1 /* 2 * Copyright (c) 1995 Wolfram Schneider <wosch@FreeBSD.org>. Berlin. 3 * Copyright (c) 1989, 1993 4 * The Regents of the University of California. All rights reserved. 5 * 6 * This code is derived from software contributed to Berkeley by 7 * James A. Woods. 8 * 9 * Redistribution and use in source and binary forms, with or without 10 * modification, are permitted provided that the following conditions 11 * are met: 12 * 1. Redistributions of source code must retain the above copyright 13 * notice, this list of conditions and the following disclaimer. 14 * 2. Redistributions in binary form must reproduce the above copyright 15 * notice, this list of conditions and the following disclaimer in the 16 * documentation and/or other materials provided with the distribution. 17 * 3. All advertising materials mentioning features or use of this software 18 * must display the following acknowledgement: 19 * This product includes software developed by the University of 20 * California, Berkeley and its contributors. 21 * 4. Neither the name of the University nor the names of its contributors 22 * may be used to endorse or promote products derived from this software 23 * without specific prior written permission. 24 * 25 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 26 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 27 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 28 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 29 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 30 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 31 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 32 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 33 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 34 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 35 * SUCH DAMAGE. 36 * 37 * @(#) Copyright (c) 1995-1996 Wolfram Schneider, Berlin. @(#) Copyright (c) 1989, 1993 The Regents of the University of California. All rights reserved. 38 * @(#)locate.c 8.1 (Berkeley) 6/6/93 39 * $FreeBSD: src/usr.bin/locate/locate/locate.c,v 1.12.2.1 2001/03/04 08:47:25 kris Exp $ 40 * $DragonFly: src/usr.bin/locate/locate/locate.c,v 1.4 2003/11/06 19:30:04 eirikn Exp $ 41 */ 42 43 /* 44 * Ref: Usenix ;login:, Vol 8, No 1, February/March, 1983, p. 8. 45 * 46 * Locate scans a file list for the full pathname of a file given only part 47 * of the name. The list has been processed with with "front-compression" 48 * and bigram coding. Front compression reduces space by a factor of 4-5, 49 * bigram coding by a further 20-25%. 50 * 51 * The codes are: 52 * 53 * 0-28 likeliest differential counts + offset to make nonnegative 54 * 30 switch code for out-of-range count to follow in next word 55 * 31 an 8 bit char followed 56 * 128-255 bigram codes (128 most common, as determined by 'updatedb') 57 * 32-127 single character (printable) ascii residue (ie, literal) 58 * 59 * A novel two-tiered string search technique is employed: 60 * 61 * First, a metacharacter-free subpattern and partial pathname is matched 62 * BACKWARDS to avoid full expansion of the pathname list. The time savings 63 * is 40-50% over forward matching, which cannot efficiently handle 64 * overlapped search patterns and compressed path residue. 65 * 66 * Then, the actual shell glob-style regular expression (if in this form) is 67 * matched against the candidate pathnames using the slower routines provided 68 * in the standard 'find'. 69 */ 70 71 #include <sys/param.h> 72 #include <ctype.h> 73 #include <err.h> 74 #include <fnmatch.h> 75 #include <locale.h> 76 #include <stdio.h> 77 #include <stdlib.h> 78 #include <string.h> 79 #include <unistd.h> 80 81 #ifdef MMAP 82 # include <sys/types.h> 83 # include <sys/stat.h> 84 # include <sys/mman.h> 85 # include <fcntl.h> 86 #endif 87 88 89 #ifdef sun 90 #include <netinet/in.h> /* SunOS byteorder(3) htohl(3) */ 91 #endif 92 93 #include "locate.h" 94 #include "pathnames.h" 95 96 #ifdef DEBUG 97 # include <sys/time.h> 98 # include <sys/types.h> 99 # include <sys/resource.h> 100 #endif 101 102 char *path_fcodes; /* locate database */ 103 int f_mmap; /* use mmap */ 104 int f_icase; /* ignore case */ 105 int f_stdin; /* read database from stdin */ 106 int f_statistic; /* print statistic */ 107 int f_silent; /* suppress output, show only count of matches */ 108 int f_limit; /* limit number of output lines, 0 == infinite */ 109 u_int counter; /* counter for matches [-c] */ 110 111 112 void usage(void); 113 void statistic(FILE *, char *); 114 void fastfind(FILE *, char *, char *); 115 void fastfind_icase(FILE *, char *, char *); 116 void fastfind_mmap(char *, caddr_t, int, char *); 117 void fastfind_mmap_icase(char *, caddr_t, int, char *); 118 void search_mmap(char *, char **); 119 void search_fopen(char *, char **); 120 unsigned long cputime(void); 121 122 extern char **colon(char **, char*, char*); 123 extern void print_matches(u_int); 124 extern int getwm(caddr_t); 125 extern int getwf(FILE *); 126 extern u_char *tolower_word(u_char *); 127 extern int check_bigram_char(int); 128 extern char *patprep(char *); 129 130 int 131 main(argc, argv) 132 int argc; 133 char **argv; 134 { 135 register int ch; 136 char **dbv = NULL; 137 #ifdef MMAP 138 f_mmap = 1; /* mmap is default */ 139 #endif 140 (void) setlocale(LC_ALL, ""); 141 142 while ((ch = getopt(argc, argv, "Scd:il:ms")) != -1) 143 switch(ch) { 144 case 'S': /* statistic lines */ 145 f_statistic = 1; 146 break; 147 case 'l': /* limit number of output lines, 0 == infinite */ 148 f_limit = atoi(optarg); 149 break; 150 case 'd': /* database */ 151 dbv = colon(dbv, optarg, _PATH_FCODES); 152 break; 153 case 'i': /* ignore case */ 154 f_icase = 1; 155 break; 156 case 'm': /* mmap */ 157 #ifdef MMAP 158 f_mmap = 1; 159 #else 160 warnx("mmap(2) not implemented"); 161 #endif 162 break; 163 case 's': /* stdio lib */ 164 f_mmap = 0; 165 break; 166 case 'c': /* suppress output, show only count of matches */ 167 f_silent = 1; 168 break; 169 default: 170 usage(); 171 } 172 argv += optind; 173 argc -= optind; 174 175 /* to few arguments */ 176 if (argc < 1 && !(f_statistic)) 177 usage(); 178 179 /* no (valid) database as argument */ 180 if (dbv == NULL || *dbv == NULL) { 181 /* try to read database from enviroment */ 182 if ((path_fcodes = getenv("LOCATE_PATH")) == NULL || 183 *path_fcodes == '\0') 184 /* use default database */ 185 dbv = colon(dbv, _PATH_FCODES, _PATH_FCODES); 186 else /* $LOCATE_PATH */ 187 dbv = colon(dbv, path_fcodes, _PATH_FCODES); 188 } 189 190 if (f_icase && UCHAR_MAX < 4096) /* init tolower lookup table */ 191 for (ch = 0; ch < UCHAR_MAX + 1; ch++) 192 myctype[ch] = tolower(ch); 193 194 /* foreach database ... */ 195 while((path_fcodes = *dbv) != NULL) { 196 dbv++; 197 198 if (!strcmp(path_fcodes, "-")) 199 f_stdin = 1; 200 else 201 f_stdin = 0; 202 203 #ifndef MMAP 204 f_mmap = 0; /* be paranoid */ 205 #endif 206 if (!f_mmap || f_stdin || f_statistic) 207 search_fopen(path_fcodes, argv); 208 else 209 search_mmap(path_fcodes, argv); 210 } 211 212 if (f_silent) 213 print_matches(counter); 214 exit(0); 215 } 216 217 218 void 219 search_fopen(db, s) 220 char *db; /* database */ 221 char **s; /* search strings */ 222 { 223 FILE *fp; 224 #ifdef DEBUG 225 long t0; 226 #endif 227 228 /* can only read stdin once */ 229 if (f_stdin) { 230 fp = stdin; 231 if (*(s+1) != NULL) { 232 warnx("read database from stdin, use only `%s' as pattern", *s); 233 *(s+1) = NULL; 234 } 235 } 236 else if ((fp = fopen(path_fcodes, "r")) == NULL) 237 err(1, "`%s'", path_fcodes); 238 239 /* count only chars or lines */ 240 if (f_statistic) { 241 statistic(fp, path_fcodes); 242 (void)fclose(fp); 243 return; 244 } 245 246 /* foreach search string ... */ 247 while(*s != NULL) { 248 #ifdef DEBUG 249 t0 = cputime(); 250 #endif 251 if (!f_stdin && 252 fseek(fp, (long)0, SEEK_SET) == -1) 253 err(1, "fseek to begin of ``%s''\n", path_fcodes); 254 255 if (f_icase) 256 fastfind_icase(fp, *s, path_fcodes); 257 else 258 fastfind(fp, *s, path_fcodes); 259 #ifdef DEBUG 260 warnx("fastfind %ld ms", cputime () - t0); 261 #endif 262 s++; 263 } 264 (void)fclose(fp); 265 } 266 267 #ifdef MMAP 268 void 269 search_mmap(db, s) 270 char *db; /* database */ 271 char **s; /* search strings */ 272 { 273 struct stat sb; 274 int fd; 275 caddr_t p; 276 off_t len; 277 #ifdef DEBUG 278 long t0; 279 #endif 280 if ((fd = open(path_fcodes, O_RDONLY)) == -1 || 281 fstat(fd, &sb) == -1) 282 err(1, "`%s'", path_fcodes); 283 len = sb.st_size; 284 285 if ((p = mmap((caddr_t)0, (size_t)len, 286 PROT_READ, MAP_SHARED, 287 fd, (off_t)0)) == MAP_FAILED) 288 err(1, "mmap ``%s''", path_fcodes); 289 290 /* foreach search string ... */ 291 while (*s != NULL) { 292 #ifdef DEBUG 293 t0 = cputime(); 294 #endif 295 if (f_icase) 296 fastfind_mmap_icase(*s, p, (int)len, path_fcodes); 297 else 298 fastfind_mmap(*s, p, (int)len, path_fcodes); 299 #ifdef DEBUG 300 warnx("fastfind %ld ms", cputime () - t0); 301 #endif 302 s++; 303 } 304 305 if (munmap(p, (size_t)len) == -1) 306 warn("munmap %s\n", path_fcodes); 307 308 (void)close(fd); 309 } 310 #endif /* MMAP */ 311 312 #ifdef DEBUG 313 unsigned long 314 cputime () 315 { 316 struct rusage rus; 317 318 getrusage(0, &rus); 319 return(rus.ru_utime.tv_sec * 1000 + rus.ru_utime.tv_usec / 1000); 320 } 321 #endif /* DEBUG */ 322 323 void 324 usage () 325 { 326 (void)fprintf(stderr, 327 "usage: locate [-Scims] [-l limit] [-d database] pattern ...\n\n"); 328 (void)fprintf(stderr, 329 "default database: `%s' or $LOCATE_PATH\n", _PATH_FCODES); 330 exit(1); 331 } 332 333 334 /* load fastfind functions */ 335 336 /* statistic */ 337 /* fastfind_mmap, fastfind_mmap_icase */ 338 #ifdef MMAP 339 #undef FF_MMAP 340 #undef FF_ICASE 341 342 #define FF_MMAP 343 #include "fastfind.c" 344 #define FF_ICASE 345 #include "fastfind.c" 346 #endif /* MMAP */ 347 348 /* fopen */ 349 /* fastfind, fastfind_icase */ 350 #undef FF_MMAP 351 #undef FF_ICASE 352 #include "fastfind.c" 353 #define FF_ICASE 354 #include "fastfind.c" 355