1 /*
2 * SPDX-License-Identifier: BSD-4-Clause
3 *
4 * Copyright (c) 1995 Wolfram Schneider <[email protected]>. Berlin.
5 * Copyright (c) 1989, 1993
6 * The Regents of the University of California. All rights reserved.
7 *
8 * This code is derived from software contributed to Berkeley by
9 * James A. Woods.
10 *
11 * Redistribution and use in source and binary forms, with or without
12 * modification, are permitted provided that the following conditions
13 * are met:
14 * 1. Redistributions of source code must retain the above copyright
15 * notice, this list of conditions and the following disclaimer.
16 * 2. Redistributions in binary form must reproduce the above copyright
17 * notice, this list of conditions and the following disclaimer in the
18 * documentation and/or other materials provided with the distribution.
19 * 3. All advertising materials mentioning features or use of this software
20 * must display the following acknowledgement:
21 * This product includes software developed by the University of
22 * California, Berkeley and its contributors.
23 * 4. Neither the name of the University nor the names of its contributors
24 * may be used to endorse or promote products derived from this software
25 * without specific prior written permission.
26 *
27 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
28 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
29 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
30 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
31 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
32 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
33 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
34 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
35 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
36 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
37 * SUCH DAMAGE.
38 *
39 * $FreeBSD$
40 */
41
42
43 #include <stdlib.h>
44 #include <string.h>
45 #include <err.h>
46 #include <sys/param.h>
47 #include <arpa/inet.h>
48 #include <stdio.h>
49
50 #include "locate.h"
51
52 char **colon(char **, char*, char*);
53 char *patprep(char *);
54 void print_matches(u_int);
55 u_char *tolower_word(u_char *);
56 int getwm(caddr_t);
57 int getwf(FILE *);
58 int check_bigram_char(int);
59
60 /*
61 * Validate bigram chars. If the test failed the database is corrupt
62 * or the database is obviously not a locate database.
63 */
64 int
check_bigram_char(ch)65 check_bigram_char(ch)
66 int ch;
67 {
68 /* legal bigram: 0, ASCII_MIN ... ASCII_MAX */
69 if (ch == 0 ||
70 (ch >= ASCII_MIN && ch <= ASCII_MAX))
71 return(ch);
72
73 errx(1,
74 "locate database header corrupt, bigram char outside 0, %d-%d: %d",
75 ASCII_MIN, ASCII_MAX, ch);
76 exit(1);
77 }
78
79 /* split a colon separated string into a char vector
80 *
81 * "bla:foo" -> {"foo", "bla"}
82 * "bla:" -> {"foo", dot}
83 * "bla" -> {"bla"}
84 * "" -> do nothing
85 *
86 */
87 char **
colon(char ** dbv,char * path,char * dot)88 colon(char **dbv, char *path, char *dot)
89 {
90 int vlen, slen;
91 char *c, *ch, *p;
92 char **pv;
93
94 if (dbv == NULL) {
95 if ((dbv = malloc(sizeof(char *))) == NULL)
96 err(1, "malloc");
97 *dbv = NULL;
98 }
99
100 /* empty string */
101 if (*path == '\0') {
102 warnx("empty database name, ignored");
103 return(dbv);
104 }
105
106 /* length of string vector */
107 for(vlen = 0, pv = dbv; *pv != NULL; pv++, vlen++);
108
109 for (ch = c = path; ; ch++) {
110 if (*ch == ':' ||
111 (!*ch && !(*(ch - 1) == ':' && ch == 1+ path))) {
112 /* single colon -> dot */
113 if (ch == c)
114 p = dot;
115 else {
116 /* a string */
117 slen = ch - c;
118 if ((p = malloc(sizeof(char) * (slen + 1)))
119 == NULL)
120 err(1, "malloc");
121 bcopy(c, p, slen);
122 *(p + slen) = '\0';
123 }
124 /* increase dbv with element p */
125 if ((dbv = realloc(dbv, sizeof(char *) * (vlen + 2)))
126 == NULL)
127 err(1, "realloc");
128 *(dbv + vlen) = p;
129 *(dbv + ++vlen) = NULL;
130 c = ch + 1;
131 }
132 if (*ch == '\0')
133 break;
134 }
135 return (dbv);
136 }
137
138 void
print_matches(counter)139 print_matches(counter)
140 u_int counter;
141 {
142 (void)printf("%d\n", counter);
143 }
144
145
146 /*
147 * extract last glob-free subpattern in name for fast pre-match; prepend
148 * '\0' for backwards match; return end of new pattern
149 */
150 static char globfree[100];
151
152 char *
patprep(name)153 patprep(name)
154 char *name;
155 {
156 register char *endmark, *p, *subp;
157
158 subp = globfree;
159 *subp++ = '\0'; /* set first element to '\0' */
160 p = name + strlen(name) - 1;
161
162 /* skip trailing metacharacters */
163 for (; p >= name; p--)
164 if (strchr(LOCATE_REG, *p) == NULL)
165 break;
166
167 /*
168 * check if maybe we are in a character class
169 *
170 * 'foo.[ch]'
171 * |----< p
172 */
173 if (p >= name &&
174 (strchr(p, '[') != NULL || strchr(p, ']') != NULL)) {
175 for (p = name; *p != '\0'; p++)
176 if (*p == ']' || *p == '[')
177 break;
178 p--;
179
180 /*
181 * cannot find a non-meta character, give up
182 * '*\*[a-z]'
183 * |-------< p
184 */
185 if (p >= name && strchr(LOCATE_REG, *p) != NULL)
186 p = name - 1;
187 }
188
189 if (p < name)
190 /* only meta chars: "???", force '/' search */
191 *subp++ = '/';
192
193 else {
194 for (endmark = p; p >= name; p--)
195 if (strchr(LOCATE_REG, *p) != NULL)
196 break;
197 for (++p;
198 (p <= endmark) && subp < (globfree + sizeof(globfree));)
199 *subp++ = *p++;
200 }
201 *subp = '\0';
202 return(--subp);
203 }
204
205 /* tolower word */
206 u_char *
tolower_word(word)207 tolower_word(word)
208 u_char *word;
209 {
210 register u_char *p;
211
212 for(p = word; *p != '\0'; p++)
213 *p = TOLOWER(*p);
214
215 return(word);
216 }
217
218
219 /*
220 * Read integer from mmap pointer.
221 * Essentially a simple ``return *(int *)p'' but avoids sigbus
222 * for integer alignment (SunOS 4.x, 5.x).
223 *
224 * Convert network byte order to host byte order if necessary.
225 * So we can read a locate database on FreeBSD/i386 (little endian)
226 * which was built on SunOS/sparc (big endian).
227 */
228
229 int
getwm(p)230 getwm(p)
231 caddr_t p;
232 {
233 union {
234 char buf[INTSIZE];
235 int i;
236 } u;
237 register int i, hi;
238
239 for (i = 0; i < (int)INTSIZE; i++)
240 u.buf[i] = *p++;
241
242 i = u.i;
243
244 if (i > MAXPATHLEN || i < -(MAXPATHLEN)) {
245 hi = ntohl(i);
246 if (hi > MAXPATHLEN || hi < -(MAXPATHLEN))
247 errx(1, "integer out of +-MAXPATHLEN (%d): %u",
248 MAXPATHLEN, abs(i) < abs(hi) ? i : hi);
249 return(hi);
250 }
251 return(i);
252 }
253
254 /*
255 * Read integer from stream.
256 *
257 * Convert network byte order to host byte order if necessary.
258 * So we can read on FreeBSD/i386 (little endian) a locate database
259 * which was built on SunOS/sparc (big endian).
260 */
261
262 int
getwf(fp)263 getwf(fp)
264 FILE *fp;
265 {
266 register int word, hword;
267
268 word = getw(fp);
269
270 if (word > MAXPATHLEN || word < -(MAXPATHLEN)) {
271 hword = ntohl(word);
272 if (hword > MAXPATHLEN || hword < -(MAXPATHLEN))
273 errx(1, "integer out of +-MAXPATHLEN (%d): %u",
274 MAXPATHLEN, abs(word) < abs(hword) ? word : hword);
275 return(hword);
276 }
277 return(word);
278 }
279