usr.bin/locate/locate/locate.c

   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 static char *path_fcodes; /* locate database */
 102 static int f_mmap;      /* use mmap */
 103 static int f_icase;     /* ignore case */
 104 static int f_stdin;     /* read database from stdin */
 105 static int f_statistic; /* print statistic */
 106 static int f_silent;    /* suppress output, show only count of matches */
 107 static int f_limit;     /* limit number of output lines, 0 == infinite */
 108 static u_int counter;   /* counter for matches [-c] */
 109
 110
 111 static void    usage(void);
 112 static void    statistic(FILE *, char *);
 113 static void    fastfind(FILE *, char *, char *);
 114 static void    fastfind_icase(FILE *, char *, char *);
 115 static void    fastfind_mmap(char *, caddr_t, int, char *);
 116 static void    fastfind_mmap_icase(char *, caddr_t, int, char *);
 117 static void     search_mmap(char *, char **);
 118 static void     search_fopen(char *, char **);
 119 #ifdef DEBUG
 120 static unsigned long cputime(void);
 121 #endif
 122
 123 extern char     **colon(char **, char*, char*);
 124 extern void     print_matches(u_int);
 125 extern int      getwm(caddr_t);
 126 extern int      getwf(FILE *);
 127 extern u_char   *tolower_word(u_char *);
 128 extern int      check_bigram_char(int);
 129 extern char     *patprep(char *);
 130
 131 int
 132 main(int argc, char **argv)
 133 {
 134         register int ch;
 135         char **dbv = NULL;
 136 #ifdef MMAP
 137         f_mmap = 1;             /* mmap is default */
 138 #endif
 139         (void) setlocale(LC_ALL, "");
 140
 141         while ((ch = getopt(argc, argv, "Scd:il:ms")) != -1)
 142                 switch(ch) {
 143                 case 'S':       /* statistic lines */
 144                         f_statistic = 1;
 145                         break;
 146                 case 'l': /* limit number of output lines, 0 == infinite */
 147                         f_limit = atoi(optarg);
 148                         break;
 149                 case 'd':       /* database */
 150                         dbv = colon(dbv, optarg, _PATH_FCODES);
 151                         break;
 152                 case 'i':       /* ignore case */
 153                         f_icase = 1;
 154                         break;
 155                 case 'm':       /* mmap */
 156 #ifdef MMAP
 157                         f_mmap = 1;
 158 #else
 159                                                 warnx("mmap(2) not implemented");
 160 #endif
 161                         break;
 162                 case 's':       /* stdio lib */
 163                         f_mmap = 0;
 164                         break;
 165                 case 'c': /* suppress output, show only count of matches */
 166                         f_silent = 1;
 167                         break;
 168                 default:
 169                         usage();
 170                 }
 171         argv += optind;
 172         argc -= optind;
 173
 174         /* to few arguments */
 175         if (argc < 1 && !(f_statistic))
 176                 usage();
 177
 178         /* no (valid) database as argument */
 179         if (dbv == NULL || *dbv == NULL) {
 180                 /* try to read database from enviroment */
 181                 if ((path_fcodes = getenv("LOCATE_PATH")) == NULL ||
 182                      *path_fcodes == '\0')
 183                         /* use default database */
 184                         dbv = colon(dbv, _PATH_FCODES, _PATH_FCODES);
 185                 else            /* $LOCATE_PATH */
 186                         dbv = colon(dbv, path_fcodes, _PATH_FCODES);
 187         }
 188
 189         if (f_icase && UCHAR_MAX < 4096) /* init tolower lookup table */
 190                 for (ch = 0; ch < UCHAR_MAX + 1; ch++)
 191                         myctype[ch] = tolower(ch);
 192
 193         /* foreach database ... */
 194         while((path_fcodes = *dbv) != NULL) {
 195                 dbv++;
 196
 197                 if (!strcmp(path_fcodes, "-"))
 198                         f_stdin = 1;
 199                 else
 200                         f_stdin = 0;
 201
 202 #ifndef MMAP
 203                 f_mmap = 0;     /* be paranoid */
 204 #endif
 205                 if (!f_mmap || f_stdin || f_statistic)
 206                         search_fopen(path_fcodes, argv);
 207                 else
 208                         search_mmap(path_fcodes, argv);
 209         }
 210
 211         if (f_silent)
 212                 print_matches(counter);
 213         exit(0);
 214 }
 215
 216
 217 static void
 218 search_fopen(char *db, char **s)
 219 {
 220         FILE *fp;
 221 #ifdef DEBUG
 222         long t0;
 223 #endif
 224
 225         /* can only read stdin once */
 226         if (f_stdin) {
 227                 fp = stdin;
 228                 if (*(s+1) != NULL) {
 229                         warnx("read database from stdin, use only `%s' as pattern", *s);
 230                         *(s+1) = NULL;
 231                 }
 232         }
 233         else if ((fp = fopen(path_fcodes, "r")) == NULL)
 234                 err(1,  "`%s'", path_fcodes);
 235
 236         /* count only chars or lines */
 237         if (f_statistic) {
 238                 statistic(fp, path_fcodes);
 239                 (void)fclose(fp);
 240                 return;
 241         }
 242
 243         /* foreach search string ... */
 244         while(*s != NULL) {
 245 #ifdef DEBUG
 246                 t0 = cputime();
 247 #endif
 248                 if (!f_stdin &&
 249                     fseek(fp, (long)0, SEEK_SET) == -1)
 250                         err(1, "fseek to begin of ``%s''\n", path_fcodes);
 251
 252                 if (f_icase)
 253                         fastfind_icase(fp, *s, path_fcodes);
 254                 else
 255                         fastfind(fp, *s, path_fcodes);
 256 #ifdef DEBUG
 257                 warnx("fastfind %ld ms", cputime () - t0);
 258 #endif
 259                 s++;
 260         }
 261         (void)fclose(fp);
 262 }
 263
 264 #ifdef MMAP
 265 static void
 266 search_mmap(char *db, char **s)
 267 {
 268         struct stat sb;
 269         int fd;
 270         caddr_t p;
 271         off_t len;
 272 #ifdef DEBUG
 273         long t0;
 274 #endif
 275         if ((fd = open(path_fcodes, O_RDONLY)) == -1 ||
 276             fstat(fd, &sb) == -1)
 277                 err(1, "`%s'", path_fcodes);
 278         len = sb.st_size;
 279
 280         if ((p = mmap((caddr_t)0, (size_t)len,
 281                       PROT_READ, MAP_SHARED,
 282                       fd, (off_t)0)) == MAP_FAILED)
 283                 err(1, "mmap ``%s''", path_fcodes);
 284
 285         /* foreach search string ... */
 286         while (*s != NULL) {
 287 #ifdef DEBUG
 288                 t0 = cputime();
 289 #endif
 290                 if (f_icase)
 291                         fastfind_mmap_icase(*s, p, (int)len, path_fcodes);
 292                 else
 293                         fastfind_mmap(*s, p, (int)len, path_fcodes);
 294 #ifdef DEBUG
 295                 warnx("fastfind %ld ms", cputime () - t0);
 296 #endif
 297                 s++;
 298         }
 299
 300         if (munmap(p, (size_t)len) == -1)
 301                 warn("munmap %s\n", path_fcodes);
 302
 303         (void)close(fd);
 304 }
 305 #endif /* MMAP */
 306
 307 #ifdef DEBUG
 308 static unsigned long
 309 cputime (void)
 310 {
 311         struct rusage rus;
 312
 313         getrusage(0, &rus);
 314         return(rus.ru_utime.tv_sec * 1000 + rus.ru_utime.tv_usec / 1000);
 315 }
 316 #endif /* DEBUG */
 317
 318 static void
 319 usage (void)
 320 {
 321         (void)fprintf(stderr,
 322         "usage: locate [-Scims] [-l limit] [-d database] pattern ...\n\n");
 323         (void)fprintf(stderr,
 324         "default database: `%s' or $LOCATE_PATH\n", _PATH_FCODES);
 325         exit(1);
 326 }
 327
 328
 329 /* load fastfind functions */
 330
 331 /* statistic */
 332 /* fastfind_mmap, fastfind_mmap_icase */
 333 #ifdef MMAP
 334 #undef FF_MMAP
 335 #undef FF_ICASE
 336
 337 #define FF_MMAP
 338 #include "fastfind.c"
 339 #define FF_ICASE
 340 #include "fastfind.c"
 341 #endif /* MMAP */
 342
 343 /* fopen */
 344 /* fastfind, fastfind_icase */
 345 #undef FF_MMAP
 346 #undef FF_ICASE
 347 #include "fastfind.c"
 348 #define FF_ICASE
 349 #include "fastfind.c"