1 /* $OpenBSD: fnmatch.c,v 1.15 2011/02/10 21:31:59 stsp Exp $ */
3 /* Copyright (c) 2011, VMware, Inc.
6 * Redistribution and use in source and binary forms, with or without
7 * modification, are permitted provided that the following conditions are met:
8 * * Redistributions of source code must retain the above copyright
9 * notice, this list of conditions and the following disclaimer.
10 * * 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 * * Neither the name of the VMware, Inc. nor the names of its contributors
14 * may be used to endorse or promote products derived from this software
15 * without specific prior written permission.
17 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
18 * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
19 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
20 * ARE DISCLAIMED. IN NO EVENT SHALL VMWARE, INC. OR CONTRIBUTORS BE LIABLE FOR
21 * ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
22 * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
23 * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
24 * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
25 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
26 * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
29 /* Authored by William A. Rowe Jr. <wrowe; apache.org, vmware.com>, April 2011
31 * Derived from The Open Group Base Specifications Issue 7, IEEE Std 1003.1-2008
33 * http://pubs.opengroup.org/onlinepubs/9699919799/functions/fnmatch.html
35 * Filename pattern matches defined in section 2.13, "Pattern Matching Notation"
36 * from chapter 2. "Shell Command Language"
37 * http://pubs.opengroup.org/onlinepubs/9699919799/utilities/V3_chap02.html#tag_18_13
38 * where; 1. A bracket expression starting with an unquoted <circumflex> '^'
39 * character CONTINUES to specify a non-matching list; 2. an explicit <period> '.'
40 * in a bracket expression matching list, e.g. "[.abc]" does NOT match a leading
41 * <period> in a filename; 3. a <left-square-bracket> '[' which does not introduce
42 * a valid bracket expression is treated as an ordinary character; 4. a differing
43 * number of consecutive slashes within pattern and string will NOT match;
44 * 5. a trailing '\' in FNM_ESCAPE mode is treated as an ordinary '\' character.
46 * Bracket expansion defined in section 9.3.5, "RE Bracket Expression",
47 * from chapter 9, "Regular Expressions"
48 * http://pubs.opengroup.org/onlinepubs/9699919799/basedefs/V1_chap09.html#tag_09_03_05
49 * with no support for collating symbols, equivalence class expressions or
50 * character class expressions. A partial range expression with a leading
51 * hyphen following a valid range expression will match only the ordinary
52 * <hyphen> and the ending character (e.g. "[a-m-z]" will match characters
53 * 'a' through 'm', a <hyphen> '-', or a 'z').
55 * Supports BSD extensions FNM_LEADING_DIR to match pattern to the end of one
56 * path segment of string, and FNM_CASEFOLD to ignore alpha case.
58 * NOTE: Only POSIX/C single byte locales are correctly supported at this time.
59 * Notably, non-POSIX locales with FNM_CASEFOLD produce undefined results,
60 * particularly in ranges of mixed case (e.g. "[A-z]") or spanning alpha and
61 * nonalpha characters within a range.
63 * XXX comments below indicate porting required for multi-byte character sets
64 * and non-POSIX locale collation orders; requires mbr* APIs to track shift
65 * state of pattern and string (rewinding pattern and string repeatedly).
67 * Certain parts of the code assume 0x00-0x3F are unique with any MBCS (e.g.
68 * UTF-8, SHIFT-JIS, etc). Any implementation allowing '\' as an alternate
69 * path delimiter must be aware that 0x5C is NOT unique within SHIFT-JIS.
74 #include <sys/types.h>
80 #endif /* HAVE_STRING_H */
83 #endif /* HAVE_STRINGS_H */
87 #include "compat/charclass.h"
88 #include "compat/fnmatch.h"
91 #define RANGE_NOMATCH 0
92 #define RANGE_ERROR (-1)
95 classmatch(const char *pattern, char test, int foldcase, const char **ep)
97 const char * const mismatch = pattern;
100 int rval = RANGE_NOMATCH;
103 if (pattern[0] != '[' || pattern[1] != ':') {
109 if ((colon = strchr(pattern, ':')) == NULL || colon[1] != ']') {
114 len = (size_t)(colon - pattern);
116 if (foldcase && strncmp(pattern, "upper:]", 7) == 0)
118 for (cc = cclasses; cc->name != NULL; cc++) {
119 if (!strncmp(pattern, cc->name, len) && cc->name[len] == '\0') {
120 if (cc->isctype((unsigned char)test))
125 if (cc->name == NULL) {
126 /* invalid character class, treat as normal text */
133 /* Most MBCS/collation/case issues handled here. Wildcard '*' is not handled.
134 * EOS '\0' and the FNM_PATHNAME '/' delimiters are not advanced over,
135 * however the "\/" sequence is advanced to '/'.
137 * Both pattern and string are **char to support pointer increment of arbitrary
138 * multibyte characters for the given locale, in a later iteration of this code
140 static int fnmatch_ch(const char **pattern, const char **string, int flags)
142 const char * const mismatch = *pattern;
143 const int nocase = !!(flags & FNM_CASEFOLD);
144 const int escape = !(flags & FNM_NOESCAPE);
145 const int slash = !!(flags & FNM_PATHNAME);
146 int result = FNM_NOMATCH;
150 if (**pattern == '[')
154 /* Handle negation, either leading ! or ^ operators (never both) */
155 negate = ((**pattern == '!') || (**pattern == '^'));
159 /* ']' is an ordinary character at the start of the range pattern */
160 if (**pattern == ']')
161 goto leadingclosebrace;
165 if (**pattern == ']') {
167 /* XXX: Fix for MBCS character width */
169 return (result ^ negate);
172 if (escape && (**pattern == '\\')) {
175 /* Patterns must be terminated with ']', not EOS */
180 /* Patterns must be terminated with ']' not '/' */
181 if (slash && (**pattern == '/'))
184 /* Match character classes. */
185 if (classmatch(*pattern, **string, nocase, pattern)
192 /* Look at only well-formed range patterns;
193 * "x-]" is not allowed unless escaped ("x-\]")
194 * XXX: Fix for locale/MBCS character width
196 if (((*pattern)[1] == '-') && ((*pattern)[2] != ']'))
199 *pattern += (escape && ((*pattern)[2] == '\\')) ? 3 : 2;
201 /* NOT a properly balanced [expr] pattern, EOS terminated
202 * or ranges containing a slash in FNM_PATHNAME mode pattern
203 * fall out to to the rewind and test '[' literal code path
205 if (!**pattern || (slash && (**pattern == '/')))
208 /* XXX: handle locale/MBCS comparison, advance by MBCS char width */
209 if ((**string >= *startch) && (**string <= **pattern))
211 else if (nocase && (isupper(**string) || isupper(*startch)
212 || isupper(**pattern))
213 && (tolower(**string) >= tolower(*startch))
214 && (tolower(**string) <= tolower(**pattern)))
221 /* XXX: handle locale/MBCS comparison, advance by MBCS char width */
222 if ((**string == **pattern))
224 else if (nocase && (isupper(**string) || isupper(**pattern))
225 && (tolower(**string) == tolower(**pattern)))
231 /* NOT a properly balanced [expr] pattern; Rewind
232 * and reset result to test '[' literal
235 result = FNM_NOMATCH;
237 else if (**pattern == '?') {
238 /* Optimize '?' match before unescaping **pattern */
239 if (!**string || (slash && (**string == '/')))
242 goto fnmatch_ch_success;
244 else if (escape && (**pattern == '\\') && (*pattern)[1]) {
248 /* XXX: handle locale/MBCS comparison, advance by the MBCS char width */
249 if (**string == **pattern)
251 else if (nocase && (isupper(**string) || isupper(**pattern))
252 && (tolower(**string) == tolower(**pattern)))
255 /* Refuse to advance over trailing slash or nulls
257 if (!**string || !**pattern || (slash && ((**string == '/') || (**pattern == '/'))))
266 int rpl_fnmatch(const char *pattern, const char *string, int flags)
268 static const char dummystring[2] = {' ', 0};
269 const int escape = !(flags & FNM_NOESCAPE);
270 const int slash = !!(flags & FNM_PATHNAME);
271 const int leading_dir = !!(flags & FNM_LEADING_DIR);
272 const char *strendseg;
273 const char *dummyptr;
274 const char *matchptr;
276 /* For '*' wild processing only; surpress 'used before initialization'
277 * warnings with dummy initialization values;
279 const char *strstartseg = NULL;
280 const char *mismatch = NULL;
283 if (strlen(pattern) > PATH_MAX || strlen(string) > PATH_MAX)
289 while (*pattern && *string)
291 /* Pre-decode "\/" which has no special significance, and
292 * match balanced slashes, starting a new segment pattern
294 if (slash && escape && (*pattern == '\\') && (pattern[1] == '/'))
296 if (slash && (*pattern == '/') && (*string == '/')) {
302 /* At the beginning of each segment, validate leading period behavior.
304 if ((flags & FNM_PERIOD) && (*string == '.'))
308 else if (escape && (*pattern == '\\') && (pattern[1] == '.'))
315 /* Determine the end of string segment
317 * Presumes '/' character is unique, not composite in any MBCS encoding
320 strendseg = strchr(string, '/');
322 strendseg = strchr(string, '\0');
325 strendseg = strchr(string, '\0');
328 /* Allow pattern '*' to be consumed even with no remaining string to match
332 if ((string > strendseg)
333 || ((string == strendseg) && (*pattern != '*')))
336 if (slash && ((*pattern == '/')
337 || (escape && (*pattern == '\\')
338 && (pattern[1] == '/'))))
341 /* Reduce groups of '*' and '?' to n '?' matches
342 * followed by one '*' test for simplicity
344 for (wild = 0; ((*pattern == '*') || (*pattern == '?')); ++pattern)
346 if (*pattern == '*') {
349 else if (string < strendseg) { /* && (*pattern == '?') */
350 /* XXX: Advance 1 char for MBCS locale */
353 else { /* (string >= strendseg) && (*pattern == '?') */
360 strstartseg = string;
363 /* Count fixed (non '*') char matches remaining in pattern
364 * excluding '/' (or "\/") and '*'
366 for (matchptr = pattern, matchlen = 0; 1; ++matchlen)
368 if ((*matchptr == '\0')
369 || (slash && ((*matchptr == '/')
370 || (escape && (*matchptr == '\\')
371 && (matchptr[1] == '/')))))
373 /* Compare precisely this many trailing string chars,
374 * the resulting match needs no wildcard loop
376 /* XXX: Adjust for MBCS */
377 if (string + matchlen > strendseg)
380 string = strendseg - matchlen;
385 if (*matchptr == '*')
387 /* Ensure at least this many trailing string chars remain
388 * for the first comparison
390 /* XXX: Adjust for MBCS */
391 if (string + matchlen > strendseg)
394 /* Begin first wild comparison at the current position */
398 /* Skip forward in pattern by a single character match
399 * Use a dummy fnmatch_ch() test to count one "[range]" escape
401 /* XXX: Adjust for MBCS */
402 if (escape && (*matchptr == '\\') && matchptr[1]) {
405 else if (*matchptr == '[') {
406 dummyptr = dummystring;
407 fnmatch_ch(&matchptr, &dummyptr, flags);
415 /* Incrementally match string against the pattern
417 while (*pattern && (string < strendseg))
419 /* Success; begin a new wild pattern search
424 if (slash && ((*string == '/')
426 || (escape && (*pattern == '\\')
427 && (pattern[1] == '/'))))
430 /* Compare ch's (the pattern is advanced over "\/" to the '/',
431 * but slashes will mismatch, and are not consumed)
433 if (!fnmatch_ch(&pattern, &string, flags))
436 /* Failed to match, loop against next char offset of string segment
437 * until not enough string chars remain to match the fixed pattern
440 /* XXX: Advance 1 char for MBCS locale */
441 string = ++strstartseg;
442 if (string + matchlen > strendseg)
453 if (*string && !((slash || leading_dir) && (*string == '/')))
456 if (*pattern && !(slash && ((*pattern == '/')
457 || (escape && (*pattern == '\\')
458 && (pattern[1] == '/')))))
461 if (leading_dir && !*pattern && *string == '/')
465 /* Where both pattern and string are at EOS, declare success
467 if (!*string && !*pattern)
470 /* pattern didn't match to the end of string */