1 | /***************************************************************************
|
---|
2 | * _ _ ____ _
|
---|
3 | * Project ___| | | | _ \| |
|
---|
4 | * / __| | | | |_) | |
|
---|
5 | * | (__| |_| | _ <| |___
|
---|
6 | * \___|\___/|_| \_\_____|
|
---|
7 | *
|
---|
8 | * Copyright (C) 1998 - 2018, Daniel Stenberg, <daniel@haxx.se>, et al.
|
---|
9 | *
|
---|
10 | * This software is licensed as described in the file COPYING, which
|
---|
11 | * you should have received as part of this distribution. The terms
|
---|
12 | * are also available at https://curl.haxx.se/docs/copyright.html.
|
---|
13 | *
|
---|
14 | * You may opt to use, copy, modify, merge, publish, distribute and/or sell
|
---|
15 | * copies of the Software, and permit persons to whom the Software is
|
---|
16 | * furnished to do so, under the terms of the COPYING file.
|
---|
17 | *
|
---|
18 | * This software is distributed on an "AS IS" basis, WITHOUT WARRANTY OF ANY
|
---|
19 | * KIND, either express or implied.
|
---|
20 | *
|
---|
21 | ***************************************************************************/
|
---|
22 |
|
---|
23 | #include "curl_setup.h"
|
---|
24 |
|
---|
25 | #include <curl/curl.h>
|
---|
26 |
|
---|
27 | #include "curl_fnmatch.h"
|
---|
28 | #include "curl_memory.h"
|
---|
29 |
|
---|
30 | /* The last #include file should be: */
|
---|
31 | #include "memdebug.h"
|
---|
32 |
|
---|
33 | #ifndef HAVE_FNMATCH
|
---|
34 |
|
---|
35 | /*
|
---|
36 | * TODO:
|
---|
37 | *
|
---|
38 | * Make this function match POSIX. Test 1307 includes a set of test patterns
|
---|
39 | * that returns different results with a POSIX fnmatch() than with this
|
---|
40 | * implementation and this is considered a bug where POSIX is the guiding
|
---|
41 | * light.
|
---|
42 | */
|
---|
43 |
|
---|
44 | #define CURLFNM_CHARSET_LEN (sizeof(char) * 256)
|
---|
45 | #define CURLFNM_CHSET_SIZE (CURLFNM_CHARSET_LEN + 15)
|
---|
46 |
|
---|
47 | #define CURLFNM_NEGATE CURLFNM_CHARSET_LEN
|
---|
48 |
|
---|
49 | #define CURLFNM_ALNUM (CURLFNM_CHARSET_LEN + 1)
|
---|
50 | #define CURLFNM_DIGIT (CURLFNM_CHARSET_LEN + 2)
|
---|
51 | #define CURLFNM_XDIGIT (CURLFNM_CHARSET_LEN + 3)
|
---|
52 | #define CURLFNM_ALPHA (CURLFNM_CHARSET_LEN + 4)
|
---|
53 | #define CURLFNM_PRINT (CURLFNM_CHARSET_LEN + 5)
|
---|
54 | #define CURLFNM_BLANK (CURLFNM_CHARSET_LEN + 6)
|
---|
55 | #define CURLFNM_LOWER (CURLFNM_CHARSET_LEN + 7)
|
---|
56 | #define CURLFNM_GRAPH (CURLFNM_CHARSET_LEN + 8)
|
---|
57 | #define CURLFNM_SPACE (CURLFNM_CHARSET_LEN + 9)
|
---|
58 | #define CURLFNM_UPPER (CURLFNM_CHARSET_LEN + 10)
|
---|
59 |
|
---|
60 | typedef enum {
|
---|
61 | CURLFNM_SCHS_DEFAULT = 0,
|
---|
62 | CURLFNM_SCHS_RIGHTBR,
|
---|
63 | CURLFNM_SCHS_RIGHTBRLEFTBR
|
---|
64 | } setcharset_state;
|
---|
65 |
|
---|
66 | typedef enum {
|
---|
67 | CURLFNM_PKW_INIT = 0,
|
---|
68 | CURLFNM_PKW_DDOT
|
---|
69 | } parsekey_state;
|
---|
70 |
|
---|
71 | typedef enum {
|
---|
72 | CCLASS_OTHER = 0,
|
---|
73 | CCLASS_DIGIT,
|
---|
74 | CCLASS_UPPER,
|
---|
75 | CCLASS_LOWER
|
---|
76 | } char_class;
|
---|
77 |
|
---|
78 | #define SETCHARSET_OK 1
|
---|
79 | #define SETCHARSET_FAIL 0
|
---|
80 |
|
---|
81 | static int parsekeyword(unsigned char **pattern, unsigned char *charset)
|
---|
82 | {
|
---|
83 | parsekey_state state = CURLFNM_PKW_INIT;
|
---|
84 | #define KEYLEN 10
|
---|
85 | char keyword[KEYLEN] = { 0 };
|
---|
86 | int found = FALSE;
|
---|
87 | int i;
|
---|
88 | unsigned char *p = *pattern;
|
---|
89 | for(i = 0; !found; i++) {
|
---|
90 | char c = *p++;
|
---|
91 | if(i >= KEYLEN)
|
---|
92 | return SETCHARSET_FAIL;
|
---|
93 | switch(state) {
|
---|
94 | case CURLFNM_PKW_INIT:
|
---|
95 | if(ISLOWER(c))
|
---|
96 | keyword[i] = c;
|
---|
97 | else if(c == ':')
|
---|
98 | state = CURLFNM_PKW_DDOT;
|
---|
99 | else
|
---|
100 | return SETCHARSET_FAIL;
|
---|
101 | break;
|
---|
102 | case CURLFNM_PKW_DDOT:
|
---|
103 | if(c == ']')
|
---|
104 | found = TRUE;
|
---|
105 | else
|
---|
106 | return SETCHARSET_FAIL;
|
---|
107 | }
|
---|
108 | }
|
---|
109 | #undef KEYLEN
|
---|
110 |
|
---|
111 | *pattern = p; /* move caller's pattern pointer */
|
---|
112 | if(strcmp(keyword, "digit") == 0)
|
---|
113 | charset[CURLFNM_DIGIT] = 1;
|
---|
114 | else if(strcmp(keyword, "alnum") == 0)
|
---|
115 | charset[CURLFNM_ALNUM] = 1;
|
---|
116 | else if(strcmp(keyword, "alpha") == 0)
|
---|
117 | charset[CURLFNM_ALPHA] = 1;
|
---|
118 | else if(strcmp(keyword, "xdigit") == 0)
|
---|
119 | charset[CURLFNM_XDIGIT] = 1;
|
---|
120 | else if(strcmp(keyword, "print") == 0)
|
---|
121 | charset[CURLFNM_PRINT] = 1;
|
---|
122 | else if(strcmp(keyword, "graph") == 0)
|
---|
123 | charset[CURLFNM_GRAPH] = 1;
|
---|
124 | else if(strcmp(keyword, "space") == 0)
|
---|
125 | charset[CURLFNM_SPACE] = 1;
|
---|
126 | else if(strcmp(keyword, "blank") == 0)
|
---|
127 | charset[CURLFNM_BLANK] = 1;
|
---|
128 | else if(strcmp(keyword, "upper") == 0)
|
---|
129 | charset[CURLFNM_UPPER] = 1;
|
---|
130 | else if(strcmp(keyword, "lower") == 0)
|
---|
131 | charset[CURLFNM_LOWER] = 1;
|
---|
132 | else
|
---|
133 | return SETCHARSET_FAIL;
|
---|
134 | return SETCHARSET_OK;
|
---|
135 | }
|
---|
136 |
|
---|
137 | /* Return the character class. */
|
---|
138 | static char_class charclass(unsigned char c)
|
---|
139 | {
|
---|
140 | if(ISUPPER(c))
|
---|
141 | return CCLASS_UPPER;
|
---|
142 | if(ISLOWER(c))
|
---|
143 | return CCLASS_LOWER;
|
---|
144 | if(ISDIGIT(c))
|
---|
145 | return CCLASS_DIGIT;
|
---|
146 | return CCLASS_OTHER;
|
---|
147 | }
|
---|
148 |
|
---|
149 | /* Include a character or a range in set. */
|
---|
150 | static void setcharorrange(unsigned char **pp, unsigned char *charset)
|
---|
151 | {
|
---|
152 | unsigned char *p = (*pp)++;
|
---|
153 | unsigned char c = *p++;
|
---|
154 |
|
---|
155 | charset[c] = 1;
|
---|
156 | if(ISALNUM(c) && *p++ == '-') {
|
---|
157 | char_class cc = charclass(c);
|
---|
158 | unsigned char endrange = *p++;
|
---|
159 |
|
---|
160 | if(endrange == '\\')
|
---|
161 | endrange = *p++;
|
---|
162 | if(endrange >= c && charclass(endrange) == cc) {
|
---|
163 | while(c++ != endrange)
|
---|
164 | if(charclass(c) == cc) /* Chars in class may be not consecutive. */
|
---|
165 | charset[c] = 1;
|
---|
166 | *pp = p;
|
---|
167 | }
|
---|
168 | }
|
---|
169 | }
|
---|
170 |
|
---|
171 | /* returns 1 (true) if pattern is OK, 0 if is bad ("p" is pattern pointer) */
|
---|
172 | static int setcharset(unsigned char **p, unsigned char *charset)
|
---|
173 | {
|
---|
174 | setcharset_state state = CURLFNM_SCHS_DEFAULT;
|
---|
175 | bool something_found = FALSE;
|
---|
176 | unsigned char c;
|
---|
177 |
|
---|
178 | memset(charset, 0, CURLFNM_CHSET_SIZE);
|
---|
179 | for(;;) {
|
---|
180 | c = **p;
|
---|
181 | if(!c)
|
---|
182 | return SETCHARSET_FAIL;
|
---|
183 |
|
---|
184 | switch(state) {
|
---|
185 | case CURLFNM_SCHS_DEFAULT:
|
---|
186 | if(c == ']') {
|
---|
187 | if(something_found)
|
---|
188 | return SETCHARSET_OK;
|
---|
189 | something_found = TRUE;
|
---|
190 | state = CURLFNM_SCHS_RIGHTBR;
|
---|
191 | charset[c] = 1;
|
---|
192 | (*p)++;
|
---|
193 | }
|
---|
194 | else if(c == '[') {
|
---|
195 | unsigned char *pp = *p + 1;
|
---|
196 |
|
---|
197 | if(*pp++ == ':' && parsekeyword(&pp, charset))
|
---|
198 | *p = pp;
|
---|
199 | else {
|
---|
200 | charset[c] = 1;
|
---|
201 | (*p)++;
|
---|
202 | }
|
---|
203 | something_found = TRUE;
|
---|
204 | }
|
---|
205 | else if(c == '^' || c == '!') {
|
---|
206 | if(!something_found) {
|
---|
207 | if(charset[CURLFNM_NEGATE]) {
|
---|
208 | charset[c] = 1;
|
---|
209 | something_found = TRUE;
|
---|
210 | }
|
---|
211 | else
|
---|
212 | charset[CURLFNM_NEGATE] = 1; /* negate charset */
|
---|
213 | }
|
---|
214 | else
|
---|
215 | charset[c] = 1;
|
---|
216 | (*p)++;
|
---|
217 | }
|
---|
218 | else if(c == '\\') {
|
---|
219 | c = *(++(*p));
|
---|
220 | if(c)
|
---|
221 | setcharorrange(p, charset);
|
---|
222 | else
|
---|
223 | charset['\\'] = 1;
|
---|
224 | something_found = TRUE;
|
---|
225 | }
|
---|
226 | else {
|
---|
227 | setcharorrange(p, charset);
|
---|
228 | something_found = TRUE;
|
---|
229 | }
|
---|
230 | break;
|
---|
231 | case CURLFNM_SCHS_RIGHTBR:
|
---|
232 | if(c == '[') {
|
---|
233 | state = CURLFNM_SCHS_RIGHTBRLEFTBR;
|
---|
234 | charset[c] = 1;
|
---|
235 | (*p)++;
|
---|
236 | }
|
---|
237 | else if(c == ']') {
|
---|
238 | return SETCHARSET_OK;
|
---|
239 | }
|
---|
240 | else if(ISPRINT(c)) {
|
---|
241 | charset[c] = 1;
|
---|
242 | (*p)++;
|
---|
243 | state = CURLFNM_SCHS_DEFAULT;
|
---|
244 | }
|
---|
245 | else
|
---|
246 | /* used 'goto fail' instead of 'return SETCHARSET_FAIL' to avoid a
|
---|
247 | * nonsense warning 'statement not reached' at end of the fnc when
|
---|
248 | * compiling on Solaris */
|
---|
249 | goto fail;
|
---|
250 | break;
|
---|
251 | case CURLFNM_SCHS_RIGHTBRLEFTBR:
|
---|
252 | if(c == ']')
|
---|
253 | return SETCHARSET_OK;
|
---|
254 | state = CURLFNM_SCHS_DEFAULT;
|
---|
255 | charset[c] = 1;
|
---|
256 | (*p)++;
|
---|
257 | break;
|
---|
258 | }
|
---|
259 | }
|
---|
260 | fail:
|
---|
261 | return SETCHARSET_FAIL;
|
---|
262 | }
|
---|
263 |
|
---|
264 | static int loop(const unsigned char *pattern, const unsigned char *string,
|
---|
265 | int maxstars)
|
---|
266 | {
|
---|
267 | unsigned char *p = (unsigned char *)pattern;
|
---|
268 | unsigned char *s = (unsigned char *)string;
|
---|
269 | unsigned char charset[CURLFNM_CHSET_SIZE] = { 0 };
|
---|
270 |
|
---|
271 | for(;;) {
|
---|
272 | unsigned char *pp;
|
---|
273 |
|
---|
274 | switch(*p) {
|
---|
275 | case '*':
|
---|
276 | if(!maxstars)
|
---|
277 | return CURL_FNMATCH_NOMATCH;
|
---|
278 | /* Regroup consecutive stars and question marks. This can be done because
|
---|
279 | '*?*?*' can be expressed as '??*'. */
|
---|
280 | for(;;) {
|
---|
281 | if(*++p == '\0')
|
---|
282 | return CURL_FNMATCH_MATCH;
|
---|
283 | if(*p == '?') {
|
---|
284 | if(!*s++)
|
---|
285 | return CURL_FNMATCH_NOMATCH;
|
---|
286 | }
|
---|
287 | else if(*p != '*')
|
---|
288 | break;
|
---|
289 | }
|
---|
290 | /* Skip string characters until we find a match with pattern suffix. */
|
---|
291 | for(maxstars--; *s; s++) {
|
---|
292 | if(loop(p, s, maxstars) == CURL_FNMATCH_MATCH)
|
---|
293 | return CURL_FNMATCH_MATCH;
|
---|
294 | }
|
---|
295 | return CURL_FNMATCH_NOMATCH;
|
---|
296 | case '?':
|
---|
297 | if(!*s)
|
---|
298 | return CURL_FNMATCH_NOMATCH;
|
---|
299 | s++;
|
---|
300 | p++;
|
---|
301 | break;
|
---|
302 | case '\0':
|
---|
303 | return *s? CURL_FNMATCH_NOMATCH: CURL_FNMATCH_MATCH;
|
---|
304 | case '\\':
|
---|
305 | if(p[1])
|
---|
306 | p++;
|
---|
307 | if(*s++ != *p++)
|
---|
308 | return CURL_FNMATCH_NOMATCH;
|
---|
309 | break;
|
---|
310 | case '[':
|
---|
311 | pp = p + 1; /* Copy in case of syntax error in set. */
|
---|
312 | if(setcharset(&pp, charset)) {
|
---|
313 | int found = FALSE;
|
---|
314 | if(!*s)
|
---|
315 | return CURL_FNMATCH_NOMATCH;
|
---|
316 | if(charset[(unsigned int)*s])
|
---|
317 | found = TRUE;
|
---|
318 | else if(charset[CURLFNM_ALNUM])
|
---|
319 | found = ISALNUM(*s);
|
---|
320 | else if(charset[CURLFNM_ALPHA])
|
---|
321 | found = ISALPHA(*s);
|
---|
322 | else if(charset[CURLFNM_DIGIT])
|
---|
323 | found = ISDIGIT(*s);
|
---|
324 | else if(charset[CURLFNM_XDIGIT])
|
---|
325 | found = ISXDIGIT(*s);
|
---|
326 | else if(charset[CURLFNM_PRINT])
|
---|
327 | found = ISPRINT(*s);
|
---|
328 | else if(charset[CURLFNM_SPACE])
|
---|
329 | found = ISSPACE(*s);
|
---|
330 | else if(charset[CURLFNM_UPPER])
|
---|
331 | found = ISUPPER(*s);
|
---|
332 | else if(charset[CURLFNM_LOWER])
|
---|
333 | found = ISLOWER(*s);
|
---|
334 | else if(charset[CURLFNM_BLANK])
|
---|
335 | found = ISBLANK(*s);
|
---|
336 | else if(charset[CURLFNM_GRAPH])
|
---|
337 | found = ISGRAPH(*s);
|
---|
338 |
|
---|
339 | if(charset[CURLFNM_NEGATE])
|
---|
340 | found = !found;
|
---|
341 |
|
---|
342 | if(!found)
|
---|
343 | return CURL_FNMATCH_NOMATCH;
|
---|
344 | p = pp + 1;
|
---|
345 | s++;
|
---|
346 | break;
|
---|
347 | }
|
---|
348 | /* Syntax error in set; mismatch! */
|
---|
349 | return CURL_FNMATCH_NOMATCH;
|
---|
350 |
|
---|
351 | default:
|
---|
352 | if(*p++ != *s++)
|
---|
353 | return CURL_FNMATCH_NOMATCH;
|
---|
354 | break;
|
---|
355 | }
|
---|
356 | }
|
---|
357 | }
|
---|
358 |
|
---|
359 | /*
|
---|
360 | * @unittest: 1307
|
---|
361 | */
|
---|
362 | int Curl_fnmatch(void *ptr, const char *pattern, const char *string)
|
---|
363 | {
|
---|
364 | (void)ptr; /* the argument is specified by the curl_fnmatch_callback
|
---|
365 | prototype, but not used by Curl_fnmatch() */
|
---|
366 | if(!pattern || !string) {
|
---|
367 | return CURL_FNMATCH_FAIL;
|
---|
368 | }
|
---|
369 | return loop((unsigned char *)pattern, (unsigned char *)string, 2);
|
---|
370 | }
|
---|
371 | #else
|
---|
372 | #include <fnmatch.h>
|
---|
373 | /*
|
---|
374 | * @unittest: 1307
|
---|
375 | */
|
---|
376 | int Curl_fnmatch(void *ptr, const char *pattern, const char *string)
|
---|
377 | {
|
---|
378 | int rc;
|
---|
379 | (void)ptr; /* the argument is specified by the curl_fnmatch_callback
|
---|
380 | prototype, but not used by Curl_fnmatch() */
|
---|
381 | if(!pattern || !string) {
|
---|
382 | return CURL_FNMATCH_FAIL;
|
---|
383 | }
|
---|
384 | rc = fnmatch(pattern, string, 0);
|
---|
385 | switch(rc) {
|
---|
386 | case 0:
|
---|
387 | return CURL_FNMATCH_MATCH;
|
---|
388 | case FNM_NOMATCH:
|
---|
389 | return CURL_FNMATCH_NOMATCH;
|
---|
390 | default:
|
---|
391 | return CURL_FNMATCH_FAIL;
|
---|
392 | }
|
---|
393 | /* not reached */
|
---|
394 | }
|
---|
395 |
|
---|
396 | #endif
|
---|