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) resemblance to that
  23    code now
  24 */
  25
  26 /**
  27  * @file
  28  * @brief MS-style Filename matching
  29  */
  30
  31 #include "replace.h"
  32 #include "lib/util/samba_util.h"
  33 #include "libcli/smb/smb_constants.h"
  34
  35 static int null_match(const char *p)
  36 {
  37         for (;*p;p++) {
  38                 if (*p != '*' &&
  39                     *p != '<' &&
  40                     *p != '"' &&
  41                     *p != '>') return -1;
  42         }
  43         return 0;
  44 }
  45
  46 /*
  47   the max_n structure is purely for efficiency, it doesn't contribute
  48   to the matching algorithm except by ensuring that the algorithm does
  49   not grow exponentially
  50 */
  51 struct max_n {
  52         const char *predot;
  53         const char *postdot;
  54 };
  55
  56
  57 /*
  58   p and n are the pattern and string being matched. The max_n array is
  59   an optimisation only. The ldot pointer is NULL if the string does
  60   not contain a '.', otherwise it points at the last dot in 'n'.
  61 */
  62 static int ms_fnmatch_core(const char *p, const char *n,
  63                            struct max_n *max_n, const char *ldot,
  64                            bool is_case_sensitive)
  65 {
  66         codepoint_t c, c2;
  67         int i;
  68         size_t size, size_n;
  69
  70         while ((c = next_codepoint(p, &size))) {
  71                 p += size;
  72
  73                 switch (c) {
  74                 case '*':
  75                         /* a '*' matches zero or more characters of any type */
  76                         if (max_n != NULL && max_n->predot &&
  77                             max_n->predot <= n) {
  78                                 return null_match(p);
  79                         }
  80                         for (i=0; n[i]; i += size_n) {
  81                                 next_codepoint(n+i, &size_n);
  82                                 if (ms_fnmatch_core(p, n+i, max_n+1, ldot, is_case_sensitive) == 0) {
  83                                         return 0;
  84                                 }
  85                         }
  86                         if (max_n != NULL && (!max_n->predot ||
  87                             max_n->predot > n)) {
  88                                 max_n->predot = n;
  89                         }
  90                         return null_match(p);
  91
  92                 case '<':
  93                         /* a '<' matches zero or more characters of
  94                            any type, but stops matching at the last
  95                            '.' in the string. */
  96                         if (max_n != NULL && max_n->predot &&
  97                             max_n->predot <= n) {
  98                                 return null_match(p);
  99                         }
 100                         if (max_n != NULL && max_n->postdot &&
 101                             max_n->postdot <= n && n <= ldot) {
 102                                 return -1;
 103                         }
 104                         for (i=0; n[i]; i += size_n) {
 105                                 next_codepoint(n+i, &size_n);
 106                                 if (ms_fnmatch_core(p, n+i, max_n+1, ldot, is_case_sensitive) == 0) return 0;
 107                                 if (n+i == ldot) {
 108                                         if (ms_fnmatch_core(p, n+i+size_n, max_n+1, ldot, is_case_sensitive) == 0) return 0;
 109                                         if (max_n != NULL) {
 110                                                 if (!max_n->postdot ||
 111                                                     max_n->postdot > n) {
 112                                                         max_n->postdot = n;
 113                                                 }
 114                                         }
 115                                         return -1;
 116                                 }
 117                         }
 118                         if (max_n != NULL && (!max_n->predot ||
 119                             max_n->predot > n)) {
 120                                 max_n->predot = n;
 121                         }
 122                         return null_match(p);
 123
 124                 case '?':
 125                         /* a '?' matches any single character */
 126                         if (! *n) {
 127                                 return -1;
 128                         }
 129                         next_codepoint(n, &size_n);
 130                         n += size_n;
 131                         break;
 132
 133                 case '>':
 134                         /* a '?' matches any single character, but
 135                            treats '.' specially */
 136                         if (n[0] == '.') {
 137                                 if (! n[1] && null_match(p) == 0) {
 138                                         return 0;
 139                                 }
 140                                 break;
 141                         }
 142                         if (! *n) return null_match(p);
 143                         next_codepoint(n, &size_n);
 144                         n += size_n;
 145                         break;
 146
 147                 case '"':
 148                         /* a bit like a soft '.' */
 149                         if (*n == 0 && null_match(p) == 0) {
 150                                 return 0;
 151                         }
 152                         if (*n != '.') return -1;
 153                         next_codepoint(n, &size_n);
 154                         n += size_n;
 155                         break;
 156
 157                 default:
 158                         c2 = next_codepoint(n, &size_n);
 159                         if (c != c2) {
 160                                 if (is_case_sensitive) {
 161                                         return -1;
 162                                 }
 163                                 if (codepoint_cmpi(c, c2) != 0) {
 164                                         return -1;
 165                                 }
 166                         }
 167                         n += size_n;
 168                         break;
 169                 }
 170         }
 171
 172         if (! *n) {
 173                 return 0;
 174         }
 175
 176         return -1;
 177 }
 178
 179 int ms_fnmatch_protocol(const char *pattern, const char *string, int protocol,
 180                         bool is_case_sensitive)
 181 {
 182         int ret = -1;
 183         size_t count, i;
 184
 185         if (strcmp(string, "..") == 0) {
 186                 string = ".";
 187         }
 188
 189         if (strpbrk(pattern, "<>*?\"") == NULL) {
 190                 /* this is not just an optimisation - it is essential
 191                    for LANMAN1 correctness */
 192                 return strcasecmp_m(pattern, string);
 193         }
 194
 195         if (protocol <= PROTOCOL_LANMAN2) {
 196                 char *p = talloc_strdup(NULL, pattern);
 197                 if (p == NULL) {
 198                         return -1;
 199                 }
 200                 /*
 201                   for older negotiated protocols it is possible to
 202                   translate the pattern to produce a "new style"
 203                   pattern that exactly matches w2k behaviour
 204                 */
 205                 for (i=0;p[i];i++) {
 206                         if (p[i] == '?') {
 207                                 p[i] = '>';
 208                         } else if (p[i] == '.' &&
 209                                    (p[i+1] == '?' ||
 210                                     p[i+1] == '*' ||
 211                                     p[i+1] == 0)) {
 212                                 p[i] = '"';
 213                         } else if (p[i] == '*' &&
 214                                    p[i+1] == '.') {
 215                                 p[i] = '<';
 216                         }
 217                 }
 218                 ret = ms_fnmatch_protocol(p, string, PROTOCOL_NT1,
 219                                           is_case_sensitive);
 220                 talloc_free(p);
 221                 return ret;
 222         }
 223
 224         for (count=i=0;pattern[i];i++) {
 225                 if (pattern[i] == '*' || pattern[i] == '<') count++;
 226         }
 227
 228         /* If the pattern includes '*' or '<' */
 229         if (count >= 1) {
 230                 struct max_n max_n[count];
 231
 232                 memset(max_n, 0, sizeof(struct max_n) * count);
 233
 234                 ret = ms_fnmatch_core(pattern, string, max_n, strrchr(string, '.'),
 235                                       is_case_sensitive);
 236         } else {
 237                 ret = ms_fnmatch_core(pattern, string, NULL, strrchr(string, '.'),
 238                                       is_case_sensitive);
 239         }
 240
 241         return ret;
 242 }
 243
 244
 245 /** a generic fnmatch function - uses for non-CIFS pattern matching */
 246 int gen_fnmatch(const char *pattern, const char *string)
 247 {
 248         return ms_fnmatch_protocol(pattern, string, PROTOCOL_NT1, false);
 249 }