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 79 fnmatch(const char *pattern, const char *string, int flags) 80 { 81 static const mbstate_t initial; 82 83 return (fnmatch1(pattern, string, string, flags, initial, initial)); 84 } 85 86 static int 87 fnmatch1(const char *pattern, const char *string, const char *stringstart, 88 int flags, mbstate_t patmbs, mbstate_t strmbs) 89 { 90 char *newp; 91 char c; 92 wchar_t pc, sc; 93 size_t pclen, sclen; 94 95 for (;;) { 96 pclen = mbrtowc(&pc, pattern, MB_LEN_MAX, &patmbs); 97 if (pclen == (size_t)-1 || pclen == (size_t)-2) 98 return (FNM_NOMATCH); 99 pattern += pclen; 100 sclen = mbrtowc(&sc, string, MB_LEN_MAX, &strmbs); 101 if (sclen == (size_t)-1 || sclen == (size_t)-2) { 102 sc = (unsigned char)*string; 103 sclen = 1; 104 memset(&strmbs, 0, sizeof(strmbs)); 105 } 106 switch (pc) { 107 case EOS: 108 if ((flags & FNM_LEADING_DIR) && sc == '/') 109 return (0); 110 return (sc == EOS ? 0 : FNM_NOMATCH); 111 case '?': 112 if (sc == EOS) 113 return (FNM_NOMATCH); 114 if (sc == '/' && (flags & FNM_PATHNAME)) 115 return (FNM_NOMATCH); 116 if (sc == '.' && (flags & FNM_PERIOD) && 117 (string == stringstart || 118 ((flags & FNM_PATHNAME) && *(string - 1) == '/'))) 119 return (FNM_NOMATCH); 120 string += sclen; 121 break; 122 case '*': 123 c = *pattern; 124 /* Collapse multiple stars. */ 125 while (c == '*') 126 c = *++pattern; 127 128 if (sc == '.' && (flags & FNM_PERIOD) && 129 (string == stringstart || 130 ((flags & FNM_PATHNAME) && *(string - 1) == '/'))) 131 return (FNM_NOMATCH); 132 133 /* Optimize for pattern with * at end or before /. */ 134 if (c == EOS) 135 if (flags & FNM_PATHNAME) 136 return ((flags & FNM_LEADING_DIR) || 137 strchr(string, '/') == NULL ? 138 0 : FNM_NOMATCH); 139 else 140 return (0); 141 else if (c == '/' && flags & FNM_PATHNAME) { 142 if ((string = strchr(string, '/')) == NULL) 143 return (FNM_NOMATCH); 144 break; 145 } 146 147 /* General case, use recursion. */ 148 while (sc != EOS) { 149 if (!fnmatch1(pattern, string, stringstart, 150 flags, patmbs, strmbs)) 151 return (0); 152 sclen = mbrtowc(&sc, string, MB_LEN_MAX, 153 &strmbs); 154 if (sclen == (size_t)-1 || 155 sclen == (size_t)-2) { 156 sc = (unsigned char)*string; 157 sclen = 1; 158 memset(&strmbs, 0, sizeof(strmbs)); 159 } 160 if (sc == '/' && flags & FNM_PATHNAME) 161 break; 162 string += sclen; 163 } 164 return (FNM_NOMATCH); 165 case '[': 166 if (sc == EOS) 167 return (FNM_NOMATCH); 168 if (sc == '/' && (flags & FNM_PATHNAME)) 169 return (FNM_NOMATCH); 170 if (sc == '.' && (flags & FNM_PERIOD) && 171 (string == stringstart || 172 ((flags & FNM_PATHNAME) && *(string - 1) == '/'))) 173 return (FNM_NOMATCH); 174 175 switch (rangematch(pattern, sc, flags, &newp, 176 &patmbs)) { 177 case RANGE_ERROR: 178 goto norm; 179 case RANGE_MATCH: 180 pattern = newp; 181 break; 182 case RANGE_NOMATCH: 183 return (FNM_NOMATCH); 184 } 185 string += sclen; 186 break; 187 case '\\': 188 if (!(flags & FNM_NOESCAPE)) { 189 pclen = mbrtowc(&pc, pattern, MB_LEN_MAX, 190 &patmbs); 191 if (pclen == (size_t)-1 || pclen == (size_t)-2) 192 return (FNM_NOMATCH); 193 pattern += pclen; 194 } 195 /* FALLTHROUGH */ 196 default: 197 norm: 198 if (pc == sc) 199 ; 200 else if ((flags & FNM_CASEFOLD) && 201 (towlower(pc) == towlower(sc))) 202 ; 203 else 204 return (FNM_NOMATCH); 205 string += sclen; 206 break; 207 } 208 } 209 /* NOTREACHED */ 210 } 211 212 static int 213 rangematch(const char *pattern, wchar_t test, int flags, char **newp, 214 mbstate_t *patmbs) 215 { 216 int negate, ok; 217 wchar_t c, c2; 218 size_t pclen; 219 const char *origpat; 220 struct xlocale_collate *table = 221 (struct xlocale_collate*)__get_locale()->components[XLC_COLLATE]; 222 223 /* 224 * A bracket expression starting with an unquoted circumflex 225 * character produces unspecified results (IEEE 1003.2-1992, 226 * 3.13.2). This implementation treats it like '!', for 227 * consistency with the regular expression syntax. 228 * J.T. Conklin (conklin@ngai.kaleida.com) 229 */ 230 if ( (negate = (*pattern == '!' || *pattern == '^')) ) 231 ++pattern; 232 233 if (flags & FNM_CASEFOLD) 234 test = towlower(test); 235 236 /* 237 * A right bracket shall lose its special meaning and represent 238 * itself in a bracket expression if it occurs first in the list. 239 * -- POSIX.2 2.8.3.2 240 */ 241 ok = 0; 242 origpat = pattern; 243 for (;;) { 244 if (*pattern == ']' && pattern > origpat) { 245 pattern++; 246 break; 247 } else if (*pattern == '\0') { 248 return (RANGE_ERROR); 249 } else if (*pattern == '/' && (flags & FNM_PATHNAME)) { 250 return (RANGE_NOMATCH); 251 } else if (*pattern == '\\' && !(flags & FNM_NOESCAPE)) 252 pattern++; 253 pclen = mbrtowc(&c, pattern, MB_LEN_MAX, patmbs); 254 if (pclen == (size_t)-1 || pclen == (size_t)-2) 255 return (RANGE_NOMATCH); 256 pattern += pclen; 257 258 if (flags & FNM_CASEFOLD) 259 c = towlower(c); 260 261 if (*pattern == '-' && *(pattern + 1) != EOS && 262 *(pattern + 1) != ']') { 263 if (*++pattern == '\\' && !(flags & FNM_NOESCAPE)) 264 if (*pattern != EOS) 265 pattern++; 266 pclen = mbrtowc(&c2, pattern, MB_LEN_MAX, patmbs); 267 if (pclen == (size_t)-1 || pclen == (size_t)-2) 268 return (RANGE_NOMATCH); 269 pattern += pclen; 270 if (c2 == EOS) 271 return (RANGE_ERROR); 272 273 if (flags & FNM_CASEFOLD) 274 c2 = towlower(c2); 275 276 if (table->__collate_load_error ? 277 c <= test && test <= c2 : 278 __collate_range_cmp(table, c, test) <= 0 279 && __collate_range_cmp(table, test, c2) <= 0 280 ) 281 ok = 1; 282 } else if (c == test) 283 ok = 1; 284 } 285 286 *newp = (char *)pattern; 287 return (ok == negate ? RANGE_NOMATCH : RANGE_MATCH); 288 } 289