[thirdparty/git.git] / wildmatch.c

/*
**  Do shell-style pattern matching for ?, \, [], and * characters.
**  It is 8bit clean.
**
**  Written by Rich $alz, mirror!rs, Wed Nov 26 19:03:17 EST 1986.
**  Rich $alz is now <rsalz@bbn.com>.
**
**  Modified by Wayne Davison to special-case '/' matching, to make '**'
**  work differently than '*', and to fix the character-class code.
*/

#include "cache.h"
#include "wildmatch.h"

typedef unsigned char uchar;

/* What character marks an inverted character class? */
#define NEGATE_CLASS	'!'
#define NEGATE_CLASS2	'^'

#define CC_EQ(class, len, litmatch) ((len) == sizeof (litmatch)-1 \
				    && *(class) == *(litmatch) \
				    && strncmp((char*)class, litmatch, len) == 0)

#if defined STDC_HEADERS || !defined isascii
# define ISASCII(c) 1
#else
# define ISASCII(c) isascii(c)
#endif

#ifdef isblank
# define ISBLANK(c) (ISASCII(c) && isblank(c))
#else
# define ISBLANK(c) ((c) == ' ' || (c) == '\t')
#endif

#ifdef isgraph
# define ISGRAPH(c) (ISASCII(c) && isgraph(c))
#else
# define ISGRAPH(c) (ISASCII(c) && isprint(c) && !isspace(c))
#endif

#define ISPRINT(c) (ISASCII(c) && isprint(c))
#define ISDIGIT(c) (ISASCII(c) && isdigit(c))
#define ISALNUM(c) (ISASCII(c) && isalnum(c))
#define ISALPHA(c) (ISASCII(c) && isalpha(c))
#define ISCNTRL(c) (ISASCII(c) && iscntrl(c))
#define ISLOWER(c) (ISASCII(c) && islower(c))
#define ISPUNCT(c) (ISASCII(c) && ispunct(c))
#define ISSPACE(c) (ISASCII(c) && isspace(c))
#define ISUPPER(c) (ISASCII(c) && isupper(c))
#define ISXDIGIT(c) (ISASCII(c) && isxdigit(c))

/* Match pattern "p" against "text" */
static int dowild(const uchar *p, const uchar *text, unsigned int flags)
{
	uchar p_ch;
	const uchar *pattern = p;

	for ( ; (p_ch = *p) != '\0'; text++, p++) {
		int matched, match_slash, negated;
		uchar t_ch, prev_ch;
		if ((t_ch = *text) == '\0' && p_ch != '*')
			return WM_ABORT_ALL;
		if ((flags & WM_CASEFOLD) && ISUPPER(t_ch))
			t_ch = tolower(t_ch);
		if ((flags & WM_CASEFOLD) && ISUPPER(p_ch))
			p_ch = tolower(p_ch);
		switch (p_ch) {
		case '\\':
			/* Literal match with following character.  Note that the test
			 * in "default" handles the p[1] == '\0' failure case. */
			p_ch = *++p;
			/* FALLTHROUGH */
		default:
			if (t_ch != p_ch)
				return WM_NOMATCH;
			continue;
		case '?':
			/* Match anything but '/'. */
			if ((flags & WM_PATHNAME) && t_ch == '/')
				return WM_NOMATCH;
			continue;
		case '*':
			if (*++p == '*') {
				const uchar *prev_p = p - 2;
				while (*++p == '*') {}
				if (!(flags & WM_PATHNAME))
					/* without WM_PATHNAME, '*' == '**' */
					match_slash = 1;
				else if ((prev_p < pattern || *prev_p == '/') &&
				    (*p == '\0' || *p == '/' ||
				     (p[0] == '\\' && p[1] == '/'))) {
					/*
					 * Assuming we already match 'foo/' and are at
					 * <star star slash>, just assume it matches
					 * nothing and go ahead match the rest of the
					 * pattern with the remaining string. This
					 * helps make foo/<*><*>/bar (<> because
					 * otherwise it breaks C comment syntax) match
					 * both foo/bar and foo/a/bar.
					 */
					if (p[0] == '/' &&
					    dowild(p + 1, text, flags) == WM_MATCH)
						return WM_MATCH;
					match_slash = 1;
				} else /* WM_PATHNAME is set */
					match_slash = 0;
			} else
				/* without WM_PATHNAME, '*' == '**' */
				match_slash = flags & WM_PATHNAME ? 0 : 1;
			if (*p == '\0') {
				/* Trailing "**" matches everything.  Trailing "*" matches
				 * only if there are no more slash characters. */
				if (!match_slash) {
					if (strchr((char *)text, '/'))
						return WM_ABORT_TO_STARSTAR;
				}
				return WM_MATCH;
			} else if (!match_slash && *p == '/') {
				/*
				 * _one_ asterisk followed by a slash
				 * with WM_PATHNAME matches the next
				 * directory
				 */
				const char *slash = strchr((char*)text, '/');
				if (!slash)
					return WM_ABORT_ALL;
				text = (const uchar*)slash;
				/* the slash is consumed by the top-level for loop */
				break;
			}
			while (1) {
				if (t_ch == '\0')
					break;
				/*
				 * Try to advance faster when an asterisk is
				 * followed by a literal. We know in this case
				 * that the string before the literal
				 * must belong to "*".
				 * If match_slash is false, do not look past
				 * the first slash as it cannot belong to '*'.
				 */
				if (!is_glob_special(*p)) {
					p_ch = *p;
					if ((flags & WM_CASEFOLD) && ISUPPER(p_ch))
						p_ch = tolower(p_ch);
					while ((t_ch = *text) != '\0' &&
					       (match_slash || t_ch != '/')) {
						if ((flags & WM_CASEFOLD) && ISUPPER(t_ch))
							t_ch = tolower(t_ch);
						if (t_ch == p_ch)
							break;
						text++;
					}
					if (t_ch != p_ch) {
						if (match_slash)
							return WM_ABORT_ALL;
						else
							return WM_ABORT_TO_STARSTAR;
					}
				}
				if ((matched = dowild(p, text, flags)) != WM_NOMATCH) {
					if (!match_slash || matched != WM_ABORT_TO_STARSTAR)
						return matched;
				} else if (!match_slash && t_ch == '/')
					return WM_ABORT_TO_STARSTAR;
				t_ch = *++text;
			}
			return WM_ABORT_ALL;
		case '[':
			p_ch = *++p;
#ifdef NEGATE_CLASS2
			if (p_ch == NEGATE_CLASS2)
				p_ch = NEGATE_CLASS;
#endif
			/* Assign literal 1/0 because of "matched" comparison. */
			negated = p_ch == NEGATE_CLASS ? 1 : 0;
			if (negated) {
				/* Inverted character class. */
				p_ch = *++p;
			}
			prev_ch = 0;
			matched = 0;
			do {
				if (!p_ch)
					return WM_ABORT_ALL;
				if (p_ch == '\\') {
					p_ch = *++p;
					if (!p_ch)
						return WM_ABORT_ALL;
					if (t_ch == p_ch)
						matched = 1;
				} else if (p_ch == '-' && prev_ch && p[1] && p[1] != ']') {
					p_ch = *++p;
					if (p_ch == '\\') {
						p_ch = *++p;
						if (!p_ch)
							return WM_ABORT_ALL;
					}
					if (t_ch <= p_ch && t_ch >= prev_ch)
						matched = 1;
					else if ((flags & WM_CASEFOLD) && ISLOWER(t_ch)) {
						uchar t_ch_upper = toupper(t_ch);
						if (t_ch_upper <= p_ch && t_ch_upper >= prev_ch)
							matched = 1;
					}
					p_ch = 0; /* This makes "prev_ch" get set to 0. */
				} else if (p_ch == '[' && p[1] == ':') {
					const uchar *s;
					int i;
					for (s = p += 2; (p_ch = *p) && p_ch != ']'; p++) {} /*SHARED ITERATOR*/
					if (!p_ch)
						return WM_ABORT_ALL;
					i = p - s - 1;
					if (i < 0 || p[-1] != ':') {
						/* Didn't find ":]", so treat like a normal set. */
						p = s - 2;
						p_ch = '[';
						if (t_ch == p_ch)
							matched = 1;
						continue;
					}
					if (CC_EQ(s,i, "alnum")) {
						if (ISALNUM(t_ch))
							matched = 1;
					} else if (CC_EQ(s,i, "alpha")) {
						if (ISALPHA(t_ch))
							matched = 1;
					} else if (CC_EQ(s,i, "blank")) {
						if (ISBLANK(t_ch))
							matched = 1;
					} else if (CC_EQ(s,i, "cntrl")) {
						if (ISCNTRL(t_ch))
							matched = 1;
					} else if (CC_EQ(s,i, "digit")) {
						if (ISDIGIT(t_ch))
							matched = 1;
					} else if (CC_EQ(s,i, "graph")) {
						if (ISGRAPH(t_ch))
							matched = 1;
					} else if (CC_EQ(s,i, "lower")) {
						if (ISLOWER(t_ch))
							matched = 1;
					} else if (CC_EQ(s,i, "print")) {
						if (ISPRINT(t_ch))
							matched = 1;
					} else if (CC_EQ(s,i, "punct")) {
						if (ISPUNCT(t_ch))
							matched = 1;
					} else if (CC_EQ(s,i, "space")) {
						if (ISSPACE(t_ch))
							matched = 1;
					} else if (CC_EQ(s,i, "upper")) {
						if (ISUPPER(t_ch))
							matched = 1;
						else if ((flags & WM_CASEFOLD) && ISLOWER(t_ch))
							matched = 1;
					} else if (CC_EQ(s,i, "xdigit")) {
						if (ISXDIGIT(t_ch))
							matched = 1;
					} else /* malformed [:class:] string */
						return WM_ABORT_ALL;
					p_ch = 0; /* This makes "prev_ch" get set to 0. */
				} else if (t_ch == p_ch)
					matched = 1;
			} while (prev_ch = p_ch, (p_ch = *++p) != ']');
			if (matched == negated ||
			    ((flags & WM_PATHNAME) && t_ch == '/'))
				return WM_NOMATCH;
			continue;
		}
	}

	return *text ? WM_NOMATCH : WM_MATCH;
}

/* Match the "pattern" against the "text" string. */
int wildmatch(const char *pattern, const char *text, unsigned int flags)
{
	return dowild((const uchar*)pattern, (const uchar*)text, flags);
}
Commit	Line	Data
5230f605 NTND	1	/*
	2	** Do shell-style pattern matching for ?, \, [], and * characters.
	3	** It is 8bit clean.
	4	**
	5	** Written by Rich $alz, mirror!rs, Wed Nov 26 19:03:17 EST 1986.
	6	** Rich $alz is now <rsalz@bbn.com>.
	7	**
	8	Modified by Wayne Davison to special-case '/' matching, to make ''
	9	** work differently than '*', and to fix the character-class code.
	10	*/
	11
feabcc17 NTND	12	#include "cache.h"
	13	#include "wildmatch.h"
	14
	15	typedef unsigned char uchar;
5230f605 NTND	16
	17	/* What character marks an inverted character class? */
	18	#define NEGATE_CLASS '!'
	19	#define NEGATE_CLASS2 '^'
	20
5230f605 NTND	21	#define CC_EQ(class, len, litmatch) ((len) == sizeof (litmatch)-1 \
	22	&& (class) == (litmatch) \
	23	&& strncmp((char*)class, litmatch, len) == 0)
	24
	25	#if defined STDC_HEADERS \|\| !defined isascii
	26	# define ISASCII(c) 1
	27	#else
	28	# define ISASCII(c) isascii(c)
	29	#endif
	30
	31	#ifdef isblank
	32	# define ISBLANK(c) (ISASCII(c) && isblank(c))
	33	#else
	34	# define ISBLANK(c) ((c) == ' ' \|\| (c) == '\t')
	35	#endif
	36
	37	#ifdef isgraph
	38	# define ISGRAPH(c) (ISASCII(c) && isgraph(c))
	39	#else
	40	# define ISGRAPH(c) (ISASCII(c) && isprint(c) && !isspace(c))
	41	#endif
	42
	43	#define ISPRINT(c) (ISASCII(c) && isprint(c))
	44	#define ISDIGIT(c) (ISASCII(c) && isdigit(c))
	45	#define ISALNUM(c) (ISASCII(c) && isalnum(c))
	46	#define ISALPHA(c) (ISASCII(c) && isalpha(c))
	47	#define ISCNTRL(c) (ISASCII(c) && iscntrl(c))
	48	#define ISLOWER(c) (ISASCII(c) && islower(c))
	49	#define ISPUNCT(c) (ISASCII(c) && ispunct(c))
	50	#define ISSPACE(c) (ISASCII(c) && isspace(c))
	51	#define ISUPPER(c) (ISASCII(c) && isupper(c))
	52	#define ISXDIGIT(c) (ISASCII(c) && isxdigit(c))
	53
b0e02876	54	/* Match pattern "p" against "text" */
0c528168	55	static int dowild(const uchar p, const uchar text, unsigned int flags)
5230f605	56	{
327f2f3e	57	uchar p_ch;
3a078dec	58	const uchar *pattern = p;
5230f605	59
327f2f3e	60	for ( ; (p_ch = *p) != '\0'; text++, p++) {
b6a3d335	61	int matched, match_slash, negated;
327f2f3e NTND	62	uchar t_ch, prev_ch;
327f2f3e NTND	63	if ((t_ch = text) == '\0' && p_ch != '')
9b3497ca	64	return WM_ABORT_ALL;
0c528168	65	if ((flags & WM_CASEFOLD) && ISUPPER(t_ch))
327f2f3e	66	t_ch = tolower(t_ch);
0c528168	67	if ((flags & WM_CASEFOLD) && ISUPPER(p_ch))
164bf83a	68	p_ch = tolower(p_ch);
327f2f3e NTND	69	switch (p_ch) {
	70	case '\\':
	71	/* Literal match with following character. Note that the test
	72	* in "default" handles the p[1] == '\0' failure case. */
5230f605	73	p_ch = *++p;
327f2f3e NTND	74	/* FALLTHROUGH */
	75	default:
	76	if (t_ch != p_ch)
9b3497ca	77	return WM_NOMATCH;
327f2f3e NTND	78	continue;
	79	case '?':
	80	/* Match anything but '/'. */
c41244e7	81	if ((flags & WM_PATHNAME) && t_ch == '/')
9b3497ca	82	return WM_NOMATCH;
5230f605	83	continue;
327f2f3e NTND	84	case '*':
327f2f3e NTND	85	if (++p == '') {
40bbee0a	86	const uchar *prev_p = p - 2;
327f2f3e	87	while (++p == '') {}
c41244e7 NTND	88	if (!(flags & WM_PATHNAME))
	89	/* without WM_PATHNAME, '' == '' /
	90	match_slash = 1;
	91	else if ((prev_p < pattern \|\| *prev_p == '/') &&
40bbee0a NTND	92	(p == '\0' \|\| p == '/' \|\|
40bbee0a NTND	93	(p[0] == '\\' && p[1] == '/'))) {
4c251e5c NTND	94	/*
	95	* Assuming we already match 'foo/' and are at
	96	* <star star slash>, just assume it matches
	97	* nothing and go ahead match the rest of the
	98	* pattern with the remaining string. This
	99	* helps make foo/<><>/bar (<> because
	100	* otherwise it breaks C comment syntax) match
	101	* both foo/bar and foo/a/bar.
	102	*/
	103	if (p[0] == '/' &&
0c528168	104	dowild(p + 1, text, flags) == WM_MATCH)
9b3497ca NTND	105	return WM_MATCH;
9b3497ca NTND	106	match_slash = 1;
e5bbe09e NTND	107	} else /* WM_PATHNAME is set */
e5bbe09e NTND	108	match_slash = 0;
327f2f3e	109	} else
c41244e7 NTND	110	/* without WM_PATHNAME, '' == '' /
c41244e7 NTND	111	match_slash = flags & WM_PATHNAME ? 0 : 1;
327f2f3e NTND	112	if (*p == '\0') {
	113	/* Trailing "*" matches everything. Trailing "" matches
	114	* only if there are no more slash characters. */
b6a3d335	115	if (!match_slash) {
afe8a907	116	if (strchr((char *)text, '/'))
1f2e05f0	117	return WM_ABORT_TO_STARSTAR;
327f2f3e	118	}
9b3497ca	119	return WM_MATCH;
46983441 NTND	120	} else if (!match_slash && *p == '/') {
	121	/*
	122	* _one_ asterisk followed by a slash
	123	* with WM_PATHNAME matches the next
	124	* directory
	125	*/
	126	const char slash = strchr((char)text, '/');
	127	if (!slash)
1f2e05f0	128	return WM_ABORT_ALL;
46983441 NTND	129	text = (const uchar*)slash;
	130	/* the slash is consumed by the top-level for loop */
	131	break;
327f2f3e NTND	132	}
	133	while (1) {
	134	if (t_ch == '\0')
	135	break;
6f1a31f0 NTND	136	/*
	137	* Try to advance faster when an asterisk is
	138	* followed by a literal. We know in this case
832c0e5e	139	* that the string before the literal
6f1a31f0 NTND	140	* must belong to "*".
	141	* If match_slash is false, do not look past
	142	* the first slash as it cannot belong to '*'.
	143	*/
	144	if (!is_glob_special(*p)) {
	145	p_ch = *p;
	146	if ((flags & WM_CASEFOLD) && ISUPPER(p_ch))
	147	p_ch = tolower(p_ch);
	148	while ((t_ch = *text) != '\0' &&
	149	(match_slash \|\| t_ch != '/')) {
	150	if ((flags & WM_CASEFOLD) && ISUPPER(t_ch))
	151	t_ch = tolower(t_ch);
	152	if (t_ch == p_ch)
	153	break;
	154	text++;
	155	}
1f2e05f0 PW	156	if (t_ch != p_ch) {
	157	if (match_slash)
	158	return WM_ABORT_ALL;
	159	else
	160	return WM_ABORT_TO_STARSTAR;
	161	}
6f1a31f0	162	}
0c528168	163	if ((matched = dowild(p, text, flags)) != WM_NOMATCH) {
9b3497ca	164	if (!match_slash \|\| matched != WM_ABORT_TO_STARSTAR)
327f2f3e	165	return matched;
b6a3d335	166	} else if (!match_slash && t_ch == '/')
9b3497ca	167	return WM_ABORT_TO_STARSTAR;
327f2f3e NTND	168	t_ch = *++text;
327f2f3e NTND	169	}
9b3497ca	170	return WM_ABORT_ALL;
327f2f3e NTND	171	case '[':
	172	p_ch = *++p;
	173	#ifdef NEGATE_CLASS2
	174	if (p_ch == NEGATE_CLASS2)
	175	p_ch = NEGATE_CLASS;
	176	#endif
9b3497ca NTND	177	/* Assign literal 1/0 because of "matched" comparison. */
9b3497ca NTND	178	negated = p_ch == NEGATE_CLASS ? 1 : 0;
b6a3d335	179	if (negated) {
327f2f3e NTND	180	/* Inverted character class. */
	181	p_ch = *++p;
	182	}
	183	prev_ch = 0;
9b3497ca	184	matched = 0;
327f2f3e NTND	185	do {
327f2f3e NTND	186	if (!p_ch)
9b3497ca	187	return WM_ABORT_ALL;
327f2f3e NTND	188	if (p_ch == '\\') {
	189	p_ch = *++p;
	190	if (!p_ch)
9b3497ca	191	return WM_ABORT_ALL;
327f2f3e	192	if (t_ch == p_ch)
9b3497ca	193	matched = 1;
327f2f3e NTND	194	} else if (p_ch == '-' && prev_ch && p[1] && p[1] != ']') {
	195	p_ch = *++p;
	196	if (p_ch == '\\') {
	197	p_ch = *++p;
	198	if (!p_ch)
9b3497ca	199	return WM_ABORT_ALL;
327f2f3e NTND	200	}
327f2f3e NTND	201	if (t_ch <= p_ch && t_ch >= prev_ch)
9b3497ca	202	matched = 1;
b79c0c37 AR	203	else if ((flags & WM_CASEFOLD) && ISLOWER(t_ch)) {
	204	uchar t_ch_upper = toupper(t_ch);
	205	if (t_ch_upper <= p_ch && t_ch_upper >= prev_ch)
	206	matched = 1;
	207	}
327f2f3e NTND	208	p_ch = 0; /* This makes "prev_ch" get set to 0. */
	209	} else if (p_ch == '[' && p[1] == ':') {
	210	const uchar *s;
	211	int i;
	212	for (s = p += 2; (p_ch = p) && p_ch != ']'; p++) {} /SHARED ITERATOR*/
	213	if (!p_ch)
9b3497ca	214	return WM_ABORT_ALL;
327f2f3e NTND	215	i = p - s - 1;
	216	if (i < 0 \|\| p[-1] != ':') {
	217	/* Didn't find ":]", so treat like a normal set. */
	218	p = s - 2;
	219	p_ch = '[';
	220	if (t_ch == p_ch)
9b3497ca	221	matched = 1;
327f2f3e NTND	222	continue;
	223	}
	224	if (CC_EQ(s,i, "alnum")) {
	225	if (ISALNUM(t_ch))
9b3497ca	226	matched = 1;
327f2f3e NTND	227	} else if (CC_EQ(s,i, "alpha")) {
327f2f3e NTND	228	if (ISALPHA(t_ch))
9b3497ca	229	matched = 1;
327f2f3e NTND	230	} else if (CC_EQ(s,i, "blank")) {
327f2f3e NTND	231	if (ISBLANK(t_ch))
9b3497ca	232	matched = 1;
327f2f3e NTND	233	} else if (CC_EQ(s,i, "cntrl")) {
327f2f3e NTND	234	if (ISCNTRL(t_ch))
9b3497ca	235	matched = 1;
327f2f3e NTND	236	} else if (CC_EQ(s,i, "digit")) {
327f2f3e NTND	237	if (ISDIGIT(t_ch))
9b3497ca	238	matched = 1;
327f2f3e NTND	239	} else if (CC_EQ(s,i, "graph")) {
327f2f3e NTND	240	if (ISGRAPH(t_ch))
9b3497ca	241	matched = 1;
327f2f3e NTND	242	} else if (CC_EQ(s,i, "lower")) {
327f2f3e NTND	243	if (ISLOWER(t_ch))
9b3497ca	244	matched = 1;
327f2f3e NTND	245	} else if (CC_EQ(s,i, "print")) {
327f2f3e NTND	246	if (ISPRINT(t_ch))
9b3497ca	247	matched = 1;
327f2f3e NTND	248	} else if (CC_EQ(s,i, "punct")) {
327f2f3e NTND	249	if (ISPUNCT(t_ch))
9b3497ca	250	matched = 1;
327f2f3e NTND	251	} else if (CC_EQ(s,i, "space")) {
327f2f3e NTND	252	if (ISSPACE(t_ch))
9b3497ca	253	matched = 1;
327f2f3e NTND	254	} else if (CC_EQ(s,i, "upper")) {
327f2f3e NTND	255	if (ISUPPER(t_ch))
9b3497ca	256	matched = 1;
b79c0c37 AR	257	else if ((flags & WM_CASEFOLD) && ISLOWER(t_ch))
b79c0c37 AR	258	matched = 1;
327f2f3e NTND	259	} else if (CC_EQ(s,i, "xdigit")) {
327f2f3e NTND	260	if (ISXDIGIT(t_ch))
9b3497ca	261	matched = 1;
327f2f3e	262	} else /* malformed [:class:] string */
9b3497ca	263	return WM_ABORT_ALL;
327f2f3e NTND	264	p_ch = 0; /* This makes "prev_ch" get set to 0. */
327f2f3e NTND	265	} else if (t_ch == p_ch)
9b3497ca	266	matched = 1;
327f2f3e	267	} while (prev_ch = p_ch, (p_ch = *++p) != ']');
c41244e7 NTND	268	if (matched == negated \|\|
c41244e7 NTND	269	((flags & WM_PATHNAME) && t_ch == '/'))
9b3497ca	270	return WM_NOMATCH;
327f2f3e NTND	271	continue;
327f2f3e NTND	272	}
5230f605	273	}
5230f605	274
9b3497ca	275	return *text ? WM_NOMATCH : WM_MATCH;
5230f605 NTND	276	}
	277
	278	/* Match the "pattern" against the "text" string. */
55d34269	279	int wildmatch(const char pattern, const char text, unsigned int flags)
5230f605	280	{
0c528168	281	return dowild((const uchar)pattern, (const uchar)text, flags);
5230f605	282	}