1 /* $OpenBSD: util.c,v 1.39 2010/07/02 22:18:03 tedu Exp $ */ 2 3 /*- 4 * Copyright (c) 1999 James Howard and Dag-Erling Co�dan Sm�rgrav 5 * Copyright (C) 2008-2010 Gabor Kovesdan <gabor@FreeBSD.org> 6 * All rights reserved. 7 * 8 * Redistribution and use in source and binary forms, with or without 9 * modification, are permitted provided that the following conditions 10 * are met: 11 * 1. Redistributions of source code must retain the above copyright 12 * notice, this list of conditions and the following disclaimer. 13 * 2. Redistributions in binary form must reproduce the above copyright 14 * notice, this list of conditions and the following disclaimer in the 15 * documentation and/or other materials provided with the distribution. 16 * 17 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND 18 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 19 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 20 * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE 21 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 22 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 23 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 24 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 25 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 26 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 27 * SUCH DAMAGE. 28 */ 29 30 #include <sys/cdefs.h> 31 __FBSDID("$FreeBSD$"); 32 33 #include <sys/stat.h> 34 #include <sys/types.h> 35 36 #include <ctype.h> 37 #include <err.h> 38 #include <errno.h> 39 #include <fnmatch.h> 40 #include <fts.h> 41 #include <libgen.h> 42 #include <stdio.h> 43 #include <stdlib.h> 44 #include <string.h> 45 #include <unistd.h> 46 #include <wchar.h> 47 #include <wctype.h> 48 49 #include "grep.h" 50 51 static int linesqueued; 52 static int procline(struct str *l, int); 53 54 /* 55 * Processes a directory when a recursive search is performed with 56 * the -R option. Each appropriate file is passed to procfile(). 57 */ 58 int 59 grep_tree(char **argv) 60 { 61 FTS *fts; 62 FTSENT *p; 63 char *d, *dir = NULL; 64 unsigned int i; 65 int c, fts_flags; 66 bool ok; 67 68 c = fts_flags = 0; 69 70 switch(linkbehave) { 71 case LINK_EXPLICIT: 72 fts_flags = FTS_COMFOLLOW; 73 break; 74 case LINK_SKIP: 75 fts_flags = FTS_PHYSICAL; 76 break; 77 default: 78 fts_flags = FTS_LOGICAL; 79 80 } 81 82 fts_flags |= FTS_NOSTAT | FTS_NOCHDIR; 83 84 if (!(fts = fts_open(argv, fts_flags, NULL))) 85 err(2, "fts_open"); 86 while ((p = fts_read(fts)) != NULL) { 87 switch (p->fts_info) { 88 case FTS_DNR: 89 /* FALLTHROUGH */ 90 case FTS_ERR: 91 errx(2, "%s: %s", p->fts_path, strerror(p->fts_errno)); 92 break; 93 case FTS_D: 94 /* FALLTHROUGH */ 95 case FTS_DP: 96 break; 97 case FTS_DC: 98 /* Print a warning for recursive directory loop */ 99 warnx("warning: %s: recursive directory loop", 100 p->fts_path); 101 break; 102 default: 103 /* Check for file exclusion/inclusion */ 104 ok = true; 105 if (exclflag) { 106 if ((d = strrchr(p->fts_path, '/')) != NULL) { 107 dir = grep_malloc(sizeof(char) * 108 (d - p->fts_path + 2)); 109 strlcpy(dir, p->fts_path, 110 (d - p->fts_path + 1)); 111 } 112 for (i = 0; i < epatterns; ++i) { 113 switch(epattern[i].type) { 114 case FILE_PAT: 115 if (fnmatch(epattern[i].pat, 116 basename(p->fts_path), 0) == 0) 117 ok = epattern[i].mode != EXCL_PAT; 118 break; 119 case DIR_PAT: 120 if (dir != NULL && strstr(dir, 121 epattern[i].pat) != NULL) 122 ok = epattern[i].mode != EXCL_PAT; 123 break; 124 } 125 } 126 free(dir); 127 dir = NULL; 128 } 129 130 if (ok) 131 c += procfile(p->fts_path); 132 break; 133 } 134 } 135 136 fts_close(fts); 137 return (c); 138 } 139 140 /* 141 * Opens a file and processes it. Each file is processed line-by-line 142 * passing the lines to procline(). 143 */ 144 int 145 procfile(const char *fn) 146 { 147 struct file *f; 148 struct stat sb; 149 struct str ln; 150 mode_t s; 151 int c, t; 152 153 if (mflag && (mcount <= 0)) 154 return (0); 155 156 if (strcmp(fn, "-") == 0) { 157 fn = label != NULL ? label : getstr(1); 158 f = grep_stdin_open(); 159 } else { 160 if (!stat(fn, &sb)) { 161 /* Check if we need to process the file */ 162 s = sb.st_mode & S_IFMT; 163 if (s == S_IFDIR && dirbehave == DIR_SKIP) 164 return (0); 165 if ((s == S_IFIFO || s == S_IFCHR || s == S_IFBLK 166 || s == S_IFSOCK) && devbehave == DEV_SKIP) 167 return (0); 168 } 169 f = grep_open(fn); 170 } 171 if (f == NULL) { 172 if (!sflag) 173 warn("%s", fn); 174 if (errno == ENOENT) 175 notfound = true; 176 return (0); 177 } 178 179 ln.file = grep_malloc(strlen(fn) + 1); 180 strcpy(ln.file, fn); 181 ln.line_no = 0; 182 ln.len = 0; 183 linesqueued = 0; 184 tail = 0; 185 ln.off = -1; 186 187 for (c = 0; c == 0 || !(lflag || qflag); ) { 188 ln.off += ln.len + 1; 189 if ((ln.dat = grep_fgetln(f, &ln.len)) == NULL) { 190 if (ln.line_no == 0 && matchall) 191 exit(0); 192 else 193 break; 194 } 195 if (ln.len > 0 && ln.dat[ln.len - 1] == '\n') 196 --ln.len; 197 ln.line_no++; 198 199 /* Return if we need to skip a binary file */ 200 if (f->binary && binbehave == BINFILE_SKIP) { 201 grep_close(f); 202 free(ln.file); 203 free(f); 204 return (0); 205 } 206 /* Process the file line-by-line */ 207 if ((t = procline(&ln, f->binary)) == 0 && Bflag > 0) { 208 enqueue(&ln); 209 linesqueued++; 210 } 211 c += t; 212 213 /* Count the matches if we have a match limit */ 214 if (mflag) { 215 mcount -= t; 216 if (mcount <= 0) 217 break; 218 } 219 } 220 if (Bflag > 0) 221 clearqueue(); 222 grep_close(f); 223 224 if (cflag) { 225 if (!hflag) 226 printf("%s:", ln.file); 227 printf("%u\n", c); 228 } 229 if (lflag && !qflag && c != 0) 230 printf("%s\n", fn); 231 if (Lflag && !qflag && c == 0) 232 printf("%s\n", fn); 233 if (c && !cflag && !lflag && !Lflag && 234 binbehave == BINFILE_BIN && f->binary && !qflag) 235 printf(getstr(9), fn); 236 237 free(ln.file); 238 free(f); 239 return (c); 240 } 241 242 #define iswword(x) (iswalnum((x)) || (x) == L'_') 243 244 /* 245 * Processes a line comparing it with the specified patterns. Each pattern 246 * is looped to be compared along with the full string, saving each and every 247 * match, which is necessary to colorize the output and to count the 248 * matches. The matching lines are passed to printline() to display the 249 * appropriate output. 250 */ 251 static int 252 procline(struct str *l, int nottext) 253 { 254 regmatch_t matches[MAX_LINE_MATCHES]; 255 regmatch_t pmatch; 256 size_t st = 0; 257 unsigned int i; 258 int c = 0, m = 0, r = 0; 259 260 if (!matchall) { 261 /* Loop to process the whole line */ 262 while (st <= l->len) { 263 pmatch.rm_so = st; 264 pmatch.rm_eo = l->len; 265 266 /* Loop to compare with all the patterns */ 267 for (i = 0; i < patterns; i++) { 268 /* 269 * XXX: grep_search() is a workaround for speed up and should be 270 * removed in the future. See fastgrep.c. 271 */ 272 if (fg_pattern[i].pattern) { 273 r = grep_search(&fg_pattern[i], 274 (unsigned char *)l->dat, 275 l->len, &pmatch); 276 r = (r == 0) ? 0 : REG_NOMATCH; 277 st = pmatch.rm_eo; 278 } else { 279 r = regexec(&r_pattern[i], l->dat, 1, 280 &pmatch, eflags); 281 r = (r == 0) ? 0 : REG_NOMATCH; 282 st = pmatch.rm_eo; 283 } 284 if (r == REG_NOMATCH) 285 continue; 286 /* Check for full match */ 287 if (r == 0 && xflag) 288 if (pmatch.rm_so != 0 || 289 (size_t)pmatch.rm_eo != l->len) 290 r = REG_NOMATCH; 291 /* Check for whole word match */ 292 if (r == 0 && wflag && pmatch.rm_so != 0 && 293 (size_t)pmatch.rm_eo != l->len) { 294 wchar_t *wbegin; 295 wint_t wend; 296 size_t size; 297 298 size = mbstowcs(NULL, l->dat, 299 pmatch.rm_so); 300 301 if (size == ((size_t) - 1)) 302 r = REG_NOMATCH; 303 else { 304 wbegin = grep_malloc(size); 305 if (mbstowcs(wbegin, l->dat, 306 pmatch.rm_so) == ((size_t) - 1)) 307 r = REG_NOMATCH; 308 else if (sscanf(&l->dat[pmatch.rm_eo], 309 "%lc", &wend) != 1) 310 r = REG_NOMATCH; 311 else if (iswword(wbegin[wcslen(wbegin)]) || 312 iswword(wend)) 313 r = REG_NOMATCH; 314 free(wbegin); 315 } 316 } 317 if (r == 0) { 318 if (m == 0) 319 c++; 320 if (m < MAX_LINE_MATCHES) 321 matches[m++] = pmatch; 322 /* matches - skip further patterns */ 323 if ((color != NULL && !oflag) || qflag || lflag) 324 break; 325 } 326 } 327 328 if (vflag) { 329 c = !c; 330 break; 331 } 332 /* One pass if we are not recording matches */ 333 if ((color != NULL && !oflag) || qflag || lflag) 334 break; 335 336 if (st == (size_t)pmatch.rm_so) 337 break; /* No matches */ 338 } 339 } else 340 c = !vflag; 341 342 if (c && binbehave == BINFILE_BIN && nottext) 343 return (c); /* Binary file */ 344 345 /* Dealing with the context */ 346 if ((tail || c) && !cflag && !qflag && !lflag && !Lflag) { 347 if (c) { 348 if (!first && !prev && !tail && Aflag) 349 printf("--\n"); 350 tail = Aflag; 351 if (Bflag > 0) { 352 if (!first && !prev) 353 printf("--\n"); 354 printqueue(); 355 } 356 linesqueued = 0; 357 printline(l, ':', matches, m); 358 } else { 359 printline(l, '-', matches, m); 360 tail--; 361 } 362 } 363 364 if (c) { 365 prev = true; 366 first = false; 367 } else 368 prev = false; 369 370 return (c); 371 } 372 373 /* 374 * Safe malloc() for internal use. 375 */ 376 void * 377 grep_malloc(size_t size) 378 { 379 void *ptr; 380 381 if ((ptr = malloc(size)) == NULL) 382 err(2, "malloc"); 383 return (ptr); 384 } 385 386 /* 387 * Safe calloc() for internal use. 388 */ 389 void * 390 grep_calloc(size_t nmemb, size_t size) 391 { 392 void *ptr; 393 394 if ((ptr = calloc(nmemb, size)) == NULL) 395 err(2, "calloc"); 396 return (ptr); 397 } 398 399 /* 400 * Safe realloc() for internal use. 401 */ 402 void * 403 grep_realloc(void *ptr, size_t size) 404 { 405 406 if ((ptr = realloc(ptr, size)) == NULL) 407 err(2, "realloc"); 408 return (ptr); 409 } 410 411 /* 412 * Prints a matching line according to the command line options. 413 */ 414 void 415 printline(struct str *line, int sep, regmatch_t *matches, int m) 416 { 417 size_t a = 0; 418 int i, n = 0; 419 420 if (!hflag) { 421 if (nullflag == 0) 422 fputs(line->file, stdout); 423 else { 424 printf("%s", line->file); 425 putchar(0); 426 } 427 ++n; 428 } 429 if (nflag) { 430 if (n > 0) 431 putchar(sep); 432 printf("%d", line->line_no); 433 ++n; 434 } 435 if (bflag) { 436 if (n > 0) 437 putchar(sep); 438 printf("%lld", (long long)line->off); 439 ++n; 440 } 441 if (n) 442 putchar(sep); 443 /* --color and -o */ 444 if ((oflag || color) && m > 0) { 445 for (i = 0; i < m; i++) { 446 if (!oflag) 447 fwrite(line->dat + a, matches[i].rm_so - a, 1, 448 stdout); 449 if (color) 450 fprintf(stdout, "\33[%sm\33[K", color); 451 452 fwrite(line->dat + matches[i].rm_so, 453 matches[i].rm_eo - matches[i].rm_so, 1, 454 stdout); 455 if (color) 456 fprintf(stdout, "\33[m\33[K"); 457 a = matches[i].rm_eo; 458 if (oflag) 459 putchar('\n'); 460 } 461 if (!oflag) { 462 if (line->len - a > 0) 463 fwrite(line->dat + a, line->len - a, 1, stdout); 464 putchar('\n'); 465 } 466 } else { 467 fwrite(line->dat, line->len, 1, stdout); 468 putchar('\n'); 469 } 470 } 471