2, "realloc");
 304         return ptr;
 305 }
 306
 307 void
 308 printline(str_t *line, int sep, regmatch_t *matches, int m)
 309 {
 310         int i, n = 0;
 311         size_t a = 0;
 312
 313         if (output_filenames) {
 314                 fputs(line->file, stdout);
 315                 ++n;
 316         }
 317         if (nflag) {
 318                 if (n)
 319                         putchar(sep);
 320                 printf("%d", line->line_no);
 321                 ++n;
 322         }
 323         if (bflag) {
 324                 if (n)
 325                         putchar(sep);
 326                 printf("%lu", (unsigned long)line->off);
 327         }
 328         if (n)
 329                 putchar(sep);
 330
 331         if ((oflag || colours) && m > 0) {
 332
 333                 for (i = 0; i < m; i++) {
 334
 335                         if (!oflag)
 336                                 fwrite(line->dat + a, matches[i].rm_so - a, 1, stdout);
 337
 338                         if (colours)
 339                                 fprintf(stdout, "\33[%sm", grep_colour);
 340                         fwrite(line->dat + matches[i].rm_so,
 341                                 matches[i].rm_eo - matches[i].rm_so, 1, stdout);
 342
 343                         if (colours)
 344                                 fprintf(stdout, "\33[00m");
 345                         a = matches[i].rm_eo;
 346                         if (oflag)
 347                                 putchar('\n');
 348                 }
 349                 if (!oflag) {
 350                         if (line->len - a > 0)
 351                                 fwrite(line->dat + a, line->len - a, 1, stdout);
 352                         putchar('\n');
 353                 }
 354
 355
 356         } else {
 357                 fwrite(line->dat, line->len, 1, stdout);
 358                 putchar(line_endchar);
 359         }
 360
 361 }