1 /*
2 * Copyright (c) 1989, 1993, 1994
3 * The Regents of the University of California. All rights reserved.
4 *
5 * This code is derived from software contributed to Berkeley by
6 * Guido van Rossum.
7 *
8 * Copyright (c) 2011 The FreeBSD Foundation
9 * All rights reserved.
10 * Portions of this software were developed by David Chisnall
11 * under sponsorship from the FreeBSD Foundation.
12 *
13 * Redistribution and use in source and binary forms, with or without
14 * modification, are permitted provided that the following conditions
15 * are met:
16 * 1. Redistributions of source code must retain the above copyright
17 * notice, this list of conditions and the following disclaimer.
18 * 2. Redistributions in binary form must reproduce the above copyright
19 * notice, this list of conditions and the following disclaimer in the
20 * documentation and/or other materials provided with the distribution.
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
38 #if defined(LIBC_SCCS) && !defined(lint)
39 static char sccsid[] = "@(#)fnmatch.c 8.2 (Berkeley) 4/16/94";
40 #endif /* LIBC_SCCS and not lint */
41 #include <sys/cdefs.h>
42 __FBSDID("$FreeBSD$");
43
44 /*
45 * Function fnmatch() as specified in POSIX 1003.2-1992, section B.6.
46 * Compares a filename or pathname to a pattern.
47 */
48
49 /*
50 * Some notes on multibyte character support:
51 * 1. Patterns with illegal byte sequences match nothing.
52 * 2. Illegal byte sequences in the "string" argument are handled by treating
53 * them as single-byte characters with a value of the first byte of the
54 * sequence cast to wchar_t.
55 * 3. Multibyte conversion state objects (mbstate_t) are passed around and
56 * used for most, but not all, conversions. Further work will be required
57 * to support state-dependent encodings.
58 */
59
60 #include <fnmatch.h>
61 #include <limits.h>
62 #include <string.h>
63 #include <wchar.h>
64 #include <wctype.h>
65
66 #include "collate.h"
67
68 #define EOS '\0'
69
70 #define RANGE_MATCH 1
71 #define RANGE_NOMATCH 0
72 #define RANGE_ERROR (-1)
73
74 static int rangematch(const char *, wchar_t, int, char **, mbstate_t *);
75 static int fnmatch1(const char *, const char *, const char *, int, mbstate_t,
76 mbstate_t);
77
78 int
fnmatch(pattern,string,flags)79 fnmatch(pattern, string, flags)
80 const char *pattern, *string;
81 int flags;
82 {
83 static const mbstate_t initial;
84
85 return (fnmatch1(pattern, string, string, flags, initial, initial));
86 }
87
88 static int
fnmatch1(pattern,string,stringstart,flags,patmbs,strmbs)89 fnmatch1(pattern, string, stringstart, flags, patmbs, strmbs)
90 const char *pattern, *string, *stringstart;
91 int flags;
92 mbstate_t patmbs, strmbs;
93 {
94 char *newp;
95 char c;
96 wchar_t pc, sc;
97 size_t pclen, sclen;
98
99 for (;;) {
100 pclen = mbrtowc(&pc, pattern, MB_LEN_MAX, &patmbs);
101 if (pclen == (size_t)-1 || pclen == (size_t)-2)
102 return (FNM_NOMATCH);
103 pattern += pclen;
104 sclen = mbrtowc(&sc, string, MB_LEN_MAX, &strmbs);
105 if (sclen == (size_t)-1 || sclen == (size_t)-2) {
106 sc = (unsigned char)*string;
107 sclen = 1;
108 memset(&strmbs, 0, sizeof(strmbs));
109 }
110 switch (pc) {
111 case EOS:
112 if ((flags & FNM_LEADING_DIR) && sc == '/')
113 return (0);
114 return (sc == EOS ? 0 : FNM_NOMATCH);
115 case '?':
116 if (sc == EOS)
117 return (FNM_NOMATCH);
118 if (sc == '/' && (flags & FNM_PATHNAME))
119 return (FNM_NOMATCH);
120 if (sc == '.' && (flags & FNM_PERIOD) &&
121 (string == stringstart ||
122 ((flags & FNM_PATHNAME) && *(string - 1) == '/')))
123 return (FNM_NOMATCH);
124 string += sclen;
125 break;
126 case '*':
127 c = *pattern;
128 /* Collapse multiple stars. */
129 while (c == '*')
130 c = *++pattern;
131
132 if (sc == '.' && (flags & FNM_PERIOD) &&
133 (string == stringstart ||
134 ((flags & FNM_PATHNAME) && *(string - 1) == '/')))
135 return (FNM_NOMATCH);
136
137 /* Optimize for pattern with * at end or before /. */
138 if (c == EOS)
139 if (flags & FNM_PATHNAME)
140 return ((flags & FNM_LEADING_DIR) ||
141 strchr(string, '/') == NULL ?
142 0 : FNM_NOMATCH);
143 else
144 return (0);
145 else if (c == '/' && flags & FNM_PATHNAME) {
146 if ((string = strchr(string, '/')) == NULL)
147 return (FNM_NOMATCH);
148 break;
149 }
150
151 /* General case, use recursion. */
152 while (sc != EOS) {
153 if (!fnmatch1(pattern, string, stringstart,
154 flags, patmbs, strmbs))
155 return (0);
156 sclen = mbrtowc(&sc, string, MB_LEN_MAX,
157 &strmbs);
158 if (sclen == (size_t)-1 ||
159 sclen == (size_t)-2) {
160 sc = (unsigned char)*string;
161 sclen = 1;
162 memset(&strmbs, 0, sizeof(strmbs));
163 }
164 if (sc == '/' && flags & FNM_PATHNAME)
165 break;
166 string += sclen;
167 }
168 return (FNM_NOMATCH);
169 case '[':
170 if (sc == EOS)
171 return (FNM_NOMATCH);
172 if (sc == '/' && (flags & FNM_PATHNAME))
173 return (FNM_NOMATCH);
174 if (sc == '.' && (flags & FNM_PERIOD) &&
175 (string == stringstart ||
176 ((flags & FNM_PATHNAME) && *(string - 1) == '/')))
177 return (FNM_NOMATCH);
178
179 switch (rangematch(pattern, sc, flags, &newp,
180 &patmbs)) {
181 case RANGE_ERROR:
182 goto norm;
183 case RANGE_MATCH:
184 pattern = newp;
185 break;
186 case RANGE_NOMATCH:
187 return (FNM_NOMATCH);
188 }
189 string += sclen;
190 break;
191 case '\\':
192 if (!(flags & FNM_NOESCAPE)) {
193 pclen = mbrtowc(&pc, pattern, MB_LEN_MAX,
194 &patmbs);
195 if (pclen == (size_t)-1 || pclen == (size_t)-2)
196 return (FNM_NOMATCH);
197 pattern += pclen;
198 }
199 /* FALLTHROUGH */
200 default:
201 norm:
202 if (pc == sc)
203 ;
204 else if ((flags & FNM_CASEFOLD) &&
205 (towlower(pc) == towlower(sc)))
206 ;
207 else
208 return (FNM_NOMATCH);
209 string += sclen;
210 break;
211 }
212 }
213 /* NOTREACHED */
214 }
215
216 static int
rangematch(pattern,test,flags,newp,patmbs)217 rangematch(pattern, test, flags, newp, patmbs)
218 const char *pattern;
219 wchar_t test;
220 int flags;
221 char **newp;
222 mbstate_t *patmbs;
223 {
224 int negate, ok;
225 wchar_t c, c2;
226 size_t pclen;
227 const char *origpat;
228 struct xlocale_collate *table =
229 (struct xlocale_collate*)__get_locale()->components[XLC_COLLATE];
230
231 /*
232 * A bracket expression starting with an unquoted circumflex
233 * character produces unspecified results (IEEE 1003.2-1992,
234 * 3.13.2). This implementation treats it like '!', for
235 * consistency with the regular expression syntax.
236 * J.T. Conklin (conklin@ngai.kaleida.com)
237 */
238 if ( (negate = (*pattern == '!' || *pattern == '^')) )
239 ++pattern;
240
241 if (flags & FNM_CASEFOLD)
242 test = towlower(test);
243
244 /*
245 * A right bracket shall lose its special meaning and represent
246 * itself in a bracket expression if it occurs first in the list.
247 * -- POSIX.2 2.8.3.2
248 */
249 ok = 0;
250 origpat = pattern;
251 for (;;) {
252 if (*pattern == ']' && pattern > origpat) {
253 pattern++;
254 break;
255 } else if (*pattern == '\0') {
256 return (RANGE_ERROR);
257 } else if (*pattern == '/' && (flags & FNM_PATHNAME)) {
258 return (RANGE_NOMATCH);
259 } else if (*pattern == '\\' && !(flags & FNM_NOESCAPE))
260 pattern++;
261 pclen = mbrtowc(&c, pattern, MB_LEN_MAX, patmbs);
262 if (pclen == (size_t)-1 || pclen == (size_t)-2)
263 return (RANGE_NOMATCH);
264 pattern += pclen;
265
266 if (flags & FNM_CASEFOLD)
267 c = towlower(c);
268
269 if (*pattern == '-' && *(pattern + 1) != EOS &&
270 *(pattern + 1) != ']') {
271 if (*++pattern == '\\' && !(flags & FNM_NOESCAPE))
272 if (*pattern != EOS)
273 pattern++;
274 pclen = mbrtowc(&c2, pattern, MB_LEN_MAX, patmbs);
275 if (pclen == (size_t)-1 || pclen == (size_t)-2)
276 return (RANGE_NOMATCH);
277 pattern += pclen;
278 if (c2 == EOS)
279 return (RANGE_ERROR);
280
281 if (flags & FNM_CASEFOLD)
282 c2 = towlower(c2);
283
284 if (table->__collate_load_error ?
285 c <= test && test <= c2 :
286 __collate_range_cmp(table, c, test) <= 0
287 && __collate_range_cmp(table, test, c2) <= 0
288 )
289 ok = 1;
290 } else if (c == test)
291 ok = 1;
292 }
293
294 *newp = (char *)pattern;
295 return (ok == negate ? RANGE_NOMATCH : RANGE_MATCH);
296 }
297