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 u_char myctype[UCHAR_MAX + 1];
111
112 static void usage(void) __dead2;
113 static void statistic(FILE *, char *);
114 static void fastfind(FILE *, char *, char *);
115 static void fastfind_icase(FILE *, char *, char *);
116 static void fastfind_mmap(char *, caddr_t, int, char *);
117 static void fastfind_mmap_icase(char *, caddr_t, int, char *);
118 static void search_mmap(char *, char **);
119 static void search_fopen(char *, char **);
120 #ifdef DEBUG
121 static unsigned long cputime(void);
122 #endif
123
124 extern char **colon(char **, char*, char*);
125 extern void print_matches(u_int);
126 extern int getwm(caddr_t);
127 extern int getwf(FILE *);
128 extern u_char *tolower_word(u_char *);
129 extern int check_bigram_char(int);
130 extern char *patprep(char *);
131
132 int
main(int argc,char ** argv)133 main(int argc, 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 environment */
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 static void
search_fopen(char * db,char ** s)219 search_fopen(char *db, char **s)
220 {
221 FILE *fp;
222 #ifdef DEBUG
223 long t0;
224 #endif
225
226 /* can only read stdin once */
227 if (f_stdin) {
228 fp = stdin;
229 if (*(s+1) != NULL) {
230 warnx("read database from stdin, use only `%s' as pattern", *s);
231 *(s+1) = NULL;
232 }
233 }
234 else if ((fp = fopen(path_fcodes, "r")) == NULL)
235 err(1, "`%s'", path_fcodes);
236
237 /* count only chars or lines */
238 if (f_statistic) {
239 statistic(fp, path_fcodes);
240 (void)fclose(fp);
241 return;
242 }
243
244 /* foreach search string ... */
245 while(*s != NULL) {
246 #ifdef DEBUG
247 t0 = cputime();
248 #endif
249 if (!f_stdin &&
250 fseek(fp, (long)0, SEEK_SET) == -1)
251 err(1, "fseek to begin of ``%s''\n", path_fcodes);
252
253 if (f_icase)
254 fastfind_icase(fp, *s, path_fcodes);
255 else
256 fastfind(fp, *s, path_fcodes);
257 #ifdef DEBUG
258 warnx("fastfind %ld ms", cputime () - t0);
259 #endif
260 s++;
261 }
262 (void)fclose(fp);
263 }
264
265 #ifdef MMAP
266 static void
search_mmap(char * db,char ** s)267 search_mmap(char *db, char **s)
268 {
269 struct stat sb;
270 int fd;
271 caddr_t p;
272 off_t len;
273 #ifdef DEBUG
274 long t0;
275 #endif
276 if ((fd = open(path_fcodes, O_RDONLY)) == -1 ||
277 fstat(fd, &sb) == -1)
278 err(1, "`%s'", path_fcodes);
279 len = sb.st_size;
280
281 if ((p = mmap((caddr_t)0, (size_t)len,
282 PROT_READ, MAP_SHARED,
283 fd, (off_t)0)) == MAP_FAILED)
284 err(1, "mmap ``%s''", path_fcodes);
285
286 /* foreach search string ... */
287 while (*s != NULL) {
288 #ifdef DEBUG
289 t0 = cputime();
290 #endif
291 if (f_icase)
292 fastfind_mmap_icase(*s, p, (int)len, path_fcodes);
293 else
294 fastfind_mmap(*s, p, (int)len, path_fcodes);
295 #ifdef DEBUG
296 warnx("fastfind %ld ms", cputime () - t0);
297 #endif
298 s++;
299 }
300
301 if (munmap(p, (size_t)len) == -1)
302 warn("munmap %s\n", path_fcodes);
303
304 (void)close(fd);
305 }
306 #endif /* MMAP */
307
308 #ifdef DEBUG
309 static unsigned long
cputime(void)310 cputime (void)
311 {
312 struct rusage rus;
313
314 getrusage(0, &rus);
315 return(rus.ru_utime.tv_sec * 1000 + rus.ru_utime.tv_usec / 1000);
316 }
317 #endif /* DEBUG */
318
319 static void
usage(void)320 usage (void)
321 {
322 (void)fprintf(stderr,
323 "usage: locate [-Scims] [-l limit] [-d database] pattern ...\n\n");
324 (void)fprintf(stderr,
325 "default database: `%s' or $LOCATE_PATH\n", _PATH_FCODES);
326 exit(1);
327 }
328
329
330 /* load fastfind functions */
331
332 /* statistic */
333 /* fastfind_mmap, fastfind_mmap_icase */
334 #ifdef MMAP
335 #undef FF_MMAP
336 #undef FF_ICASE
337
338 #define FF_MMAP
339 #include "fastfind.c"
340 #define FF_ICASE
341 #include "fastfind.c"
342 #endif /* MMAP */
343
344 /* fopen */
345 /* fastfind, fastfind_icase */
346 #undef FF_MMAP
347 #undef FF_ICASE
348 #include "fastfind.c"
349 #define FF_ICASE
350 #include "fastfind.c"
351