1/*-
2 * Copyright (c) 2003-2007 Tim Kientzle
3 * All rights reserved.
4 *
5 * Redistribution and use in source and binary forms, with or without
6 * modification, are permitted provided that the following conditions
7 * are met:
8 * 1. Redistributions of source code must retain the above copyright
9 *    notice, this list of conditions and the following disclaimer.
10 * 2. Redistributions in binary form must reproduce the above copyright
11 *    notice, this list of conditions and the following disclaimer in the
12 *    documentation and/or other materials provided with the distribution.
13 *
14 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR(S) ``AS IS'' AND ANY EXPRESS OR
15 * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
16 * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
17 * IN NO EVENT SHALL THE AUTHOR(S) BE LIABLE FOR ANY DIRECT, INDIRECT,
18 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
19 * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
20 * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
21 * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
22 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
23 * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
24 */
25
26#include "lafe_platform.h"
27__FBSDID("$FreeBSD: src/usr.bin/cpio/matching.c,v 1.2 2008/06/21 02:20:20 kientzle Exp $");
28
29#ifdef HAVE_ERRNO_H
30#include <errno.h>
31#endif
32#ifdef HAVE_STDLIB_H
33#include <stdlib.h>
34#endif
35#ifdef HAVE_STRING_H
36#include <string.h>
37#endif
38
39#include "err.h"
40#include "line_reader.h"
41#include "matching.h"
42#include "pathmatch.h"
43
44struct match {
45	struct match	 *next;
46	int		  matches;
47	char		  pattern[1];
48};
49
50struct lafe_matching {
51	struct match	 *exclusions;
52	int		  exclusions_count;
53	struct match	 *inclusions;
54	int		  inclusions_count;
55	int		  inclusions_unmatched_count;
56};
57
58static void	add_pattern(struct match **list, const char *pattern);
59static void	initialize_matching(struct lafe_matching **);
60static int	match_exclusion(struct match *, const char *pathname);
61static int	match_inclusion(struct match *, const char *pathname);
62
63/*
64 * The matching logic here needs to be re-thought.  I started out to
65 * try to mimic gtar's matching logic, but it's not entirely
66 * consistent.  In particular 'tar -t' and 'tar -x' interpret patterns
67 * on the command line as anchored, but --exclude doesn't.
68 */
69
70/*
71 * Utility functions to manage exclusion/inclusion patterns
72 */
73
74int
75lafe_exclude(struct lafe_matching **matching, const char *pattern)
76{
77
78	if (*matching == NULL)
79		initialize_matching(matching);
80	add_pattern(&((*matching)->exclusions), pattern);
81	(*matching)->exclusions_count++;
82	return (0);
83}
84
85int
86lafe_exclude_from_file(struct lafe_matching **matching, const char *pathname)
87{
88	struct lafe_line_reader *lr;
89	const char *p;
90	int ret = 0;
91
92	lr = lafe_line_reader(pathname, 0);
93	while ((p = lafe_line_reader_next(lr)) != NULL) {
94		if (lafe_exclude(matching, p) != 0)
95			ret = -1;
96	}
97	lafe_line_reader_free(lr);
98	return (ret);
99}
100
101int
102lafe_include(struct lafe_matching **matching, const char *pattern)
103{
104
105	if (*matching == NULL)
106		initialize_matching(matching);
107	add_pattern(&((*matching)->inclusions), pattern);
108	(*matching)->inclusions_count++;
109	(*matching)->inclusions_unmatched_count++;
110	return (0);
111}
112
113int
114lafe_include_from_file(struct lafe_matching **matching, const char *pathname,
115    int nullSeparator)
116{
117	struct lafe_line_reader *lr;
118	const char *p;
119	int ret = 0;
120
121	lr = lafe_line_reader(pathname, nullSeparator);
122	while ((p = lafe_line_reader_next(lr)) != NULL) {
123		if (lafe_include(matching, p) != 0)
124			ret = -1;
125	}
126	lafe_line_reader_free(lr);
127	return (ret);
128}
129
130static void
131add_pattern(struct match **list, const char *pattern)
132{
133	struct match *match;
134	size_t len;
135
136	len = strlen(pattern);
137	match = malloc(sizeof(*match) + len + 1);
138	if (match == NULL)
139		lafe_errc(1, errno, "Out of memory");
140	strcpy(match->pattern, pattern);
141	/* Both "foo/" and "foo" should match "foo/bar". */
142	if (len && match->pattern[len - 1] == '/')
143		match->pattern[strlen(match->pattern)-1] = '\0';
144	match->next = *list;
145	*list = match;
146	match->matches = 0;
147}
148
149
150int
151lafe_excluded(struct lafe_matching *matching, const char *pathname)
152{
153	struct match *match;
154	struct match *matched;
155
156	if (matching == NULL)
157		return (0);
158
159	/* Exclusions take priority */
160	for (match = matching->exclusions; match != NULL; match = match->next){
161		if (match_exclusion(match, pathname))
162			return (1);
163	}
164
165	/* Then check for inclusions */
166	matched = NULL;
167	for (match = matching->inclusions; match != NULL; match = match->next){
168		if (match_inclusion(match, pathname)) {
169			/*
170			 * If this pattern has never been matched,
171			 * then we're done.
172			 */
173			if (match->matches == 0) {
174				match->matches++;
175				matching->inclusions_unmatched_count--;
176				return (0);
177			}
178			/*
179			 * Otherwise, remember the match but keep checking
180			 * in case we can tick off an unmatched pattern.
181			 */
182			matched = match;
183		}
184	}
185	/*
186	 * We didn't find a pattern that had never been matched, but
187	 * we did find a match, so count it and exit.
188	 */
189	if (matched != NULL) {
190		matched->matches++;
191		return (0);
192	}
193
194	/* If there were inclusions, default is to exclude. */
195	if (matching->inclusions != NULL)
196	    return (1);
197
198	/* No explicit inclusions, default is to match. */
199	return (0);
200}
201
202/*
203 * This is a little odd, but it matches the default behavior of
204 * gtar.  In particular, 'a*b' will match 'foo/a1111/222b/bar'
205 *
206 */
207static int
208match_exclusion(struct match *match, const char *pathname)
209{
210	return (lafe_pathmatch(match->pattern,
211		    pathname,
212		    PATHMATCH_NO_ANCHOR_START | PATHMATCH_NO_ANCHOR_END));
213}
214
215/*
216 * Again, mimic gtar:  inclusions are always anchored (have to match
217 * the beginning of the path) even though exclusions are not anchored.
218 */
219static int
220match_inclusion(struct match *match, const char *pathname)
221{
222#if 0
223	return (lafe_pathmatch(match->pattern, pathname, 0));
224#else
225	return (lafe_pathmatch(match->pattern, pathname, PATHMATCH_NO_ANCHOR_END));
226#endif
227}
228
229void
230lafe_cleanup_exclusions(struct lafe_matching **matching)
231{
232	struct match *p, *q;
233
234	if (*matching == NULL)
235		return;
236
237	for (p = (*matching)->inclusions; p != NULL; ) {
238		q = p;
239		p = p->next;
240		free(q);
241	}
242
243	for (p = (*matching)->exclusions; p != NULL; ) {
244		q = p;
245		p = p->next;
246		free(q);
247	}
248
249	free(*matching);
250	*matching = NULL;
251}
252
253static void
254initialize_matching(struct lafe_matching **matching)
255{
256	*matching = calloc(sizeof(**matching), 1);
257	if (*matching == NULL)
258		lafe_errc(1, errno, "No memory");
259}
260
261int
262lafe_unmatched_inclusions(struct lafe_matching *matching)
263{
264
265	if (matching == NULL)
266		return (0);
267	return (matching->inclusions_unmatched_count);
268}
269
270int
271lafe_unmatched_inclusions_warn(struct lafe_matching *matching, const char *msg)
272{
273	struct match *p;
274
275	if (matching == NULL)
276		return (0);
277
278	for (p = matching->inclusions; p != NULL; p = p->next) {
279		if (p->matches == 0)
280			lafe_warnc(0, "%s: %s", p->pattern, msg);
281	}
282
283	return (matching->inclusions_unmatched_count);
284}
285