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 */ 41 42 /* 43 * Ref: Usenix ;login:, Vol 8, No 1, February/March, 1983, p. 8. 44 * 45 * Locate scans a file list for the full pathname of a file given only part 46 * of the name. The list has been processed with with "front-compression" 47 * and bigram coding. Front compression reduces space by a factor of 4-5, 48 * bigram coding by a further 20-25%. 49 * 50 * The codes are: 51 * 52 * 0-28 likeliest differential counts + offset to make nonnegative 53 * 30 switch code for out-of-range count to follow in next word 54 * 31 an 8 bit char followed 55 * 128-255 bigram codes (128 most common, as determined by 'updatedb') 56 * 32-127 single character (printable) ascii residue (ie, literal) 57 * 58 * A novel two-tiered string search technique is employed: 59 * 60 * First, a metacharacter-free subpattern and partial pathname is matched 61 * BACKWARDS to avoid full expansion of the pathname list. The time savings 62 * is 40-50% over forward matching, which cannot efficiently handle 63 * overlapped search patterns and compressed path residue. 64 * 65 * Then, the actual shell glob-style regular expression (if in this form) is 66 * matched against the candidate pathnames using the slower routines provided 67 * in the standard 'find'. 68 */ 69 70 #include <sys/param.h> 71 #include <ctype.h> 72 #include <err.h> 73 #include <fnmatch.h> 74 #include <locale.h> 75 #include <stdio.h> 76 #include <stdlib.h> 77 #include <string.h> 78 #include <unistd.h> 79 80 #ifdef MMAP 81 # include <sys/types.h> 82 # include <sys/stat.h> 83 # include <sys/mman.h> 84 # include <fcntl.h> 85 #endif 86 87 88 #ifdef sun 89 #include <netinet/in.h> /* SunOS byteorder(3) htohl(3) */ 90 #endif 91 92 #include "locate.h" 93 #include "pathnames.h" 94 95 #ifdef DEBUG 96 # include <sys/time.h> 97 # include <sys/types.h> 98 # include <sys/resource.h> 99 #endif 100 101 char *path_fcodes; /* locate database */ 102 int f_mmap; /* use mmap */ 103 int f_icase; /* ignore case */ 104 int f_stdin; /* read database from stdin */ 105 int f_statistic; /* print statistic */ 106 int f_silent; /* suppress output, show only count of matches */ 107 int f_limit; /* limit number of output lines, 0 == infinite */ 108 u_int counter; /* counter for matches [-c] */ 109 110 111 void usage(void); 112 void statistic(FILE *, char *); 113 void fastfind(FILE *, char *, char *); 114 void fastfind_icase(FILE *, char *, char *); 115 void fastfind_mmap(char *, caddr_t, int, char *); 116 void fastfind_mmap_icase(char *, caddr_t, int, char *); 117 void search_mmap(char *, char **); 118 void search_fopen(char *, char **); 119 unsigned long cputime(void); 120 121 extern char **colon(char **, char*, char*); 122 extern void print_matches(u_int); 123 extern int getwm(caddr_t); 124 extern int getwf(FILE *); 125 extern u_char *tolower_word(u_char *); 126 extern int check_bigram_char(int); 127 extern char *patprep(char *); 128 129 int 130 main(int argc, char **argv) 131 { 132 register int ch; 133 char **dbv = NULL; 134 #ifdef MMAP 135 f_mmap = 1; /* mmap is default */ 136 #endif 137 (void) setlocale(LC_ALL, ""); 138 139 while ((ch = getopt(argc, argv, "Scd:il:ms")) != -1) 140 switch(ch) { 141 case 'S': /* statistic lines */ 142 f_statistic = 1; 143 break; 144 case 'l': /* limit number of output lines, 0 == infinite */ 145 f_limit = atoi(optarg); 146 break; 147 case 'd': /* database */ 148 dbv = colon(dbv, optarg, _PATH_FCODES); 149 break; 150 case 'i': /* ignore case */ 151 f_icase = 1; 152 break; 153 case 'm': /* mmap */ 154 #ifdef MMAP 155 f_mmap = 1; 156 #else 157 warnx("mmap(2) not implemented"); 158 #endif 159 break; 160 case 's': /* stdio lib */ 161 f_mmap = 0; 162 break; 163 case 'c': /* suppress output, show only count of matches */ 164 f_silent = 1; 165 break; 166 default: 167 usage(); 168 } 169 argv += optind; 170 argc -= optind; 171 172 /* to few arguments */ 173 if (argc < 1 && !(f_statistic)) 174 usage(); 175 176 /* no (valid) database as argument */ 177 if (dbv == NULL || *dbv == NULL) { 178 /* try to read database from enviroment */ 179 if ((path_fcodes = getenv("LOCATE_PATH")) == NULL || 180 *path_fcodes == '\0') 181 /* use default database */ 182 dbv = colon(dbv, _PATH_FCODES, _PATH_FCODES); 183 else /* $LOCATE_PATH */ 184 dbv = colon(dbv, path_fcodes, _PATH_FCODES); 185 } 186 187 if (f_icase && UCHAR_MAX < 4096) /* init tolower lookup table */ 188 for (ch = 0; ch < UCHAR_MAX + 1; ch++) 189 myctype[ch] = tolower(ch); 190 191 /* foreach database ... */ 192 while((path_fcodes = *dbv) != NULL) { 193 dbv++; 194 195 if (!strcmp(path_fcodes, "-")) 196 f_stdin = 1; 197 else 198 f_stdin = 0; 199 200 #ifndef MMAP 201 f_mmap = 0; /* be paranoid */ 202 #endif 203 if (!f_mmap || f_stdin || f_statistic) 204 search_fopen(path_fcodes, argv); 205 else 206 search_mmap(path_fcodes, argv); 207 } 208 209 if (f_silent) 210 print_matches(counter); 211 exit(0); 212 } 213 214 215 void 216 search_fopen(char *db, char **s) 217 { 218 FILE *fp; 219 #ifdef DEBUG 220 long t0; 221 #endif 222 223 /* can only read stdin once */ 224 if (f_stdin) { 225 fp = stdin; 226 if (*(s+1) != NULL) { 227 warnx("read database from stdin, use only `%s' as pattern", *s); 228 *(s+1) = NULL; 229 } 230 } 231 else if ((fp = fopen(path_fcodes, "r")) == NULL) 232 err(1, "`%s'", path_fcodes); 233 234 /* count only chars or lines */ 235 if (f_statistic) { 236 statistic(fp, path_fcodes); 237 (void)fclose(fp); 238 return; 239 } 240 241 /* foreach search string ... */ 242 while(*s != NULL) { 243 #ifdef DEBUG 244 t0 = cputime(); 245 #endif 246 if (!f_stdin && 247 fseek(fp, (long)0, SEEK_SET) == -1) 248 err(1, "fseek to begin of ``%s''\n", path_fcodes); 249 250 if (f_icase) 251 fastfind_icase(fp, *s, path_fcodes); 252 else 253 fastfind(fp, *s, path_fcodes); 254 #ifdef DEBUG 255 warnx("fastfind %ld ms", cputime () - t0); 256 #endif 257 s++; 258 } 259 (void)fclose(fp); 260 } 261 262 #ifdef MMAP 263 void 264 search_mmap(char *db, char **s) 265 { 266 struct stat sb; 267 int fd; 268 caddr_t p; 269 off_t len; 270 #ifdef DEBUG 271 long t0; 272 #endif 273 if ((fd = open(path_fcodes, O_RDONLY)) == -1 || 274 fstat(fd, &sb) == -1) 275 err(1, "`%s'", path_fcodes); 276 len = sb.st_size; 277 278 if ((p = mmap((caddr_t)0, (size_t)len, 279 PROT_READ, MAP_SHARED, 280 fd, (off_t)0)) == MAP_FAILED) 281 err(1, "mmap ``%s''", path_fcodes); 282 283 /* foreach search string ... */ 284 while (*s != NULL) { 285 #ifdef DEBUG 286 t0 = cputime(); 287 #endif 288 if (f_icase) 289 fastfind_mmap_icase(*s, p, (int)len, path_fcodes); 290 else 291 fastfind_mmap(*s, p, (int)len, path_fcodes); 292 #ifdef DEBUG 293 warnx("fastfind %ld ms", cputime () - t0); 294 #endif 295 s++; 296 } 297 298 if (munmap(p, (size_t)len) == -1) 299 warn("munmap %s\n", path_fcodes); 300 301 (void)close(fd); 302 } 303 #endif /* MMAP */ 304 305 #ifdef DEBUG 306 unsigned long 307 cputime (void) 308 { 309 struct rusage rus; 310 311 getrusage(0, &rus); 312 return(rus.ru_utime.tv_sec * 1000 + rus.ru_utime.tv_usec / 1000); 313 } 314 #endif /* DEBUG */ 315 316 void 317 usage (void) 318 { 319 (void)fprintf(stderr, 320 "usage: locate [-Scims] [-l limit] [-d database] pattern ...\n\n"); 321 (void)fprintf(stderr, 322 "default database: `%s' or $LOCATE_PATH\n", _PATH_FCODES); 323 exit(1); 324 } 325 326 327 /* load fastfind functions */ 328 329 /* statistic */ 330 /* fastfind_mmap, fastfind_mmap_icase */ 331 #ifdef MMAP 332 #undef FF_MMAP 333 #undef FF_ICASE 334 335 #define FF_MMAP 336 #include "fastfind.c" 337 #define FF_ICASE 338 #include "fastfind.c" 339 #endif /* MMAP */ 340 341 /* fopen */ 342 /* fastfind, fastfind_icase */ 343 #undef FF_MMAP 344 #undef FF_ICASE 345 #include "fastfind.c" 346 #define FF_ICASE 347 #include "fastfind.c" 348