lib/libc/gen/fnmatch.c

   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  * Redistribution and use in source and binary forms, with or without
   9  * modification, are permitted provided that the following conditions
  10  * are met:
  11  * 1. Redistributions of source code must retain the above copyright
  12  *    notice, this list of conditions and the following disclaimer.
  13  * 2. Redistributions in binary form must reproduce the above copyright
  14  *    notice, this list of conditions and the following disclaimer in the
  15  *    documentation and/or other materials provided with the distribution.
  16  * 4. Neither the name of the University nor the names of its contributors
  17  *    may be used to endorse or promote products derived from this software
  18  *    without specific prior written permission.
  19  *
  20  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
  21  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
  22  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
  23  * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
  24  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
  25  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
  26  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
  27  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
  28  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
  29  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
  30  * SUCH DAMAGE.
  31  *
  32  * @(#)fnmatch.c        8.2 (Berkeley) 4/16/94
  33  * $FreeBSD: src/lib/libc/gen/fnmatch.c,v 1.18 2007/01/09 00:27:53 imp Exp $
  34  * $DragonFly: src/lib/libc/gen/fnmatch.c,v 1.6 2005/11/13 00:07:42 swildner Exp $
  35  */
  36
  37 /*
  38  * Function fnmatch() as specified in POSIX 1003.2-1992, section B.6.
  39  * Compares a filename or pathname to a pattern.
  40  */
  41
  42 /*
  43  * Some notes on multibyte character support:
  44  * 1. Patterns with illegal byte sequences match nothing.
  45  * 2. Illegal byte sequences in the "string" argument are handled by treating
  46  *    them as single-byte characters with a value of the first byte of the
  47  *    sequence cast to wchar_t.
  48  * 3. Multibyte conversion state objects (mbstate_t) are passed around and
  49  *    used for most, but not all, conversions. Further work will be required
  50  *    to support state-dependent encodings.
  51  */
  52
  53 #include <fnmatch.h>
  54 #include <limits.h>
  55 #include <string.h>
  56 #include <wchar.h>
  57 #include <wctype.h>
  58
  59 #include "collate.h"
  60
  61 #define EOS     '\0'
  62
  63 #define RANGE_MATCH     1
  64 #define RANGE_NOMATCH   0
  65 #define RANGE_ERROR     (-1)
  66
  67 static int rangematch(const char *, wchar_t, int, char **, mbstate_t *);
  68 static int fnmatch1(const char *, const char *, int, mbstate_t, mbstate_t);
  69
  70 int
  71 fnmatch(const char *pattern, const char *string, int flags)
  72 {
  73         static const mbstate_t initial;
  74
  75         return (fnmatch1(pattern, string, flags, initial, initial));
  76 }
  77
  78 static int
  79 fnmatch1(const char *pattern, const char *string, int flags, mbstate_t patmbs,
  80          mbstate_t strmbs)
  81 {
  82         const char *stringstart;
  83         char *newp;
  84         char c;
  85         wchar_t pc, sc;
  86         size_t pclen, sclen;
  87
  88         for (stringstart = string;;) {
  89                 pclen = mbrtowc(&pc, pattern, MB_LEN_MAX, &patmbs);
  90                 if (pclen == (size_t)-1 || pclen == (size_t)-2)
  91                         return (FNM_NOMATCH);
  92                 pattern += pclen;
  93                 sclen = mbrtowc(&sc, string, MB_LEN_MAX, &strmbs);
  94                 if (sclen == (size_t)-1 || sclen == (size_t)-2) {
  95                         sc = (unsigned char)*string;
  96                         sclen = 1;
  97                         memset(&strmbs, 0, sizeof(strmbs));
  98                 }
  99                 switch (pc) {
 100                 case EOS:
 101                         if ((flags & FNM_LEADING_DIR) && sc == '/')
 102                                 return (0);
 103                         return (sc == EOS ? 0 : FNM_NOMATCH);
 104                 case '?':
 105                         if (sc == EOS)
 106                                 return (FNM_NOMATCH);
 107                         if (sc == '/' && (flags & FNM_PATHNAME))
 108                                 return (FNM_NOMATCH);
 109                         if (sc == '.' && (flags & FNM_PERIOD) &&
 110                             (string == stringstart ||
 111                             ((flags & FNM_PATHNAME) && *(string - 1) == '/')))
 112                                 return (FNM_NOMATCH);
 113                         string += sclen;
 114                         break;
 115                 case '*':
 116                         c = *pattern;
 117                         /* Collapse multiple stars. */
 118                         while (c == '*')
 119                                 c = *++pattern;
 120
 121                         if (sc == '.' && (flags & FNM_PERIOD) &&
 122                             (string == stringstart ||
 123                             ((flags & FNM_PATHNAME) && *(string - 1) == '/')))
 124                                 return (FNM_NOMATCH);
 125
 126                         /* Optimize for pattern with * at end or before /. */
 127                         if (c == EOS)
 128                                 if (flags & FNM_PATHNAME)
 129                                         return ((flags & FNM_LEADING_DIR) ||
 130                                             strchr(string, '/') == NULL ?
 131                                             0 : FNM_NOMATCH);
 132                                 else
 133                                         return (0);
 134                         else if (c == '/' && flags & FNM_PATHNAME) {
 135                                 if ((string = strchr(string, '/')) == NULL)
 136                                         return (FNM_NOMATCH);
 137                                 break;
 138                         }
 139
 140                         /* General case, use recursion. */
 141                         while (sc != EOS) {
 142                                 if (!fnmatch1(pattern, string,
 143                                     flags & ~FNM_PERIOD, patmbs, strmbs))
 144                                         return (0);
 145                                 sclen = mbrtowc(&sc, string, MB_LEN_MAX,
 146                                     &strmbs);
 147                                 if (sclen == (size_t)-1 ||
 148                                     sclen == (size_t)-2) {
 149                                         sc = (unsigned char)*string;
 150                                         sclen = 1;
 151                                         memset(&strmbs, 0, sizeof(strmbs));
 152                                 }
 153                                 if (sc == '/' && flags & FNM_PATHNAME)
 154                                         break;
 155                                 string += sclen;
 156                         }
 157                         return (FNM_NOMATCH);
 158                 case '[':
 159                         if (sc == EOS)
 160                                 return (FNM_NOMATCH);
 161                         if (sc == '/' && (flags & FNM_PATHNAME))
 162                                 return (FNM_NOMATCH);
 163                         if (sc == '.' && (flags & FNM_PERIOD) &&
 164                             (string == stringstart ||
 165                             ((flags & FNM_PATHNAME) && *(string - 1) == '/')))
 166                                 return (FNM_NOMATCH);
 167
 168                         switch (rangematch(pattern, sc, flags, &newp,
 169                             &patmbs)) {
 170                         case RANGE_ERROR:
 171                                 goto norm;
 172                         case RANGE_MATCH:
 173                                 pattern = newp;
 174                                 break;
 175                         case RANGE_NOMATCH:
 176                                 return (FNM_NOMATCH);
 177                         }
 178                         string += sclen;
 179                         break;
 180                 case '\\':
 181                         if (!(flags & FNM_NOESCAPE)) {
 182                                 pclen = mbrtowc(&pc, pattern, MB_LEN_MAX,
 183                                     &patmbs);
 184                                 if (pclen == (size_t)-1 || pclen == (size_t)-2)
 185                                         return (FNM_NOMATCH);
 186                                 if (pclen == 0)
 187                                         pc = '\\';
 188                                 pattern += pclen;
 189                         }
 190                         /* FALLTHROUGH */
 191                 default:
 192                 norm:
 193                         if (pc == sc)
 194                                 ;
 195                         else if ((flags & FNM_CASEFOLD) &&
 196                                  (towlower(pc) == towlower(sc)))
 197                                 ;
 198                         else
 199                                 return (FNM_NOMATCH);
 200                         string += sclen;
 201                         break;
 202                 }
 203         }
 204         /* NOTREACHED */
 205 }
 206
 207 static int
 208 rangematch(const char *pattern, wchar_t test, int flags, char **newp,
 209            mbstate_t *patmbs)
 210 {
 211         int negate, ok;
 212         wchar_t c, c2;
 213         size_t pclen;
 214         const char *origpat;
 215
 216         /*
 217          * A bracket expression starting with an unquoted circumflex
 218          * character produces unspecified results (IEEE 1003.2-1992,
 219          * 3.13.2).  This implementation treats it like '!', for
 220          * consistency with the regular expression syntax.
 221          * J.T. Conklin (conklin@ngai.kaleida.com)
 222          */
 223         if ((negate = (*pattern == '!' || *pattern == '^')))
 224                 ++pattern;
 225
 226         if (flags & FNM_CASEFOLD)
 227                 test = towlower(test);
 228
 229         /*
 230          * A right bracket shall lose its special meaning and represent
 231          * itself in a bracket expression if it occurs first in the list.
 232          * -- POSIX.2 2.8.3.2
 233          */
 234         ok = 0;
 235         origpat = pattern;
 236         for (;;) {
 237                 if (*pattern == ']' && pattern > origpat) {
 238                         pattern++;
 239                         break;
 240                 } else if (*pattern == '\0') {
 241                         return (RANGE_ERROR);
 242                 } else if (*pattern == '/' && (flags & FNM_PATHNAME)) {
 243                         return (RANGE_NOMATCH);
 244                 } else if (*pattern == '\\' && !(flags & FNM_NOESCAPE)) {
 245                         pattern++;
 246                 }
 247                 pclen = mbrtowc(&c, pattern, MB_LEN_MAX, patmbs);
 248                 if (pclen == (size_t)-1 || pclen == (size_t)-2)
 249                         return (RANGE_NOMATCH);
 250                 pattern += pclen;
 251
 252                 if (flags & FNM_CASEFOLD)
 253                         c = towlower(c);
 254
 255                 if (*pattern == '-' && *(pattern + 1) != EOS &&
 256                     *(pattern + 1) != ']') {
 257                         if (*++pattern == '\\' && !(flags & FNM_NOESCAPE)) {
 258                                 if (*pattern != EOS)
 259                                         pattern++;
 260                         }
 261                         pclen = mbrtowc(&c2, pattern, MB_LEN_MAX, patmbs);
 262                         if (pclen == (size_t)-1 || pclen == (size_t)-2)
 263                                 return (RANGE_NOMATCH);
 264                         pattern += pclen;
 265                         if (c2 == EOS)
 266                                 return (RANGE_ERROR);
 267
 268                         if (flags & FNM_CASEFOLD)
 269                                 c2 = towlower(c2);
 270
 271                         if (__collate_load_error ?
 272                             c <= test && test <= c2 :
 273                                __collate_range_cmp(c, test) <= 0
 274                             && __collate_range_cmp(test, c2) <= 0
 275                            )
 276                                 ok = 1;
 277                 } else if (c == test) {
 278                         ok = 1;
 279                 }
 280         }
 281
 282         *newp = __DECONST(char *, pattern);
 283         return (ok == negate ? RANGE_NOMATCH : RANGE_MATCH);
 284 }