| 1 | /* Copyright (C) 1991, 1992, 1993 Free Software Foundation, Inc. |
| 2 | * |
| 3 | * This library is free software; you can redistribute it and/or |
| 4 | * modify it under the terms of the GNU Lesser General Public |
| 5 | * License as published by the Free Software Foundation; either |
| 6 | * version 2 of the License, or (at your option) any later version. |
| 7 | * |
| 8 | * This library is distributed in the hope that it will be useful, |
| 9 | * but WITHOUT ANY WARRANTY; without even the implied warranty of |
| 10 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU |
| 11 | * Lesser General Public License for more details. |
| 12 | * |
| 13 | * You should have received a copy of the GNU Lesser General Public |
| 14 | * License along with this library; if not, write to the |
| 15 | * Free Software Foundation, Inc., 59 Temple Place - Suite 330, |
| 16 | * Boston, MA 02111-1307, USA. |
| 17 | */ |
| 18 | |
| 19 | /* |
| 20 | * Modified by the GTK+ Team and others 1997-2000. See the AUTHORS |
| 21 | * file for a list of people on the GTK+ Team. See the ChangeLog |
| 22 | * files for a list of changes. These files are distributed with |
| 23 | * GTK+ at ftp://ftp.gtk.org/pub/gtk/. |
| 24 | */ |
| 25 | |
| 26 | /* |
| 27 | * Stripped down, converted to UTF-8 and test cases added |
| 28 | * |
| 29 | * Owen Taylor, 13 December 2002; |
| 30 | */ |
| 31 | |
| 32 | #include <config.h> |
| 33 | #include <string.h> |
| 34 | |
| 35 | #include <glib.h> |
| 36 | |
| 37 | /* We need to make sure that all constants are defined |
| 38 | * to properly compile this file |
| 39 | */ |
| 40 | #ifndef _GNU_SOURCE |
| 41 | #define _GNU_SOURCE |
| 42 | #endif |
| 43 | |
| 44 | static gunichar |
| 45 | get_char (const char **str) |
| 46 | { |
| 47 | gunichar c = g_utf8_get_char (*str); |
| 48 | *str = g_utf8_next_char (*str); |
| 49 | |
| 50 | #ifdef G_PLATFORM_WIN32 |
| 51 | c = g_unichar_tolower (c); |
| 52 | #endif |
| 53 | |
| 54 | return c; |
| 55 | } |
| 56 | |
| 57 | #if defined(G_OS_WIN32) || defined(G_WITH_CYGWIN) |
| 58 | #define DO_ESCAPE 0 |
| 59 | #else |
| 60 | #define DO_ESCAPE 1 |
| 61 | #endif |
| 62 | |
| 63 | static gunichar |
| 64 | get_unescaped_char (const char **str, |
| 65 | gboolean *was_escaped) |
| 66 | { |
| 67 | gunichar c = get_char (str); |
| 68 | |
| 69 | *was_escaped = DO_ESCAPE && c == '\\'; |
| 70 | if (*was_escaped) |
| 71 | c = get_char (str); |
| 72 | |
| 73 | return c; |
| 74 | } |
| 75 | |
| 76 | /* Match STRING against the filename pattern PATTERN, returning zero if |
| 77 | it matches, nonzero if not. */ |
| 78 | |
| 79 | static gboolean |
| 80 | gtk_fnmatch_intern (const char *pattern, |
| 81 | const char *string, |
| 82 | gboolean component_start, |
| 83 | gboolean no_leading_period) |
| 84 | { |
| 85 | const char *p = pattern, *n = string; |
| 86 | |
| 87 | while (*p) |
| 88 | { |
| 89 | const char *last_n = n; |
| 90 | |
| 91 | gunichar c = get_char (&p); |
| 92 | gunichar nc = get_char (&n); |
| 93 | |
| 94 | switch (c) |
| 95 | { |
| 96 | case '?': |
| 97 | if (nc == '\0') |
| 98 | return FALSE; |
| 99 | else if (nc == G_DIR_SEPARATOR) |
| 100 | return FALSE; |
| 101 | else if (nc == '.' && component_start && no_leading_period) |
| 102 | return FALSE; |
| 103 | break; |
| 104 | case '\\': |
| 105 | if (DO_ESCAPE) |
| 106 | c = get_char (&p); |
| 107 | if (nc != c) |
| 108 | return FALSE; |
| 109 | break; |
| 110 | case '*': |
| 111 | if (nc == '.' && component_start && no_leading_period) |
| 112 | return FALSE; |
| 113 | |
| 114 | { |
| 115 | const char *last_p = p; |
| 116 | |
| 117 | for (last_p = p, c = get_char (&p); |
| 118 | c == '?' || c == '*'; |
| 119 | last_p = p, c = get_char (&p)) |
| 120 | { |
| 121 | if (c == '?') |
| 122 | { |
| 123 | if (nc == '\0') |
| 124 | return FALSE; |
| 125 | else if (nc == G_DIR_SEPARATOR) |
| 126 | return FALSE; |
| 127 | else |
| 128 | { |
| 129 | last_n = n; nc = get_char (&n); |
| 130 | } |
| 131 | } |
| 132 | } |
| 133 | |
| 134 | /* If the pattern ends with wildcards, we have a |
| 135 | * guaranteed match unless there is a dir separator |
| 136 | * in the remainder of the string. |
| 137 | */ |
| 138 | if (c == '\0') |
| 139 | { |
| 140 | if (strchr (last_n, G_DIR_SEPARATOR) != NULL) |
| 141 | return FALSE; |
| 142 | else |
| 143 | return TRUE; |
| 144 | } |
| 145 | |
| 146 | if (DO_ESCAPE && c == '\\') |
| 147 | c = get_char (&p); |
| 148 | |
| 149 | for (p = last_p; nc != '\0';) |
| 150 | { |
| 151 | if ((c == '[' || nc == c) && |
| 152 | gtk_fnmatch_intern (p, last_n, component_start, no_leading_period)) |
| 153 | return TRUE; |
| 154 | |
| 155 | component_start = (nc == G_DIR_SEPARATOR); |
| 156 | last_n = n; |
| 157 | nc = get_char (&n); |
| 158 | } |
| 159 | |
| 160 | return FALSE; |
| 161 | } |
| 162 | |
| 163 | case '[': |
| 164 | { |
| 165 | /* Nonzero if the sense of the character class is inverted. */ |
| 166 | gboolean not; |
| 167 | gboolean was_escaped; |
| 168 | |
| 169 | if (nc == '\0' || nc == G_DIR_SEPARATOR) |
| 170 | return FALSE; |
| 171 | |
| 172 | if (nc == '.' && component_start && no_leading_period) |
| 173 | return FALSE; |
| 174 | |
| 175 | not = (*p == '!' || *p == '^'); |
| 176 | if (not) |
| 177 | ++p; |
| 178 | |
| 179 | c = get_unescaped_char (&p, &was_escaped); |
| 180 | for (;;) |
| 181 | { |
| 182 | register gunichar cstart = c, cend = c; |
| 183 | if (c == '\0') |
| 184 | /* [ (unterminated) loses. */ |
| 185 | return FALSE; |
| 186 | |
| 187 | c = get_unescaped_char (&p, &was_escaped); |
| 188 | |
| 189 | if (!was_escaped && c == '-' && *p != ']') |
| 190 | { |
| 191 | cend = get_unescaped_char (&p, &was_escaped); |
| 192 | if (cend == '\0') |
| 193 | return FALSE; |
| 194 | |
| 195 | c = get_char (&p); |
| 196 | } |
| 197 | |
| 198 | if (nc >= cstart && nc <= cend) |
| 199 | goto matched; |
| 200 | |
| 201 | if (!was_escaped && c == ']') |
| 202 | break; |
| 203 | } |
| 204 | if (!not) |
| 205 | return FALSE; |
| 206 | break; |
| 207 | |
| 208 | matched:; |
| 209 | /* Skip the rest of the [...] that already matched. */ |
| 210 | /* XXX 1003.2d11 is unclear if was_escaped is right. */ |
| 211 | while (was_escaped || c != ']') |
| 212 | { |
| 213 | if (c == '\0') |
| 214 | /* [... (unterminated) loses. */ |
| 215 | return FALSE; |
| 216 | |
| 217 | c = get_unescaped_char (&p, &was_escaped); |
| 218 | } |
| 219 | if (not) |
| 220 | return FALSE; |
| 221 | } |
| 222 | break; |
| 223 | |
| 224 | default: |
| 225 | if (c != nc) |
| 226 | return FALSE; |
| 227 | } |
| 228 | |
| 229 | component_start = (nc == G_DIR_SEPARATOR); |
| 230 | } |
| 231 | |
| 232 | if (*n == '\0') |
| 233 | return TRUE; |
| 234 | |
| 235 | return FALSE; |
| 236 | } |
| 237 | |
| 238 | /* Match STRING against the filename pattern PATTERN, returning zero if |
| 239 | * it matches, nonzero if not. |
| 240 | * |
| 241 | * GTK+ used to use a old version of GNU fnmatch() that was buggy |
| 242 | * in various ways and didn't handle UTF-8. The following is |
| 243 | * converted to UTF-8. To simplify the process of making it |
| 244 | * correct, this is special-cased to the combinations of flags |
| 245 | * that gtkfilesel.c uses. |
| 246 | * |
| 247 | * FNM_FILE_NAME - always set |
| 248 | * FNM_LEADING_DIR - never set |
| 249 | * FNM_NOESCAPE - set only on windows |
| 250 | * FNM_CASEFOLD - set only on windows |
| 251 | */ |
| 252 | gboolean |
| 253 | _gtk_fnmatch (const char *pattern, |
| 254 | const char *string, |
| 255 | gboolean no_leading_period) |
| 256 | { |
| 257 | return gtk_fnmatch_intern (pattern, string, TRUE, no_leading_period); |
| 258 | } |
| 259 | |
| 260 | #undef FNMATCH_TEST_CASES |
| 261 | #ifdef FNMATCH_TEST_CASES |
| 262 | |
| 263 | #define TEST(pat, str, no_leading_period, result) \ |
| 264 | g_assert (_gtk_fnmatch ((pat), (str), (no_leading_period)) == result) |
| 265 | |
| 266 | int main (int argc, char **argv) |
| 267 | { |
| 268 | TEST ("[a-]", "-", TRUE, TRUE); |
| 269 | |
| 270 | TEST ("a", "a", TRUE, TRUE); |
| 271 | TEST ("a", "b", TRUE, FALSE); |
| 272 | |
| 273 | /* Test what ? matches */ |
| 274 | TEST ("?", "a", TRUE, TRUE); |
| 275 | TEST ("?", ".", TRUE, FALSE); |
| 276 | TEST ("a?", "a.", TRUE, TRUE); |
| 277 | TEST ("a/?", "a/b", TRUE, TRUE); |
| 278 | TEST ("a/?", "a/.", TRUE, FALSE); |
| 279 | TEST ("?", "/", TRUE, FALSE); |
| 280 | |
| 281 | /* Test what * matches */ |
| 282 | TEST ("*", "a", TRUE, TRUE); |
| 283 | TEST ("*", ".", TRUE, FALSE); |
| 284 | TEST ("a*", "a.", TRUE, TRUE); |
| 285 | TEST ("a/*", "a/b", TRUE, TRUE); |
| 286 | TEST ("a/*", "a/.", TRUE, FALSE); |
| 287 | TEST ("*", "/", TRUE, FALSE); |
| 288 | |
| 289 | /* Range tests */ |
| 290 | TEST ("[ab]", "a", TRUE, TRUE); |
| 291 | TEST ("[ab]", "c", TRUE, FALSE); |
| 292 | TEST ("[^ab]", "a", TRUE, FALSE); |
| 293 | TEST ("[!ab]", "a", TRUE, FALSE); |
| 294 | TEST ("[^ab]", "c", TRUE, TRUE); |
| 295 | TEST ("[!ab]", "c", TRUE, TRUE); |
| 296 | TEST ("[a-c]", "b", TRUE, TRUE); |
| 297 | TEST ("[a-c]", "d", TRUE, FALSE); |
| 298 | TEST ("[a-]", "-", TRUE, TRUE); |
| 299 | TEST ("[]]", "]", TRUE, TRUE); |
| 300 | TEST ("[^]]", "a", TRUE, TRUE); |
| 301 | TEST ("[!]]", "a", TRUE, TRUE); |
| 302 | |
| 303 | /* Various unclosed ranges */ |
| 304 | TEST ("[ab", "a", TRUE, FALSE); |
| 305 | TEST ("[a-", "a", TRUE, FALSE); |
| 306 | TEST ("[ab", "c", TRUE, FALSE); |
| 307 | TEST ("[a-", "c", TRUE, FALSE); |
| 308 | TEST ("[^]", "a", TRUE, FALSE); |
| 309 | |
| 310 | /* Ranges and special no-wildcard matches */ |
| 311 | TEST ("[.]", ".", TRUE, FALSE); |
| 312 | TEST ("a[.]", "a.", TRUE, TRUE); |
| 313 | TEST ("a/[.]", "a/.", TRUE, FALSE); |
| 314 | TEST ("[/]", "/", TRUE, FALSE); |
| 315 | TEST ("[^/]", "a", TRUE, TRUE); |
| 316 | |
| 317 | /* Basic tests of * (and combinations of * and ?) */ |
| 318 | TEST ("a*b", "ab", TRUE, TRUE); |
| 319 | TEST ("a*b", "axb", TRUE, TRUE); |
| 320 | TEST ("a*b", "axxb", TRUE, TRUE); |
| 321 | TEST ("a**b", "ab", TRUE, TRUE); |
| 322 | TEST ("a**b", "axb", TRUE, TRUE); |
| 323 | TEST ("a**b", "axxb", TRUE, TRUE); |
| 324 | TEST ("a*?*b", "ab", TRUE, FALSE); |
| 325 | TEST ("a*?*b", "axb", TRUE, TRUE); |
| 326 | TEST ("a*?*b", "axxb", TRUE, TRUE); |
| 327 | |
| 328 | /* Test of *[range] */ |
| 329 | TEST ("a*[cd]", "ac", TRUE, TRUE); |
| 330 | TEST ("a*[cd]", "axc", TRUE, TRUE); |
| 331 | TEST ("a*[cd]", "axx", TRUE, FALSE); |
| 332 | |
| 333 | TEST ("a/[.]", "a/.", TRUE, FALSE); |
| 334 | TEST ("a*[.]", "a/.", TRUE, FALSE); |
| 335 | |
| 336 | /* Test of UTF-8 */ |
| 337 | |
| 338 | TEST ("ä", "ä", TRUE, TRUE); /* TEST ("ä", "ä", TRUE); */ |
| 339 | TEST ("?", "ä", TRUE, TRUE); /* TEST ("?", "ä", TRUE); */ |
| 340 | TEST ("*ö", "äö", TRUE, TRUE); /* TEST ("*ö", "äö", TRUE); */ |
| 341 | TEST ("*ö", "ääö", TRUE, TRUE); /* TEST ("*ö", "ääö", TRUE); */ |
| 342 | TEST ("[ä]", "ä", TRUE, TRUE); /* TEST ("[ä]", "ä", TRUE); */ |
| 343 | TEST ("[ä-ö]", "é", TRUE, TRUE); /* TEST ("[ä-ö]", "é", TRUE); */ |
| 344 | TEST ("[ä-ö]", "a", TRUE, FALSE); /* TEST ("[ä-ö]", "a", FALSE); */ |
| 345 | |
| 346 | #ifdef DO_ESCAPE |
| 347 | /* Tests of escaping */ |
| 348 | TEST ("\\\\", "\\", TRUE, TRUE); |
| 349 | TEST ("\\?", "?", TRUE, TRUE); |
| 350 | TEST ("\\?", "a", TRUE, FALSE); |
| 351 | TEST ("\\*", "*", TRUE, TRUE); |
| 352 | TEST ("\\*", "a", TRUE, FALSE); |
| 353 | TEST ("\\[a-b]", "[a-b]", TRUE, TRUE); |
| 354 | TEST ("[\\\\]", "\\", TRUE, TRUE); |
| 355 | TEST ("[\\^a]", "a", TRUE, TRUE); |
| 356 | TEST ("[a\\-c]", "b", TRUE, FALSE); |
| 357 | TEST ("[a\\-c]", "-", TRUE, TRUE); |
| 358 | TEST ("[a\\]", "a", TRUE, FALSE); |
| 359 | #endif /* DO_ESCAPE */ |
| 360 | |
| 361 | return 0; |
| 362 | } |
| 363 | |
| 364 | #endif /* FNMATCH_TEST_CASES */ |