]>
Commit | Line | Data |
---|---|---|
22a2d3d5 UG |
1 | /* |
2 | * Copyright (C) the libgit2 contributors. All rights reserved. | |
3 | * | |
4 | * This file is part of libgit2, distributed under the GNU GPL v2 with | |
5 | * a Linking Exception. For full terms see the included COPYING file. | |
6 | * | |
7 | * Do shell-style pattern matching for ?, \, [], and * characters. | |
8 | * It is 8bit clean. | |
9 | * | |
10 | * Written by Rich $alz, mirror!rs, Wed Nov 26 19:03:17 EST 1986. | |
11 | * Rich $alz is now <rsalz@bbn.com>. | |
12 | * | |
13 | * Modified by Wayne Davison to special-case '/' matching, to make '**' | |
14 | * work differently than '*', and to fix the character-class code. | |
15 | * | |
16 | * Imported from git.git. | |
17 | */ | |
18 | ||
19 | #include "wildmatch.h" | |
20 | ||
21 | #define GIT_SPACE 0x01 | |
22 | #define GIT_DIGIT 0x02 | |
23 | #define GIT_ALPHA 0x04 | |
24 | #define GIT_GLOB_SPECIAL 0x08 | |
25 | #define GIT_REGEX_SPECIAL 0x10 | |
26 | #define GIT_PATHSPEC_MAGIC 0x20 | |
27 | #define GIT_CNTRL 0x40 | |
28 | #define GIT_PUNCT 0x80 | |
29 | ||
30 | enum { | |
31 | S = GIT_SPACE, | |
32 | A = GIT_ALPHA, | |
33 | D = GIT_DIGIT, | |
34 | G = GIT_GLOB_SPECIAL, /* *, ?, [, \\ */ | |
35 | R = GIT_REGEX_SPECIAL, /* $, (, ), +, ., ^, {, | */ | |
36 | P = GIT_PATHSPEC_MAGIC, /* other non-alnum, except for ] and } */ | |
37 | X = GIT_CNTRL, | |
38 | U = GIT_PUNCT, | |
39 | Z = GIT_CNTRL | GIT_SPACE | |
40 | }; | |
41 | ||
42 | static const unsigned char sane_ctype[256] = { | |
43 | X, X, X, X, X, X, X, X, X, Z, Z, X, X, Z, X, X, /* 0.. 15 */ | |
44 | X, X, X, X, X, X, X, X, X, X, X, X, X, X, X, X, /* 16.. 31 */ | |
45 | S, P, P, P, R, P, P, P, R, R, G, R, P, P, R, P, /* 32.. 47 */ | |
46 | D, D, D, D, D, D, D, D, D, D, P, P, P, P, P, G, /* 48.. 63 */ | |
47 | P, A, A, A, A, A, A, A, A, A, A, A, A, A, A, A, /* 64.. 79 */ | |
48 | A, A, A, A, A, A, A, A, A, A, A, G, G, U, R, P, /* 80.. 95 */ | |
49 | P, A, A, A, A, A, A, A, A, A, A, A, A, A, A, A, /* 96..111 */ | |
50 | A, A, A, A, A, A, A, A, A, A, A, R, R, U, P, X, /* 112..127 */ | |
51 | /* Nothing in the 128.. range */ | |
52 | }; | |
53 | ||
54 | #define sane_istest(x,mask) ((sane_ctype[(unsigned char)(x)] & (mask)) != 0) | |
55 | #define is_glob_special(x) sane_istest(x,GIT_GLOB_SPECIAL) | |
56 | ||
57 | typedef unsigned char uchar; | |
58 | ||
59 | /* What character marks an inverted character class? */ | |
60 | #define NEGATE_CLASS '!' | |
61 | #define NEGATE_CLASS2 '^' | |
62 | ||
63 | #define CC_EQ(class, len, litmatch) ((len) == sizeof (litmatch)-1 \ | |
64 | && *(class) == *(litmatch) \ | |
65 | && strncmp((char*)class, litmatch, len) == 0) | |
66 | ||
67 | #if defined STDC_HEADERS || !defined isascii | |
68 | # define ISASCII(c) 1 | |
69 | #else | |
70 | # define ISASCII(c) isascii(c) | |
71 | #endif | |
72 | ||
73 | #ifdef isblank | |
74 | # define ISBLANK(c) (ISASCII(c) && isblank(c)) | |
75 | #else | |
76 | # define ISBLANK(c) ((c) == ' ' || (c) == '\t') | |
77 | #endif | |
78 | ||
79 | #ifdef isgraph | |
80 | # define ISGRAPH(c) (ISASCII(c) && isgraph(c)) | |
81 | #else | |
82 | # define ISGRAPH(c) (ISASCII(c) && isprint(c) && !isspace(c)) | |
83 | #endif | |
84 | ||
85 | #define ISPRINT(c) (ISASCII(c) && isprint(c)) | |
86 | #define ISDIGIT(c) (ISASCII(c) && isdigit(c)) | |
87 | #define ISALNUM(c) (ISASCII(c) && isalnum(c)) | |
88 | #define ISALPHA(c) (ISASCII(c) && isalpha(c)) | |
89 | #define ISCNTRL(c) (ISASCII(c) && iscntrl(c)) | |
90 | #define ISLOWER(c) (ISASCII(c) && islower(c)) | |
91 | #define ISPUNCT(c) (ISASCII(c) && ispunct(c)) | |
92 | #define ISSPACE(c) (ISASCII(c) && isspace(c)) | |
93 | #define ISUPPER(c) (ISASCII(c) && isupper(c)) | |
94 | #define ISXDIGIT(c) (ISASCII(c) && isxdigit(c)) | |
95 | ||
96 | /* Match pattern "p" against "text" */ | |
97 | static int dowild(const uchar *p, const uchar *text, unsigned int flags) | |
98 | { | |
99 | uchar p_ch; | |
100 | const uchar *pattern = p; | |
101 | ||
102 | for ( ; (p_ch = *p) != '\0'; text++, p++) { | |
103 | int matched, match_slash, negated; | |
104 | uchar t_ch, prev_ch; | |
105 | if ((t_ch = *text) == '\0' && p_ch != '*') | |
106 | return WM_ABORT_ALL; | |
107 | if ((flags & WM_CASEFOLD) && ISUPPER(t_ch)) | |
108 | t_ch = tolower(t_ch); | |
109 | if ((flags & WM_CASEFOLD) && ISUPPER(p_ch)) | |
110 | p_ch = tolower(p_ch); | |
111 | switch (p_ch) { | |
112 | case '\\': | |
113 | /* Literal match with following character. Note that the test | |
114 | * in "default" handles the p[1] == '\0' failure case. */ | |
115 | p_ch = *++p; | |
116 | /* FALLTHROUGH */ | |
117 | default: | |
118 | if (t_ch != p_ch) | |
119 | return WM_NOMATCH; | |
120 | continue; | |
121 | case '?': | |
122 | /* Match anything but '/'. */ | |
123 | if ((flags & WM_PATHNAME) && t_ch == '/') | |
124 | return WM_NOMATCH; | |
125 | continue; | |
126 | case '*': | |
127 | if (*++p == '*') { | |
128 | const uchar *prev_p = p - 2; | |
129 | while (*++p == '*') {} | |
130 | if (!(flags & WM_PATHNAME)) | |
131 | /* without WM_PATHNAME, '*' == '**' */ | |
132 | match_slash = 1; | |
133 | else if ((prev_p < pattern || *prev_p == '/') && | |
134 | (*p == '\0' || *p == '/' || | |
135 | (p[0] == '\\' && p[1] == '/'))) { | |
136 | /* | |
137 | * Assuming we already match 'foo/' and are at | |
138 | * <star star slash>, just assume it matches | |
139 | * nothing and go ahead match the rest of the | |
140 | * pattern with the remaining string. This | |
141 | * helps make foo/<*><*>/bar (<> because | |
142 | * otherwise it breaks C comment syntax) match | |
143 | * both foo/bar and foo/a/bar. | |
144 | */ | |
145 | if (p[0] == '/' && | |
146 | dowild(p + 1, text, flags) == WM_MATCH) | |
147 | return WM_MATCH; | |
148 | match_slash = 1; | |
149 | } else /* WM_PATHNAME is set */ | |
150 | match_slash = 0; | |
151 | } else | |
152 | /* without WM_PATHNAME, '*' == '**' */ | |
153 | match_slash = flags & WM_PATHNAME ? 0 : 1; | |
154 | if (*p == '\0') { | |
155 | /* Trailing "**" matches everything. Trailing "*" matches | |
156 | * only if there are no more slash characters. */ | |
157 | if (!match_slash) { | |
158 | if (strchr((char*)text, '/') != NULL) | |
159 | return WM_NOMATCH; | |
160 | } | |
161 | return WM_MATCH; | |
162 | } else if (!match_slash && *p == '/') { | |
163 | /* | |
164 | * _one_ asterisk followed by a slash | |
165 | * with WM_PATHNAME matches the next | |
166 | * directory | |
167 | */ | |
168 | const char *slash = strchr((char*)text, '/'); | |
169 | if (!slash) | |
170 | return WM_NOMATCH; | |
171 | text = (const uchar*)slash; | |
172 | /* the slash is consumed by the top-level for loop */ | |
173 | break; | |
174 | } | |
175 | while (1) { | |
176 | if (t_ch == '\0') | |
177 | break; | |
178 | /* | |
179 | * Try to advance faster when an asterisk is | |
180 | * followed by a literal. We know in this case | |
181 | * that the string before the literal | |
182 | * must belong to "*". | |
183 | * If match_slash is false, do not look past | |
184 | * the first slash as it cannot belong to '*'. | |
185 | */ | |
186 | if (!is_glob_special(*p)) { | |
187 | p_ch = *p; | |
188 | if ((flags & WM_CASEFOLD) && ISUPPER(p_ch)) | |
189 | p_ch = tolower(p_ch); | |
190 | while ((t_ch = *text) != '\0' && | |
191 | (match_slash || t_ch != '/')) { | |
192 | if ((flags & WM_CASEFOLD) && ISUPPER(t_ch)) | |
193 | t_ch = tolower(t_ch); | |
194 | if (t_ch == p_ch) | |
195 | break; | |
196 | text++; | |
197 | } | |
198 | if (t_ch != p_ch) | |
199 | return WM_NOMATCH; | |
200 | } | |
201 | if ((matched = dowild(p, text, flags)) != WM_NOMATCH) { | |
202 | if (!match_slash || matched != WM_ABORT_TO_STARSTAR) | |
203 | return matched; | |
204 | } else if (!match_slash && t_ch == '/') | |
205 | return WM_ABORT_TO_STARSTAR; | |
206 | t_ch = *++text; | |
207 | } | |
208 | return WM_ABORT_ALL; | |
209 | case '[': | |
210 | p_ch = *++p; | |
211 | #ifdef NEGATE_CLASS2 | |
212 | if (p_ch == NEGATE_CLASS2) | |
213 | p_ch = NEGATE_CLASS; | |
214 | #endif | |
215 | /* Assign literal 1/0 because of "matched" comparison. */ | |
216 | negated = p_ch == NEGATE_CLASS ? 1 : 0; | |
217 | if (negated) { | |
218 | /* Inverted character class. */ | |
219 | p_ch = *++p; | |
220 | } | |
221 | prev_ch = 0; | |
222 | matched = 0; | |
223 | do { | |
224 | if (!p_ch) | |
225 | return WM_ABORT_ALL; | |
226 | if (p_ch == '\\') { | |
227 | p_ch = *++p; | |
228 | if (!p_ch) | |
229 | return WM_ABORT_ALL; | |
230 | if (t_ch == p_ch) | |
231 | matched = 1; | |
232 | } else if (p_ch == '-' && prev_ch && p[1] && p[1] != ']') { | |
233 | p_ch = *++p; | |
234 | if (p_ch == '\\') { | |
235 | p_ch = *++p; | |
236 | if (!p_ch) | |
237 | return WM_ABORT_ALL; | |
238 | } | |
239 | if (t_ch <= p_ch && t_ch >= prev_ch) | |
240 | matched = 1; | |
241 | else if ((flags & WM_CASEFOLD) && ISLOWER(t_ch)) { | |
242 | uchar t_ch_upper = toupper(t_ch); | |
243 | if (t_ch_upper <= p_ch && t_ch_upper >= prev_ch) | |
244 | matched = 1; | |
245 | } | |
246 | p_ch = 0; /* This makes "prev_ch" get set to 0. */ | |
247 | } else if (p_ch == '[' && p[1] == ':') { | |
248 | const uchar *s; | |
249 | int i; | |
250 | for (s = p += 2; (p_ch = *p) && p_ch != ']'; p++) {} /*SHARED ITERATOR*/ | |
251 | if (!p_ch) | |
252 | return WM_ABORT_ALL; | |
253 | i = (int)(p - s - 1); | |
254 | if (i < 0 || p[-1] != ':') { | |
255 | /* Didn't find ":]", so treat like a normal set. */ | |
256 | p = s - 2; | |
257 | p_ch = '['; | |
258 | if (t_ch == p_ch) | |
259 | matched = 1; | |
260 | continue; | |
261 | } | |
262 | if (CC_EQ(s,i, "alnum")) { | |
263 | if (ISALNUM(t_ch)) | |
264 | matched = 1; | |
265 | } else if (CC_EQ(s,i, "alpha")) { | |
266 | if (ISALPHA(t_ch)) | |
267 | matched = 1; | |
268 | } else if (CC_EQ(s,i, "blank")) { | |
269 | if (ISBLANK(t_ch)) | |
270 | matched = 1; | |
271 | } else if (CC_EQ(s,i, "cntrl")) { | |
272 | if (ISCNTRL(t_ch)) | |
273 | matched = 1; | |
274 | } else if (CC_EQ(s,i, "digit")) { | |
275 | if (ISDIGIT(t_ch)) | |
276 | matched = 1; | |
277 | } else if (CC_EQ(s,i, "graph")) { | |
278 | if (ISGRAPH(t_ch)) | |
279 | matched = 1; | |
280 | } else if (CC_EQ(s,i, "lower")) { | |
281 | if (ISLOWER(t_ch)) | |
282 | matched = 1; | |
283 | } else if (CC_EQ(s,i, "print")) { | |
284 | if (ISPRINT(t_ch)) | |
285 | matched = 1; | |
286 | } else if (CC_EQ(s,i, "punct")) { | |
287 | if (ISPUNCT(t_ch)) | |
288 | matched = 1; | |
289 | } else if (CC_EQ(s,i, "space")) { | |
290 | if (ISSPACE(t_ch)) | |
291 | matched = 1; | |
292 | } else if (CC_EQ(s,i, "upper")) { | |
293 | if (ISUPPER(t_ch)) | |
294 | matched = 1; | |
295 | else if ((flags & WM_CASEFOLD) && ISLOWER(t_ch)) | |
296 | matched = 1; | |
297 | } else if (CC_EQ(s,i, "xdigit")) { | |
298 | if (ISXDIGIT(t_ch)) | |
299 | matched = 1; | |
300 | } else /* malformed [:class:] string */ | |
301 | return WM_ABORT_ALL; | |
302 | p_ch = 0; /* This makes "prev_ch" get set to 0. */ | |
303 | } else if (t_ch == p_ch) | |
304 | matched = 1; | |
305 | } while (prev_ch = p_ch, (p_ch = *++p) != ']'); | |
306 | if (matched == negated || | |
307 | ((flags & WM_PATHNAME) && t_ch == '/')) | |
308 | return WM_NOMATCH; | |
309 | continue; | |
310 | } | |
311 | } | |
312 | ||
313 | return *text ? WM_NOMATCH : WM_MATCH; | |
314 | } | |
315 | ||
316 | /* Match the "pattern" against the "text" string. */ | |
317 | int wildmatch(const char *pattern, const char *text, unsigned int flags) | |
318 | { | |
319 | return dowild((const uchar*)pattern, (const uchar*)text, flags); | |
320 | } |