gen/FreeBSD/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  * 3. All advertising materials mentioning features or use of this software
  17  *    must display the following acknowledgement:
  18  *      This product includes software developed by the University of
  19  *      California, Berkeley and its contributors.
  20  * 4. Neither the name of the University nor the names of its contributors
  21  *    may be used to endorse or promote products derived from this software
  22  *    without specific prior written permission.
  23  *
  24  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
  25  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
  26  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
  27  * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
  28  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
  29  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
  30  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
  31  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
  32  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
  33  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
  34  * SUCH DAMAGE.
  35  */
  36
  37 #if defined(LIBC_SCCS) && !defined(lint)
  38 static char sccsid[] = "@(#)fnmatch.c   8.2 (Berkeley) 4/16/94";
  39 #endif /* LIBC_SCCS and not lint */
  40 #include <sys/cdefs.h>
  41 __FBSDID("$FreeBSD: src/lib/libc/gen/fnmatch.c,v 1.16 2004/07/29 03:13:10 tjr Exp $");
  42
  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 #include <fnmatch.h>
  60 #include <limits.h>
  61 #include <string.h>
  62 #include <wchar.h>
  63 #include <wctype.h>
  64
  65 #include "collate.h"
  66
  67 #define EOS     '\0'
  68
  69 #define RANGE_MATCH     1
  70 #define RANGE_NOMATCH   0
  71 #define RANGE_ERROR     (-1)
  72
  73 static int rangematch(const char *, wchar_t, int, char **, mbstate_t *);
  74 static int fnmatch1(const char *, const char *, int, mbstate_t, mbstate_t);
  75
  76 int
  77 fnmatch(pattern, string, flags)
  78         const char *pattern, *string;
  79         int flags;
  80 {
  81         static const mbstate_t initial;
  82
  83         return (fnmatch1(pattern, string, flags, initial, initial));
  84 }
  85
  86 static int
  87 fnmatch1(pattern, string, flags, patmbs, strmbs)
  88         const char *pattern, *string;
  89         int flags;
  90         mbstate_t patmbs, strmbs;
  91 {
  92         const char *stringstart;
  93         char *newp;
  94         char c;
  95         wchar_t pc, sc;
  96         size_t pclen, sclen;
  97
  98         for (stringstart = string;;) {
  99                 pclen = mbrtowc(&pc, pattern, MB_LEN_MAX, &patmbs);
 100                 if (pclen == (size_t)-1 || pclen == (size_t)-2)
 101                         return (FNM_NOMATCH);
 102                 pattern += pclen;
 103                 sclen = mbrtowc(&sc, string, MB_LEN_MAX, &strmbs);
 104                 if (sclen == (size_t)-1 || sclen == (size_t)-2) {
 105                         sc = (unsigned char)*string;
 106                         sclen = 1;
 107                         memset(&strmbs, 0, sizeof(strmbs));
 108                 }
 109                 switch (pc) {
 110                 case EOS:
 111                         if ((flags & FNM_LEADING_DIR) && sc == '/')
 112                                 return (0);
 113                         return (sc == EOS ? 0 : FNM_NOMATCH);
 114                 case '?':
 115                         if (sc == EOS)
 116                                 return (FNM_NOMATCH);
 117                         if (sc == '/' && (flags & FNM_PATHNAME))
 118                                 return (FNM_NOMATCH);
 119                         if (sc == '.' && (flags & FNM_PERIOD) &&
 120                             (string == stringstart ||
 121                             ((flags & FNM_PATHNAME) && *(string - 1) == '/')))
 122                                 return (FNM_NOMATCH);
 123                         string += sclen;
 124                         break;
 125                 case '*':
 126                         c = *pattern;
 127                         /* Collapse multiple stars. */
 128                         while (c == '*')
 129                                 c = *++pattern;
 130
 131                         if (sc == '.' && (flags & FNM_PERIOD) &&
 132                             (string == stringstart ||
 133                             ((flags & FNM_PATHNAME) && *(string - 1) == '/')))
 134                                 return (FNM_NOMATCH);
 135
 136                         /* Optimize for pattern with * at end or before /. */
 137                         if (c == EOS)
 138                                 if (flags & FNM_PATHNAME)
 139                                         return ((flags & FNM_LEADING_DIR) ||
 140                                             strchr(string, '/') == NULL ?
 141                                             0 : FNM_NOMATCH);
 142                                 else
 143                                         return (0);
 144                         else if (c == '/' && flags & FNM_PATHNAME) {
 145                                 if ((string = strchr(string, '/')) == NULL)
 146                                         return (FNM_NOMATCH);
 147                                 break;
 148                         }
 149
 150                         /* General case, use recursion. */
 151                         while (sc != EOS) {
 152                                 if (!fnmatch1(pattern, string,
 153                                     flags & ~FNM_PERIOD, patmbs, strmbs))
 154                                         return (0);
 155                                 sclen = mbrtowc(&sc, string, MB_LEN_MAX,
 156                                     &strmbs);
 157                                 if (sclen == (size_t)-1 ||
 158                                     sclen == (size_t)-2) {
 159                                         sc = (unsigned char)*string;
 160                                         sclen = 1;
 161                                         memset(&strmbs, 0, sizeof(strmbs));
 162                                 }
 163                                 if (sc == '/' && flags & FNM_PATHNAME)
 164                                         break;
 165                                 string += sclen;
 166                         }
 167                         return (FNM_NOMATCH);
 168                 case '[':
 169                         if (sc == EOS)
 170                                 return (FNM_NOMATCH);
 171                         if (sc == '/' && (flags & FNM_PATHNAME))
 172                                 return (FNM_NOMATCH);
 173                         if (sc == '.' && (flags & FNM_PERIOD) &&
 174                             (string == stringstart ||
 175                             ((flags & FNM_PATHNAME) && *(string - 1) == '/')))
 176                                 return (FNM_NOMATCH);
 177
 178                         switch (rangematch(pattern, sc, flags, &newp,
 179                             &patmbs)) {
 180                         case RANGE_ERROR:
 181                                 goto norm;
 182                         case RANGE_MATCH:
 183                                 pattern = newp;
 184                                 break;
 185                         case RANGE_NOMATCH:
 186                                 return (FNM_NOMATCH);
 187                         }
 188                         string += sclen;
 189                         break;
 190                 case '\\':
 191                         if (!(flags & FNM_NOESCAPE)) {
 192                                 pclen = mbrtowc(&pc, pattern, MB_LEN_MAX,
 193                                     &patmbs);
 194                                 if (pclen == (size_t)-1 || pclen == (size_t)-2)
 195                                         return (FNM_NOMATCH);
 196                                 if (pclen == 0)
 197                                         pc = '\\';
 198                                 pattern += pclen;
 199                         }
 200                         /* FALLTHROUGH */
 201                 default:
 202                 norm:
 203                         if (pc == sc)
 204                                 ;
 205                         else if ((flags & FNM_CASEFOLD) &&
 206                                  (towlower(pc) == towlower(sc)))
 207                                 ;
 208                         else
 209                                 return (FNM_NOMATCH);
 210                         string += sclen;
 211                         break;
 212                 }
 213         }
 214         /* NOTREACHED */
 215 }
 216
 217 static int
 218 rangematch(pattern, test, flags, newp, patmbs)
 219         const char *pattern;
 220         wchar_t test;
 221         int flags;
 222         char **newp;
 223         mbstate_t *patmbs;
 224 {
 225         int negate, ok;
 226         wchar_t c, c2;
 227         size_t pclen;
 228         const char *origpat;
 229
 230         /*
 231          * A bracket expression starting with an unquoted circumflex
 232          * character produces unspecified results (IEEE 1003.2-1992,
 233          * 3.13.2).  This implementation treats it like '!', for
 234          * consistency with the regular expression syntax.
 235          * J.T. Conklin (conklin@ngai.kaleida.com)
 236          */
 237         if ( (negate = (*pattern == '!' || *pattern == '^')) )
 238                 ++pattern;
 239
 240         if (flags & FNM_CASEFOLD)
 241                 test = towlower(test);
 242
 243         /*
 244          * A right bracket shall lose its special meaning and represent
 245          * itself in a bracket expression if it occurs first in the list.
 246          * -- POSIX.2 2.8.3.2
 247          */
 248         ok = 0;
 249         origpat = pattern;
 250         for (;;) {
 251                 if (*pattern == ']' && pattern > origpat) {
 252                         pattern++;
 253                         break;
 254                 } else if (*pattern == '\0') {
 255                         return (RANGE_ERROR);
 256                 } else if (*pattern == '/' && (flags & FNM_PATHNAME)) {
 257                         pattern++;
 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 (__collate_load_error ?
 285                             c <= test && test <= c2 :
 286                                __collate_range_cmp(c, test) <= 0
 287                             && __collate_range_cmp(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 }