contrib/libarchive/cpio/test/test_pathmatch.c

   1 /*-
   2  * Copyright (c) 2003-2007 Tim Kientzle
   3  * All rights reserved.
   4  *
   5  * Redistribution and use in source and binary forms, with or without
   6  * modification, are permitted provided that the following conditions
   7  * are met:
   8  * 1. Redistributions of source code must retain the above copyright
   9  *    notice, this list of conditions and the following disclaimer.
  10  * 2. Redistributions in binary form must reproduce the above copyright
  11  *    notice, this list of conditions and the following disclaimer in the
  12  *    documentation and/or other materials provided with the distribution.
  13  *
  14  * THIS SOFTWARE IS PROVIDED BY THE AUTHOR(S) ``AS IS'' AND ANY EXPRESS OR
  15  * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
  16  * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
  17  * IN NO EVENT SHALL THE AUTHOR(S) BE LIABLE FOR ANY DIRECT, INDIRECT,
  18  * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
  19  * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
  20  * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
  21  * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
  22  * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
  23  * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
  24  */
  25 #include "test.h"
  26 __FBSDID("$FreeBSD$");
  27
  28 #include "pathmatch.h"
  29
  30 /*
  31  * Verify that the pattern matcher implements the wildcard logic specified
  32  * in SUSv2 for the cpio command.  This is essentially the
  33  * shell glob syntax:
  34  *   * - matches any sequence of chars, including '/'
  35  *   ? - matches any single char, including '/'
  36  *   [...] - matches any of a set of chars, '-' specifies a range,
  37  *        initial '!' is undefined
  38  *
  39  * The specification in SUSv2 is a bit incomplete, I assume the following:
  40  *   Trailing '-' in [...] is not special.
  41  *
  42  * TODO: Figure out if there's a good way to extend this to handle
  43  * Windows paths that use '\' as a path separator.  <sigh>
  44  */
  45
  46 DEFINE_TEST(test_pathmatch)
  47 {
  48         assertEqualInt(1, lafe_pathmatch("a/b/c", "a/b/c", 0));
  49         assertEqualInt(0, lafe_pathmatch("a/b/", "a/b/c", 0));
  50         assertEqualInt(0, lafe_pathmatch("a/b", "a/b/c", 0));
  51         assertEqualInt(0, lafe_pathmatch("a/b/c", "a/b/", 0));
  52         assertEqualInt(0, lafe_pathmatch("a/b/c", "a/b", 0));
  53
  54         /* Empty pattern only matches empty string. */
  55         assertEqualInt(1, lafe_pathmatch("","", 0));
  56         assertEqualInt(0, lafe_pathmatch("","a", 0));
  57         assertEqualInt(1, lafe_pathmatch("*","", 0));
  58         assertEqualInt(1, lafe_pathmatch("*","a", 0));
  59         assertEqualInt(1, lafe_pathmatch("*","abcd", 0));
  60         /* SUSv2: * matches / */
  61         assertEqualInt(1, lafe_pathmatch("*","abcd/efgh/ijkl", 0));
  62         assertEqualInt(1, lafe_pathmatch("abcd*efgh/ijkl","abcd/efgh/ijkl", 0));
  63         assertEqualInt(1, lafe_pathmatch("abcd***efgh/ijkl","abcd/efgh/ijkl", 0));
  64         assertEqualInt(1, lafe_pathmatch("abcd***/efgh/ijkl","abcd/efgh/ijkl", 0));
  65         assertEqualInt(0, lafe_pathmatch("?", "", 0));
  66         assertEqualInt(0, lafe_pathmatch("?", "\0", 0));
  67         assertEqualInt(1, lafe_pathmatch("?", "a", 0));
  68         assertEqualInt(0, lafe_pathmatch("?", "ab", 0));
  69         assertEqualInt(1, lafe_pathmatch("?", ".", 0));
  70         assertEqualInt(1, lafe_pathmatch("?", "?", 0));
  71         assertEqualInt(1, lafe_pathmatch("a", "a", 0));
  72         assertEqualInt(0, lafe_pathmatch("a", "ab", 0));
  73         assertEqualInt(0, lafe_pathmatch("a", "ab", 0));
  74         assertEqualInt(1, lafe_pathmatch("a?c", "abc", 0));
  75         /* SUSv2: ? matches / */
  76         assertEqualInt(1, lafe_pathmatch("a?c", "a/c", 0));
  77         assertEqualInt(1, lafe_pathmatch("a?*c*", "a/c", 0));
  78         assertEqualInt(1, lafe_pathmatch("*a*", "a/c", 0));
  79         assertEqualInt(1, lafe_pathmatch("*a*", "/a/c", 0));
  80         assertEqualInt(1, lafe_pathmatch("*a*", "defaaaaaaa", 0));
  81         assertEqualInt(0, lafe_pathmatch("a*", "defghi", 0));
  82         assertEqualInt(0, lafe_pathmatch("*a*", "defghi", 0));
  83
  84         /* Character classes */
  85         assertEqualInt(1, lafe_pathmatch("abc[def", "abc[def", 0));
  86         assertEqualInt(0, lafe_pathmatch("abc[def]", "abc[def", 0));
  87         assertEqualInt(0, lafe_pathmatch("abc[def", "abcd", 0));
  88         assertEqualInt(1, lafe_pathmatch("abc[def]", "abcd", 0));
  89         assertEqualInt(1, lafe_pathmatch("abc[def]", "abce", 0));
  90         assertEqualInt(1, lafe_pathmatch("abc[def]", "abcf", 0));
  91         assertEqualInt(0, lafe_pathmatch("abc[def]", "abcg", 0));
  92         assertEqualInt(1, lafe_pathmatch("abc[d*f]", "abcd", 0));
  93         assertEqualInt(1, lafe_pathmatch("abc[d*f]", "abc*", 0));
  94         assertEqualInt(0, lafe_pathmatch("abc[d*f]", "abcdefghi", 0));
  95         assertEqualInt(0, lafe_pathmatch("abc[d*", "abcdefghi", 0));
  96         assertEqualInt(1, lafe_pathmatch("abc[d*", "abc[defghi", 0));
  97         assertEqualInt(1, lafe_pathmatch("abc[d-f]", "abcd", 0));
  98         assertEqualInt(1, lafe_pathmatch("abc[d-f]", "abce", 0));
  99         assertEqualInt(1, lafe_pathmatch("abc[d-f]", "abcf", 0));
 100         assertEqualInt(0, lafe_pathmatch("abc[d-f]", "abcg", 0));
 101         assertEqualInt(0, lafe_pathmatch("abc[d-fh-k]", "abca", 0));
 102         assertEqualInt(1, lafe_pathmatch("abc[d-fh-k]", "abcd", 0));
 103         assertEqualInt(1, lafe_pathmatch("abc[d-fh-k]", "abce", 0));
 104         assertEqualInt(1, lafe_pathmatch("abc[d-fh-k]", "abcf", 0));
 105         assertEqualInt(0, lafe_pathmatch("abc[d-fh-k]", "abcg", 0));
 106         assertEqualInt(1, lafe_pathmatch("abc[d-fh-k]", "abch", 0));
 107         assertEqualInt(1, lafe_pathmatch("abc[d-fh-k]", "abci", 0));
 108         assertEqualInt(1, lafe_pathmatch("abc[d-fh-k]", "abcj", 0));
 109         assertEqualInt(1, lafe_pathmatch("abc[d-fh-k]", "abck", 0));
 110         assertEqualInt(0, lafe_pathmatch("abc[d-fh-k]", "abcl", 0));
 111         assertEqualInt(0, lafe_pathmatch("abc[d-fh-k]", "abc-", 0));
 112
 113         /* [] matches nothing, [!] is the same as ? */
 114         assertEqualInt(0, lafe_pathmatch("abc[]efg", "abcdefg", 0));
 115         assertEqualInt(0, lafe_pathmatch("abc[]efg", "abcqefg", 0));
 116         assertEqualInt(0, lafe_pathmatch("abc[]efg", "abcefg", 0));
 117         assertEqualInt(1, lafe_pathmatch("abc[!]efg", "abcdefg", 0));
 118         assertEqualInt(1, lafe_pathmatch("abc[!]efg", "abcqefg", 0));
 119         assertEqualInt(0, lafe_pathmatch("abc[!]efg", "abcefg", 0));
 120
 121         /* I assume: Trailing '-' is non-special. */
 122         assertEqualInt(0, lafe_pathmatch("abc[d-fh-]", "abcl", 0));
 123         assertEqualInt(1, lafe_pathmatch("abc[d-fh-]", "abch", 0));
 124         assertEqualInt(1, lafe_pathmatch("abc[d-fh-]", "abc-", 0));
 125         assertEqualInt(1, lafe_pathmatch("abc[d-fh-]", "abc-", 0));
 126
 127         /* ']' can be backslash-quoted within a character class. */
 128         assertEqualInt(1, lafe_pathmatch("abc[\\]]", "abc]", 0));
 129         assertEqualInt(1, lafe_pathmatch("abc[\\]d]", "abc]", 0));
 130         assertEqualInt(1, lafe_pathmatch("abc[\\]d]", "abcd", 0));
 131         assertEqualInt(1, lafe_pathmatch("abc[d\\]]", "abc]", 0));
 132         assertEqualInt(1, lafe_pathmatch("abc[d\\]]", "abcd", 0));
 133         assertEqualInt(1, lafe_pathmatch("abc[d]e]", "abcde]", 0));
 134         assertEqualInt(1, lafe_pathmatch("abc[d\\]e]", "abc]", 0));
 135         assertEqualInt(0, lafe_pathmatch("abc[d\\]e]", "abcd]e", 0));
 136         assertEqualInt(0, lafe_pathmatch("abc[d]e]", "abc]", 0));
 137
 138         /* backslash-quoted chars can appear as either end of a range. */
 139         assertEqualInt(1, lafe_pathmatch("abc[\\d-f]gh", "abcegh", 0));
 140         assertEqualInt(0, lafe_pathmatch("abc[\\d-f]gh", "abcggh", 0));
 141         assertEqualInt(0, lafe_pathmatch("abc[\\d-f]gh", "abc\\gh", 0));
 142         assertEqualInt(1, lafe_pathmatch("abc[d-\\f]gh", "abcegh", 0));
 143         assertEqualInt(1, lafe_pathmatch("abc[\\d-\\f]gh", "abcegh", 0));
 144         assertEqualInt(1, lafe_pathmatch("abc[\\d-\\f]gh", "abcegh", 0));
 145         /* backslash-quoted '-' isn't special. */
 146         assertEqualInt(0, lafe_pathmatch("abc[d\\-f]gh", "abcegh", 0));
 147         assertEqualInt(1, lafe_pathmatch("abc[d\\-f]gh", "abc-gh", 0));
 148
 149         /* Leading '!' negates a character class. */
 150         assertEqualInt(0, lafe_pathmatch("abc[!d]", "abcd", 0));
 151         assertEqualInt(1, lafe_pathmatch("abc[!d]", "abce", 0));
 152         assertEqualInt(1, lafe_pathmatch("abc[!d]", "abcc", 0));
 153         assertEqualInt(0, lafe_pathmatch("abc[!d-z]", "abcq", 0));
 154         assertEqualInt(1, lafe_pathmatch("abc[!d-gi-z]", "abch", 0));
 155         assertEqualInt(1, lafe_pathmatch("abc[!fgijkl]", "abch", 0));
 156         assertEqualInt(0, lafe_pathmatch("abc[!fghijkl]", "abch", 0));
 157
 158         /* Backslash quotes next character. */
 159         assertEqualInt(0, lafe_pathmatch("abc\\[def]", "abc\\d", 0));
 160         assertEqualInt(1, lafe_pathmatch("abc\\[def]", "abc[def]", 0));
 161         assertEqualInt(0, lafe_pathmatch("abc\\\\[def]", "abc[def]", 0));
 162         assertEqualInt(0, lafe_pathmatch("abc\\\\[def]", "abc\\[def]", 0));
 163         assertEqualInt(1, lafe_pathmatch("abc\\\\[def]", "abc\\d", 0));
 164         assertEqualInt(1, lafe_pathmatch("abcd\\", "abcd\\", 0));
 165         assertEqualInt(0, lafe_pathmatch("abcd\\", "abcd\\[", 0));
 166         assertEqualInt(0, lafe_pathmatch("abcd\\", "abcde", 0));
 167         assertEqualInt(0, lafe_pathmatch("abcd\\[", "abcd\\", 0));
 168
 169         /*
 170          * Because '.' and '/' have special meanings, we can
 171          * identify many equivalent paths even if they're expressed
 172          * differently.  (But quoting a character with '\\' suppresses
 173          * special meanings!)
 174          */
 175         assertEqualInt(0, lafe_pathmatch("a/b/", "a/bc", 0));
 176         assertEqualInt(1, lafe_pathmatch("a/./b", "a/b", 0));
 177         assertEqualInt(0, lafe_pathmatch("a\\/./b", "a/b", 0));
 178         assertEqualInt(0, lafe_pathmatch("a/\\./b", "a/b", 0));
 179         assertEqualInt(0, lafe_pathmatch("a/.\\/b", "a/b", 0));
 180         assertEqualInt(0, lafe_pathmatch("a\\/\\.\\/b", "a/b", 0));
 181         assertEqualInt(1, lafe_pathmatch("./abc/./def/", "abc/def/", 0));
 182         assertEqualInt(1, lafe_pathmatch("abc/def", "./././abc/./def", 0));
 183         assertEqualInt(1, lafe_pathmatch("abc/def/././//", "./././abc/./def/", 0));
 184         assertEqualInt(1, lafe_pathmatch(".////abc/.//def", "./././abc/./def", 0));
 185         assertEqualInt(1, lafe_pathmatch("./abc?def/", "abc/def/", 0));
 186         failure("\"?./\" is not the same as \"/./\"");
 187         assertEqualInt(0, lafe_pathmatch("./abc?./def/", "abc/def/", 0));
 188         failure("Trailing '/' should match no trailing '/'");
 189         assertEqualInt(1, lafe_pathmatch("./abc/./def/", "abc/def", 0));
 190         failure("Trailing '/./' is still the same directory.");
 191         assertEqualInt(1, lafe_pathmatch("./abc/./def/./", "abc/def", 0));
 192         failure("Trailing '/.' is still the same directory.");
 193         assertEqualInt(1, lafe_pathmatch("./abc/./def/.", "abc/def", 0));
 194         assertEqualInt(1, lafe_pathmatch("./abc/./def", "abc/def/", 0));
 195         failure("Trailing '/./' is still the same directory.");
 196         assertEqualInt(1, lafe_pathmatch("./abc/./def", "abc/def/./", 0));
 197         failure("Trailing '/.' is still the same directory.");
 198         assertEqualInt(1, lafe_pathmatch("./abc*/./def", "abc/def/.", 0));
 199
 200         /* Matches not anchored at beginning. */
 201         assertEqualInt(0,
 202             lafe_pathmatch("bcd", "abcd", PATHMATCH_NO_ANCHOR_START));
 203         assertEqualInt(1,
 204             lafe_pathmatch("abcd", "abcd", PATHMATCH_NO_ANCHOR_START));
 205         assertEqualInt(0,
 206             lafe_pathmatch("^bcd", "abcd", PATHMATCH_NO_ANCHOR_START));
 207         assertEqualInt(1,
 208             lafe_pathmatch("b/c/d", "a/b/c/d", PATHMATCH_NO_ANCHOR_START));
 209         assertEqualInt(0,
 210             lafe_pathmatch("b/c", "a/b/c/d", PATHMATCH_NO_ANCHOR_START));
 211         assertEqualInt(0,
 212             lafe_pathmatch("^b/c", "a/b/c/d", PATHMATCH_NO_ANCHOR_START));
 213
 214         /* Matches not anchored at end. */
 215         assertEqualInt(0,
 216             lafe_pathmatch("bcd", "abcd", PATHMATCH_NO_ANCHOR_END));
 217         assertEqualInt(1,
 218             lafe_pathmatch("abcd", "abcd", PATHMATCH_NO_ANCHOR_END));
 219         assertEqualInt(1,
 220             lafe_pathmatch("abcd", "abcd/", PATHMATCH_NO_ANCHOR_END));
 221         assertEqualInt(1,
 222             lafe_pathmatch("abcd", "abcd/.", PATHMATCH_NO_ANCHOR_END));
 223         assertEqualInt(0,
 224             lafe_pathmatch("abc", "abcd", PATHMATCH_NO_ANCHOR_END));
 225         assertEqualInt(1,
 226             lafe_pathmatch("a/b/c", "a/b/c/d", PATHMATCH_NO_ANCHOR_END));
 227         assertEqualInt(0,
 228             lafe_pathmatch("a/b/c$", "a/b/c/d", PATHMATCH_NO_ANCHOR_END));
 229         assertEqualInt(1,
 230             lafe_pathmatch("a/b/c$", "a/b/c", PATHMATCH_NO_ANCHOR_END));
 231         assertEqualInt(1,
 232             lafe_pathmatch("a/b/c$", "a/b/c/", PATHMATCH_NO_ANCHOR_END));
 233         assertEqualInt(1,
 234             lafe_pathmatch("a/b/c/", "a/b/c/d", PATHMATCH_NO_ANCHOR_END));
 235         assertEqualInt(0,
 236             lafe_pathmatch("a/b/c/$", "a/b/c/d", PATHMATCH_NO_ANCHOR_END));
 237         assertEqualInt(1,
 238             lafe_pathmatch("a/b/c/$", "a/b/c/", PATHMATCH_NO_ANCHOR_END));
 239         assertEqualInt(1,
 240             lafe_pathmatch("a/b/c/$", "a/b/c", PATHMATCH_NO_ANCHOR_END));
 241         assertEqualInt(0,
 242             lafe_pathmatch("b/c", "a/b/c/d", PATHMATCH_NO_ANCHOR_END));
 243 }