1 /*-
2 * SPDX-License-Identifier: BSD-3-Clause
3 *
4 * Copyright (c) 1989, 1993, 1994
5 * The Regents of the University of California. All rights reserved.
6 *
7 * This code is derived from software contributed to Berkeley by
8 * Guido van Rossum.
9 *
10 * Copyright (c) 2011 The FreeBSD Foundation
11 *
12 * Portions of this software were developed by David Chisnall
13 * under sponsorship from the FreeBSD Foundation.
14 *
15 * Redistribution and use in source and binary forms, with or without
16 * modification, are permitted provided that the following conditions
17 * are met:
18 * 1. Redistributions of source code must retain the above copyright
19 * notice, this list of conditions and the following disclaimer.
20 * 2. Redistributions in binary form must reproduce the above copyright
21 * notice, this list of conditions and the following disclaimer in the
22 * documentation and/or other materials provided with the distribution.
23 * 3. 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
40 #include "lafe_platform.h"
41
42 #ifndef HAVE_FNMATCH
43 /*
44 * Function fnmatch() as specified in POSIX 1003.2-1992, section B.6.
45 * Compares a filename or pathname to a pattern.
46 */
47
48 /*
49 * Some notes on multibyte character support:
50 * 1. Patterns with illegal byte sequences match nothing.
51 * 2. Illegal byte sequences in the "string" argument are handled by treating
52 * them as single-byte characters with a value of the first byte of the
53 * sequence cast to wchar_t.
54 * 3. Multibyte conversion state objects (mbstate_t) are passed around and
55 * used for most, but not all, conversions. Further work will be required
56 * to support state-dependent encodings.
57 */
58
59 #ifdef HAVE_LIMITS_H
60 #include <limits.h>
61 #endif
62 #ifdef HAVE_STRING_H
63 #include <string.h>
64 #endif
65 #ifdef HAVE_WCHAR_H
66 #include <wchar.h>
67 #endif
68 #ifdef HAVE_WCTYPE_H
69 #include <wctype.h>
70 #endif
71
72 #include "lafe_fnmatch.h"
73
74 #define EOS '\0'
75
76 #define RANGE_MATCH 1
77 #define RANGE_NOMATCH 0
78 #define RANGE_ERROR (-1)
79
80 static int rangematch(const char *, wchar_t, int, const char **, mbstate_t *);
81 static int fnmatch1(const char *, const char *, const char *, int, mbstate_t,
82 mbstate_t);
83
84 int
fnmatch(const char * pattern,const char * string,int flags)85 fnmatch(const char *pattern, const char *string, int flags)
86 {
87 static const mbstate_t initial;
88
89 return (fnmatch1(pattern, string, string, flags, initial, initial));
90 }
91
92 static int
fnmatch1(const char * pattern,const char * string,const char * stringstart,int flags,mbstate_t patmbs,mbstate_t strmbs)93 fnmatch1(const char *pattern, const char *string, const char *stringstart,
94 int flags, mbstate_t patmbs, mbstate_t strmbs)
95 {
96 const char *bt_pattern, *bt_string;
97 mbstate_t bt_patmbs, bt_strmbs;
98 const char *newp;
99 char c;
100 wchar_t pc, sc;
101 size_t pclen, sclen;
102
103 bt_pattern = bt_string = NULL;
104 for (;;) {
105 pclen = mbrtowc(&pc, pattern, MB_LEN_MAX, &patmbs);
106 if (pclen == (size_t)-1 || pclen == (size_t)-2)
107 return (FNM_NOMATCH);
108 pattern += pclen;
109 sclen = mbrtowc(&sc, string, MB_LEN_MAX, &strmbs);
110 if (sclen == (size_t)-1 || sclen == (size_t)-2) {
111 sc = (unsigned char)*string;
112 sclen = 1;
113 memset(&strmbs, 0, sizeof(strmbs));
114 }
115 switch (pc) {
116 case EOS:
117 if ((flags & FNM_LEADING_DIR) && sc == '/')
118 return (0);
119 if (sc == EOS)
120 return (0);
121 goto backtrack;
122 case '?':
123 if (sc == EOS)
124 return (FNM_NOMATCH);
125 if (sc == '/' && (flags & FNM_PATHNAME))
126 goto backtrack;
127 if (sc == '.' && (flags & FNM_PERIOD) &&
128 (string == stringstart ||
129 ((flags & FNM_PATHNAME) && *(string - 1) == '/')))
130 goto backtrack;
131 string += sclen;
132 break;
133 case '*':
134 c = *pattern;
135 /* Collapse multiple stars. */
136 while (c == '*')
137 c = *++pattern;
138
139 if (sc == '.' && (flags & FNM_PERIOD) &&
140 (string == stringstart ||
141 ((flags & FNM_PATHNAME) && *(string - 1) == '/')))
142 goto backtrack;
143
144 /* Optimize for pattern with * at end or before /. */
145 if (c == EOS)
146 if (flags & FNM_PATHNAME)
147 return ((flags & FNM_LEADING_DIR) ||
148 strchr(string, '/') == NULL ?
149 0 : FNM_NOMATCH);
150 else
151 return (0);
152 else if (c == '/' && flags & FNM_PATHNAME) {
153 if ((string = strchr(string, '/')) == NULL)
154 return (FNM_NOMATCH);
155 break;
156 }
157
158 /*
159 * First try the shortest match for the '*' that
160 * could work. We can forget any earlier '*' since
161 * there is no way having it match more characters
162 * can help us, given that we are already here.
163 */
164 bt_pattern = pattern, bt_patmbs = patmbs;
165 bt_string = string, bt_strmbs = strmbs;
166 break;
167 case '[':
168 if (sc == EOS)
169 return (FNM_NOMATCH);
170 if (sc == '/' && (flags & FNM_PATHNAME))
171 goto backtrack;
172 if (sc == '.' && (flags & FNM_PERIOD) &&
173 (string == stringstart ||
174 ((flags & FNM_PATHNAME) && *(string - 1) == '/')))
175 goto backtrack;
176
177 switch (rangematch(pattern, sc, flags, &newp,
178 &patmbs)) {
179 case RANGE_ERROR:
180 goto norm;
181 case RANGE_MATCH:
182 pattern = newp;
183 break;
184 case RANGE_NOMATCH:
185 goto backtrack;
186 }
187 string += sclen;
188 break;
189 case '\\':
190 if (!(flags & FNM_NOESCAPE)) {
191 pclen = mbrtowc(&pc, pattern, MB_LEN_MAX,
192 &patmbs);
193 if (pclen == 0 || pclen == (size_t)-1 ||
194 pclen == (size_t)-2)
195 return (FNM_NOMATCH);
196 pattern += pclen;
197 }
198 /* FALLTHROUGH */
199 default:
200 norm:
201 string += sclen;
202 if (pc == sc)
203 ;
204 else if ((flags & FNM_CASEFOLD) &&
205 (towlower(pc) == towlower(sc)))
206 ;
207 else {
208 backtrack:
209 /*
210 * If we have a mismatch (other than hitting
211 * the end of the string), go back to the last
212 * '*' seen and have it match one additional
213 * character.
214 */
215 if (bt_pattern == NULL)
216 return (FNM_NOMATCH);
217 sclen = mbrtowc(&sc, bt_string, MB_LEN_MAX,
218 &bt_strmbs);
219 if (sclen == (size_t)-1 ||
220 sclen == (size_t)-2) {
221 sc = (unsigned char)*bt_string;
222 sclen = 1;
223 memset(&bt_strmbs, 0,
224 sizeof(bt_strmbs));
225 }
226 if (sc == EOS)
227 return (FNM_NOMATCH);
228 if (sc == '/' && flags & FNM_PATHNAME)
229 return (FNM_NOMATCH);
230 bt_string += sclen;
231 pattern = bt_pattern, patmbs = bt_patmbs;
232 string = bt_string, strmbs = bt_strmbs;
233 }
234 break;
235 }
236 }
237 /* NOTREACHED */
238 }
239
240 static int
rangematch(const char * pattern,wchar_t test,int flags,const char ** newp,mbstate_t * patmbs)241 rangematch(const char *pattern, wchar_t test, int flags, const char **newp,
242 mbstate_t *patmbs)
243 {
244 int negate, ok;
245 wchar_t s1[2], s2[2], t[2];
246 size_t pclen;
247 const char *origpat;
248
249 s1[1] = L'\0';
250 s2[1] = L'\0';
251
252 t[0] = test;
253 t[1] = L'\0';
254
255 /*
256 * A bracket expression starting with an unquoted circumflex
257 * character produces unspecified results (IEEE 1003.2-1992,
258 * 3.13.2). This implementation treats it like '!', for
259 * consistency with the regular expression syntax.
260 * J.T. Conklin (conklin@ngai.kaleida.com)
261 */
262 if ((negate = (*pattern == '!' || *pattern == '^')))
263 ++pattern;
264
265 if (flags & FNM_CASEFOLD)
266 t[0] = towlower(t[0]);
267
268 /*
269 * A right bracket shall lose its special meaning and represent
270 * itself in a bracket expression if it occurs first in the list.
271 * -- POSIX.2 2.8.3.2
272 */
273 ok = 0;
274 origpat = pattern;
275 for (;;) {
276 if (*pattern == ']' && pattern > origpat) {
277 pattern++;
278 break;
279 } else if (*pattern == '\0') {
280 return (RANGE_ERROR);
281 } else if (*pattern == '/' && (flags & FNM_PATHNAME)) {
282 return (RANGE_NOMATCH);
283 } else if (*pattern == '\\' && !(flags & FNM_NOESCAPE))
284 pattern++;
285 pclen = mbrtowc(s1, pattern, MB_LEN_MAX, patmbs);
286 if (pclen == (size_t)-1 || pclen == (size_t)-2)
287 return (RANGE_NOMATCH);
288 pattern += pclen;
289
290 if (flags & FNM_CASEFOLD)
291 s1[0] = towlower(s1[0]);
292
293 if (*pattern == '-' && *(pattern + 1) != EOS &&
294 *(pattern + 1) != ']') {
295 if (*++pattern == '\\' && !(flags & FNM_NOESCAPE))
296 if (*pattern != EOS)
297 pattern++;
298 pclen = mbrtowc(s2, pattern, MB_LEN_MAX, patmbs);
299 if (pclen == (size_t)-1 || pclen == (size_t)-2)
300 return (RANGE_NOMATCH);
301 pattern += pclen;
302 if (s2[0] == EOS)
303 return (RANGE_ERROR);
304
305 if (flags & FNM_CASEFOLD)
306 s2[0] = towlower(s2[0]);
307
308 if (wcscoll(s1, t) && wcscoll(t, s2) <= 0)
309 ok = 1;
310 } else if (s1[0] == t[0])
311 ok = 1;
312 }
313
314 *newp = pattern;
315 return (ok == negate ? RANGE_NOMATCH : RANGE_MATCH);
316 }
317 #endif
318