lib/util/ms_fnmatch.c

   1 /*
   2    Unix SMB/CIFS implementation.
   3    filename matching routine
   4    Copyright (C) Andrew Tridgell 1992-2004
   5
   6    This program is free software; you can redistribute it and/or modify
   7    it under the terms of the GNU General Public License as published by
   8    the Free Software Foundation; either version 3 of the License, or
   9    (at your option) any later version.
  10
  11    This program is distributed in the hope that it will be useful,
  12    but WITHOUT ANY WARRANTY; without even the implied warranty of
  13    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  14    GNU General Public License for more details.
  15
  16    You should have received a copy of the GNU General Public License
  17    along with this program.  If not, see <http://www.gnu.org/licenses/>.
  18 */
  19
  20 /*
  21    This module was originally based on fnmatch.c copyright by the Free
  22    Software Foundation. It bears little (if any) resemblence to that
  23    code now
  24 */
  25
  26 /**
  27  * @file
  28  * @brief MS-style Filename matching
  29  */
  30
  31 #include "includes.h"
  32 #include "libcli/smb/smb_constants.h"
  33
  34 static int null_match(const char *p)
  35 {
  36         for (;*p;p++) {
  37                 if (*p != '*' &&
  38                     *p != '<' &&
  39                     *p != '"' &&
  40                     *p != '>') return -1;
  41         }
  42         return 0;
  43 }
  44
  45 /*
  46   the max_n structure is purely for efficiency, it doesn't contribute
  47   to the matching algorithm except by ensuring that the algorithm does
  48   not grow exponentially
  49 */
  50 struct max_n {
  51         const char *predot;
  52         const char *postdot;
  53 };
  54
  55
  56 /*
  57   p and n are the pattern and string being matched. The max_n array is
  58   an optimisation only. The ldot pointer is NULL if the string does
  59   not contain a '.', otherwise it points at the last dot in 'n'.
  60 */
  61 static int ms_fnmatch_core(const char *p, const char *n,
  62                            struct max_n *max_n, const char *ldot,
  63                            bool is_case_sensitive)
  64 {
  65         codepoint_t c, c2;
  66         int i;
  67         size_t size, size_n;
  68
  69         while ((c = next_codepoint(p, &size))) {
  70                 p += size;
  71
  72                 switch (c) {
  73                 case '*':
  74                         /* a '*' matches zero or more characters of any type */
  75                         if (max_n->predot && max_n->predot <= n) {
  76                                 return null_match(p);
  77                         }
  78                         for (i=0; n[i]; i += size_n) {
  79                                 next_codepoint(n+i, &size_n);
  80                                 if (ms_fnmatch_core(p, n+i, max_n+1, ldot, is_case_sensitive) == 0) {
  81                                         return 0;
  82                                 }
  83                         }
  84                         if (!max_n->predot || max_n->predot > n) max_n->predot = n;
  85                         return null_match(p);
  86
  87                 case '<':
  88                         /* a '<' matches zero or more characters of
  89                            any type, but stops matching at the last
  90                            '.' in the string. */
  91                         if (max_n->predot && max_n->predot <= n) {
  92                                 return null_match(p);
  93                         }
  94                         if (max_n->postdot && max_n->postdot <= n && n <= ldot) {
  95                                 return -1;
  96                         }
  97                         for (i=0; n[i]; i += size_n) {
  98                                 next_codepoint(n+i, &size_n);
  99                                 if (ms_fnmatch_core(p, n+i, max_n+1, ldot, is_case_sensitive) == 0) return 0;
 100                                 if (n+i == ldot) {
 101                                         if (ms_fnmatch_core(p, n+i+size_n, max_n+1, ldot, is_case_sensitive) == 0) return 0;
 102                                         if (!max_n->postdot || max_n->postdot > n) max_n->postdot = n;
 103                                         return -1;
 104                                 }
 105                         }
 106                         if (!max_n->predot || max_n->predot > n) max_n->predot = n;
 107                         return null_match(p);
 108
 109                 case '?':
 110                         /* a '?' matches any single character */
 111                         if (! *n) {
 112                                 return -1;
 113                         }
 114                         next_codepoint(n, &size_n);
 115                         n += size_n;
 116                         break;
 117
 118                 case '>':
 119                         /* a '?' matches any single character, but
 120                            treats '.' specially */
 121                         if (n[0] == '.') {
 122                                 if (! n[1] && null_match(p) == 0) {
 123                                         return 0;
 124                                 }
 125                                 break;
 126                         }
 127                         if (! *n) return null_match(p);
 128                         next_codepoint(n, &size_n);
 129                         n += size_n;
 130                         break;
 131
 132                 case '"':
 133                         /* a bit like a soft '.' */
 134                         if (*n == 0 && null_match(p) == 0) {
 135                                 return 0;
 136                         }
 137                         if (*n != '.') return -1;
 138                         next_codepoint(n, &size_n);
 139                         n += size_n;
 140                         break;
 141
 142                 default:
 143                         c2 = next_codepoint(n, &size_n);
 144                         if (c != c2) {
 145                                 if (is_case_sensitive) {
 146                                         return -1;
 147                                 }
 148                                 if (codepoint_cmpi(c, c2) != 0) {
 149                                         return -1;
 150                                 }
 151                         }
 152                         n += size_n;
 153                         break;
 154                 }
 155         }
 156
 157         if (! *n) {
 158                 return 0;
 159         }
 160
 161         return -1;
 162 }
 163
 164 int ms_fnmatch_protocol(const char *pattern, const char *string, int protocol,
 165                         bool is_case_sensitive)
 166 {
 167         int ret = -1;
 168         size_t count, i;
 169
 170         if (strcmp(string, "..") == 0) {
 171                 string = ".";
 172         }
 173
 174         if (strpbrk(pattern, "<>*?\"") == NULL) {
 175                 /* this is not just an optimisation - it is essential
 176                    for LANMAN1 correctness */
 177                 return strcasecmp_m(pattern, string);
 178         }
 179
 180         if (protocol <= PROTOCOL_LANMAN2) {
 181                 char *p = talloc_strdup(NULL, pattern);
 182                 if (p == NULL) {
 183                         return -1;
 184                 }
 185                 /*
 186                   for older negotiated protocols it is possible to
 187                   translate the pattern to produce a "new style"
 188                   pattern that exactly matches w2k behaviour
 189                 */
 190                 for (i=0;p[i];i++) {
 191                         if (p[i] == '?') {
 192                                 p[i] = '>';
 193                         } else if (p[i] == '.' &&
 194                                    (p[i+1] == '?' ||
 195                                     p[i+1] == '*' ||
 196                                     p[i+1] == 0)) {
 197                                 p[i] = '"';
 198                         } else if (p[i] == '*' &&
 199                                    p[i+1] == '.') {
 200                                 p[i] = '<';
 201                         }
 202                 }
 203                 ret = ms_fnmatch_protocol(p, string, PROTOCOL_NT1,
 204                                           is_case_sensitive);
 205                 talloc_free(p);
 206                 return ret;
 207         }
 208
 209         for (count=i=0;pattern[i];i++) {
 210                 if (pattern[i] == '*' || pattern[i] == '<') count++;
 211         }
 212
 213         /* If the pattern includes '*' or '<' */
 214         if (count >= 1) {
 215                 struct max_n max_n[count];
 216
 217                 memset(max_n, 0, sizeof(struct max_n) * count);
 218
 219                 ret = ms_fnmatch_core(pattern, string, max_n, strrchr(string, '.'),
 220                                       is_case_sensitive);
 221         } else {
 222                 ret = ms_fnmatch_core(pattern, string, NULL, strrchr(string, '.'),
 223                                       is_case_sensitive);
 224         }
 225
 226         return ret;
 227 }
 228
 229
 230 /** a generic fnmatch function - uses for non-CIFS pattern matching */
 231 int gen_fnmatch(const char *pattern, const char *string)
 232 {
 233         return ms_fnmatch_protocol(pattern, string, PROTOCOL_NT1, false);
 234 }