[thirdparty/git.git] / attr.c

#include "cache.h"
#include "attr.h"

const char git_attr__true[] = "(builtin)true";
const char git_attr__false[] = "\0(builtin)false";
static const char git_attr__unknown[] = "(builtin)unknown";
#define ATTR__TRUE git_attr__true
#define ATTR__FALSE git_attr__false
#define ATTR__UNSET NULL
#define ATTR__UNKNOWN git_attr__unknown

/*
 * The basic design decision here is that we are not going to have
 * insanely large number of attributes.
 *
 * This is a randomly chosen prime.
 */
#define HASHSIZE 257

#ifndef DEBUG_ATTR
#define DEBUG_ATTR 0
#endif

struct git_attr {
	struct git_attr *next;
	unsigned h;
	int attr_nr;
	char name[FLEX_ARRAY];
};
static int attr_nr;

static struct git_attr_check *check_all_attr;
static struct git_attr *(git_attr_hash[HASHSIZE]);

static unsigned hash_name(const char *name, int namelen)
{
	unsigned val = 0;
	unsigned char c;

	while (namelen--) {
		c = *name++;
		val = ((val << 7) | (val >> 22)) ^ c;
	}
	return val;
}

static int invalid_attr_name(const char *name, int namelen)
{
	/*
	 * Attribute name cannot begin with '-' and from
	 * [-A-Za-z0-9_.].  We'd specifically exclude '=' for now,
	 * as we might later want to allow non-binary value for
	 * attributes, e.g. "*.svg	merge=special-merge-program-for-svg"
	 */
	if (*name == '-')
		return -1;
	while (namelen--) {
		char ch = *name++;
		if (! (ch == '-' || ch == '.' || ch == '_' ||
		       ('0' <= ch && ch <= '9') ||
		       ('a' <= ch && ch <= 'z') ||
		       ('A' <= ch && ch <= 'Z')) )
			return -1;
	}
	return 0;
}

struct git_attr *git_attr(const char *name, int len)
{
	unsigned hval = hash_name(name, len);
	unsigned pos = hval % HASHSIZE;
	struct git_attr *a;

	for (a = git_attr_hash[pos]; a; a = a->next) {
		if (a->h == hval &&
		    !memcmp(a->name, name, len) && !a->name[len])
			return a;
	}

	if (invalid_attr_name(name, len))
		return NULL;

	a = xmalloc(sizeof(*a) + len + 1);
	memcpy(a->name, name, len);
	a->name[len] = 0;
	a->h = hval;
	a->next = git_attr_hash[pos];
	a->attr_nr = attr_nr++;
	git_attr_hash[pos] = a;

	check_all_attr = xrealloc(check_all_attr,
				  sizeof(*check_all_attr) * attr_nr);
	check_all_attr[a->attr_nr].attr = a;
	check_all_attr[a->attr_nr].value = ATTR__UNKNOWN;
	return a;
}

/*
 * .gitattributes file is one line per record, each of which is
 *
 * (1) glob pattern.
 * (2) whitespace
 * (3) whitespace separated list of attribute names, each of which
 *     could be prefixed with '-' to mean "set to false", '!' to mean
 *     "unset".
 */

/* What does a matched pattern decide? */
struct attr_state {
	struct git_attr *attr;
	const char *setto;
};

struct match_attr {
	union {
		char *pattern;
		struct git_attr *attr;
	} u;
	char is_macro;
	unsigned num_attr;
	struct attr_state state[FLEX_ARRAY];
};

static const char blank[] = " \t\r\n";

static const char *parse_attr(const char *src, int lineno, const char *cp,
			      int *num_attr, struct match_attr *res)
{
	const char *ep, *equals;
	int len;

	ep = cp + strcspn(cp, blank);
	equals = strchr(cp, '=');
	if (equals && ep < equals)
		equals = NULL;
	if (equals)
		len = equals - cp;
	else
		len = ep - cp;
	if (!res) {
		if (*cp == '-' || *cp == '!') {
			cp++;
			len--;
		}
		if (invalid_attr_name(cp, len)) {
			fprintf(stderr,
				"%.*s is not a valid attribute name: %s:%d\n",
				len, cp, src, lineno);
			return NULL;
		}
	} else {
		struct attr_state *e;

		e = &(res->state[*num_attr]);
		if (*cp == '-' || *cp == '!') {
			e->setto = (*cp == '-') ? ATTR__FALSE : ATTR__UNSET;
			cp++;
			len--;
		}
		else if (!equals)
			e->setto = ATTR__TRUE;
		else {
			char *value;
			int vallen = ep - equals;
			value = xmalloc(vallen);
			memcpy(value, equals+1, vallen-1);
			value[vallen-1] = 0;
			e->setto = value;
		}
		e->attr = git_attr(cp, len);
	}
	(*num_attr)++;
	return ep + strspn(ep, blank);
}

static struct match_attr *parse_attr_line(const char *line, const char *src,
					  int lineno, int macro_ok)
{
	int namelen;
	int num_attr;
	const char *cp, *name;
	struct match_attr *res = NULL;
	int pass;
	int is_macro;

	cp = line + strspn(line, blank);
	if (!*cp || *cp == '#')
		return NULL;
	name = cp;
	namelen = strcspn(name, blank);
	if (strlen(ATTRIBUTE_MACRO_PREFIX) < namelen &&
	    !prefixcmp(name, ATTRIBUTE_MACRO_PREFIX)) {
		if (!macro_ok) {
			fprintf(stderr, "%s not allowed: %s:%d\n",
				name, src, lineno);
			return NULL;
		}
		is_macro = 1;
		name += strlen(ATTRIBUTE_MACRO_PREFIX);
		name += strspn(name, blank);
		namelen = strcspn(name, blank);
		if (invalid_attr_name(name, namelen)) {
			fprintf(stderr,
				"%.*s is not a valid attribute name: %s:%d\n",
				namelen, name, src, lineno);
			return NULL;
		}
	}
	else
		is_macro = 0;

	for (pass = 0; pass < 2; pass++) {
		/* pass 0 counts and allocates, pass 1 fills */
		num_attr = 0;
		cp = name + namelen;
		cp = cp + strspn(cp, blank);
		while (*cp)
			cp = parse_attr(src, lineno, cp, &num_attr, res);
		if (pass)
			break;
		res = xcalloc(1,
			      sizeof(*res) +
			      sizeof(struct attr_state) * num_attr +
			      (is_macro ? 0 : namelen + 1));
		if (is_macro)
			res->u.attr = git_attr(name, namelen);
		else {
			res->u.pattern = (char*)&(res->state[num_attr]);
			memcpy(res->u.pattern, name, namelen);
			res->u.pattern[namelen] = 0;
		}
		res->is_macro = is_macro;
		res->num_attr = num_attr;
	}
	return res;
}

/*
 * Like info/exclude and .gitignore, the attribute information can
 * come from many places.
 *
 * (1) .gitattribute file of the same directory;
 * (2) .gitattribute file of the parent directory if (1) does not have
 *      any match; this goes recursively upwards, just like .gitignore.
 * (3) $GIT_DIR/info/attributes, which overrides both of the above.
 *
 * In the same file, later entries override the earlier match, so in the
 * global list, we would have entries from info/attributes the earliest
 * (reading the file from top to bottom), .gitattribute of the root
 * directory (again, reading the file from top to bottom) down to the
 * current directory, and then scan the list backwards to find the first match.
 * This is exactly the same as what excluded() does in dir.c to deal with
 * .gitignore
 */

static struct attr_stack {
	struct attr_stack *prev;
	char *origin;
	unsigned num_matches;
	unsigned alloc;
	struct match_attr **attrs;
} *attr_stack;

static void free_attr_elem(struct attr_stack *e)
{
	int i;
	free(e->origin);
	for (i = 0; i < e->num_matches; i++) {
		struct match_attr *a = e->attrs[i];
		int j;
		for (j = 0; j < a->num_attr; j++) {
			const char *setto = a->state[j].setto;
			if (setto == ATTR__TRUE ||
			    setto == ATTR__FALSE ||
			    setto == ATTR__UNSET ||
			    setto == ATTR__UNKNOWN)
				;
			else
				free((char*) setto);
		}
		free(a);
	}
	free(e);
}

static const char *builtin_attr[] = {
	"[attr]binary -diff -crlf",
	NULL,
};

static void handle_attr_line(struct attr_stack *res,
			     const char *line,
			     const char *src,
			     int lineno,
			     int macro_ok)
{
	struct match_attr *a;

	a = parse_attr_line(line, src, lineno, macro_ok);
	if (!a)
		return;
	if (res->alloc <= res->num_matches) {
		res->alloc = alloc_nr(res->num_matches);
		res->attrs = xrealloc(res->attrs,
				      sizeof(struct match_attr *) *
				      res->alloc);
	}
	res->attrs[res->num_matches++] = a;
}

static struct attr_stack *read_attr_from_array(const char **list)
{
	struct attr_stack *res;
	const char *line;
	int lineno = 0;

	res = xcalloc(1, sizeof(*res));
	while ((line = *(list++)) != NULL)
		handle_attr_line(res, line, "[builtin]", ++lineno, 1);
	return res;
}

static struct attr_stack *read_attr_from_file(const char *path, int macro_ok)
{
	FILE *fp = fopen(path, "r");
	struct attr_stack *res;
	char buf[2048];
	int lineno = 0;

	if (!fp)
		return NULL;
	res = xcalloc(1, sizeof(*res));
	while (fgets(buf, sizeof(buf), fp))
		handle_attr_line(res, buf, path, ++lineno, macro_ok);
	fclose(fp);
	return res;
}

static void *read_index_data(const char *path)
{
	int pos, len;
	unsigned long sz;
	enum object_type type;
	void *data;

	len = strlen(path);
	pos = cache_name_pos(path, len);
	if (pos < 0) {
		/*
		 * We might be in the middle of a merge, in which
		 * case we would read stage #2 (ours).
		 */
		int i;
		for (i = -pos - 1;
		     (pos < 0 && i < active_nr &&
		      !strcmp(active_cache[i]->name, path));
		     i++)
			if (ce_stage(active_cache[i]) == 2)
				pos = i;
	}
	if (pos < 0)
		return NULL;
	data = read_sha1_file(active_cache[pos]->sha1, &type, &sz);
	if (!data || type != OBJ_BLOB) {
		free(data);
		return NULL;
	}
	return data;
}

static struct attr_stack *read_attr(const char *path, int macro_ok)
{
	struct attr_stack *res;
	char *buf, *sp;
	int lineno = 0;

	res = read_attr_from_file(path, macro_ok);
	if (res)
		return res;

	res = xcalloc(1, sizeof(*res));

	/*
	 * There is no checked out .gitattributes file there, but
	 * we might have it in the index.  We allow operation in a
	 * sparsely checked out work tree, so read from it.
	 */
	buf = read_index_data(path);
	if (!buf)
		return res;

	for (sp = buf; *sp; ) {
		char *ep;
		int more;
		for (ep = sp; *ep && *ep != '\n'; ep++)
			;
		more = (*ep == '\n');
		*ep = '\0';
		handle_attr_line(res, sp, path, ++lineno, macro_ok);
		sp = ep + more;
	}
	free(buf);
	return res;
}

#if DEBUG_ATTR
static void debug_info(const char *what, struct attr_stack *elem)
{
	fprintf(stderr, "%s: %s\n", what, elem->origin ? elem->origin : "()");
}
static void debug_set(const char *what, const char *match, struct git_attr *attr, void *v)
{
	const char *value = v;

	if (ATTR_TRUE(value))
		value = "set";
	else if (ATTR_FALSE(value))
		value = "unset";
	else if (ATTR_UNSET(value))
		value = "unspecified";

	fprintf(stderr, "%s: %s => %s (%s)\n",
		what, attr->name, (char *) value, match);
}
#define debug_push(a) debug_info("push", (a))
#define debug_pop(a) debug_info("pop", (a))
#else
#define debug_push(a) do { ; } while (0)
#define debug_pop(a) do { ; } while (0)
#define debug_set(a,b,c,d) do { ; } while (0)
#endif

static void bootstrap_attr_stack(void)
{
	if (!attr_stack) {
		struct attr_stack *elem;

		elem = read_attr_from_array(builtin_attr);
		elem->origin = NULL;
		elem->prev = attr_stack;
		attr_stack = elem;

		elem = read_attr(GITATTRIBUTES_FILE, 1);
		elem->origin = strdup("");
		elem->prev = attr_stack;
		attr_stack = elem;
		debug_push(elem);

		elem = read_attr_from_file(git_path(INFOATTRIBUTES_FILE), 1);
		if (!elem)
			elem = xcalloc(1, sizeof(*elem));
		elem->origin = NULL;
		elem->prev = attr_stack;
		attr_stack = elem;
	}
}

static void prepare_attr_stack(const char *path, int dirlen)
{
	struct attr_stack *elem, *info;
	int len;
	char pathbuf[PATH_MAX];

	/*
	 * At the bottom of the attribute stack is the built-in
	 * set of attribute definitions.  Then, contents from
	 * .gitattribute files from directories closer to the
	 * root to the ones in deeper directories are pushed
	 * to the stack.  Finally, at the very top of the stack
	 * we always keep the contents of $GIT_DIR/info/attributes.
	 *
	 * When checking, we use entries from near the top of the
	 * stack, preferring $GIT_DIR/info/attributes, then
	 * .gitattributes in deeper directories to shallower ones,
	 * and finally use the built-in set as the default.
	 */
	if (!attr_stack)
		bootstrap_attr_stack();

	/*
	 * Pop the "info" one that is always at the top of the stack.
	 */
	info = attr_stack;
	attr_stack = info->prev;

	/*
	 * Pop the ones from directories that are not the prefix of
	 * the path we are checking.
	 */
	while (attr_stack && attr_stack->origin) {
		int namelen = strlen(attr_stack->origin);

		elem = attr_stack;
		if (namelen <= dirlen &&
		    !strncmp(elem->origin, path, namelen))
			break;

		debug_pop(elem);
		attr_stack = elem->prev;
		free_attr_elem(elem);
	}

	/*
	 * Read from parent directories and push them down
	 */
	while (1) {
		char *cp;

		len = strlen(attr_stack->origin);
		if (dirlen <= len)
			break;
		memcpy(pathbuf, path, dirlen);
		memcpy(pathbuf + dirlen, "/", 2);
		cp = strchr(pathbuf + len + 1, '/');
		strcpy(cp + 1, GITATTRIBUTES_FILE);
		elem = read_attr(pathbuf, 0);
		*cp = '\0';
		elem->origin = strdup(pathbuf);
		elem->prev = attr_stack;
		attr_stack = elem;
		debug_push(elem);
	}

	/*
	 * Finally push the "info" one at the top of the stack.
	 */
	info->prev = attr_stack;
	attr_stack = info;
}

static int path_matches(const char *pathname, int pathlen,
			const char *pattern,
			const char *base, int baselen)
{
	if (!strchr(pattern, '/')) {
		/* match basename */
		const char *basename = strrchr(pathname, '/');
		basename = basename ? basename + 1 : pathname;
		return (fnmatch(pattern, basename, 0) == 0);
	}
	/*
	 * match with FNM_PATHNAME; the pattern has base implicitly
	 * in front of it.
	 */
	if (*pattern == '/')
		pattern++;
	if (pathlen < baselen ||
	    (baselen && pathname[baselen - 1] != '/') ||
	    strncmp(pathname, base, baselen))
		return 0;
	return fnmatch(pattern, pathname + baselen, FNM_PATHNAME) == 0;
}

static int fill_one(const char *what, struct match_attr *a, int rem)
{
	struct git_attr_check *check = check_all_attr;
	int i;

	for (i = 0; 0 < rem && i < a->num_attr; i++) {
		struct git_attr *attr = a->state[i].attr;
		const char **n = &(check[attr->attr_nr].value);
		const char *v = a->state[i].setto;

		if (*n == ATTR__UNKNOWN) {
			debug_set(what, a->u.pattern, attr, v);
			*n = v;
			rem--;
		}
	}
	return rem;
}

static int fill(const char *path, int pathlen, struct attr_stack *stk, int rem)
{
	int i;
	const char *base = stk->origin ? stk->origin : "";

	for (i = stk->num_matches - 1; 0 < rem && 0 <= i; i--) {
		struct match_attr *a = stk->attrs[i];
		if (a->is_macro)
			continue;
		if (path_matches(path, pathlen,
				 a->u.pattern, base, strlen(base)))
			rem = fill_one("fill", a, rem);
	}
	return rem;
}

static int macroexpand(struct attr_stack *stk, int rem)
{
	int i;
	struct git_attr_check *check = check_all_attr;

	for (i = stk->num_matches - 1; 0 < rem && 0 <= i; i--) {
		struct match_attr *a = stk->attrs[i];
		if (!a->is_macro)
			continue;
		if (check[a->u.attr->attr_nr].value != ATTR__TRUE)
			continue;
		rem = fill_one("expand", a, rem);
	}
	return rem;
}

int git_checkattr(const char *path, int num, struct git_attr_check *check)
{
	struct attr_stack *stk;
	const char *cp;
	int dirlen, pathlen, i, rem;

	bootstrap_attr_stack();
	for (i = 0; i < attr_nr; i++)
		check_all_attr[i].value = ATTR__UNKNOWN;

	pathlen = strlen(path);
	cp = strrchr(path, '/');
	if (!cp)
		dirlen = 0;
	else
		dirlen = cp - path;
	prepare_attr_stack(path, dirlen);
	rem = attr_nr;
	for (stk = attr_stack; 0 < rem && stk; stk = stk->prev)
		rem = fill(path, pathlen, stk, rem);

	for (stk = attr_stack; 0 < rem && stk; stk = stk->prev)
		rem = macroexpand(stk, rem);

	for (i = 0; i < num; i++) {
		const char *value = check_all_attr[check[i].attr->attr_nr].value;
		if (value == ATTR__UNKNOWN)
			value = ATTR__UNSET;
		check[i].value = value;
	}

	return 0;
}
Commit	Line	Data
d0bfd026 JH	1	#include "cache.h"
	2	#include "attr.h"
	3
a5e92abd JH	4	const char git_attr__true[] = "(builtin)true";
	5	const char git_attr__false[] = "\0(builtin)false";
	6	static const char git_attr__unknown[] = "(builtin)unknown";
	7	#define ATTR__TRUE git_attr__true
	8	#define ATTR__FALSE git_attr__false
	9	#define ATTR__UNSET NULL
	10	#define ATTR__UNKNOWN git_attr__unknown
515106fa	11
d0bfd026 JH	12	/*
	13	* The basic design decision here is that we are not going to have
	14	* insanely large number of attributes.
	15	*
	16	* This is a randomly chosen prime.
	17	*/
	18	#define HASHSIZE 257
	19
	20	#ifndef DEBUG_ATTR
	21	#define DEBUG_ATTR 0
	22	#endif
	23
	24	struct git_attr {
	25	struct git_attr *next;
	26	unsigned h;
f48fd688	27	int attr_nr;
d0bfd026 JH	28	char name[FLEX_ARRAY];
d0bfd026 JH	29	};
f48fd688	30	static int attr_nr;
d0bfd026	31
f48fd688	32	static struct git_attr_check *check_all_attr;
d0bfd026 JH	33	static struct git_attr *(git_attr_hash[HASHSIZE]);
	34
	35	static unsigned hash_name(const char *name, int namelen)
	36	{
	37	unsigned val = 0;
	38	unsigned char c;
	39
	40	while (namelen--) {
	41	c = *name++;
	42	val = ((val << 7) \| (val >> 22)) ^ c;
	43	}
	44	return val;
	45	}
	46
e4aee10a JH	47	static int invalid_attr_name(const char *name, int namelen)
	48	{
	49	/*
	50	* Attribute name cannot begin with '-' and from
	51	* [-A-Za-z0-9_.]. We'd specifically exclude '=' for now,
	52	* as we might later want to allow non-binary value for
	53	* attributes, e.g. "*.svg merge=special-merge-program-for-svg"
	54	*/
	55	if (*name == '-')
	56	return -1;
	57	while (namelen--) {
	58	char ch = *name++;
	59	if (! (ch == '-' \|\| ch == '.' \|\| ch == '_' \|\|
	60	('0' <= ch && ch <= '9') \|\|
	61	('a' <= ch && ch <= 'z') \|\|
	62	('A' <= ch && ch <= 'Z')) )
	63	return -1;
	64	}
	65	return 0;
	66	}
	67
d0bfd026 JH	68	struct git_attr git_attr(const char name, int len)
	69	{
	70	unsigned hval = hash_name(name, len);
	71	unsigned pos = hval % HASHSIZE;
	72	struct git_attr *a;
	73
	74	for (a = git_attr_hash[pos]; a; a = a->next) {
	75	if (a->h == hval &&
	76	!memcmp(a->name, name, len) && !a->name[len])
	77	return a;
	78	}
	79
e4aee10a JH	80	if (invalid_attr_name(name, len))
	81	return NULL;
	82
d0bfd026 JH	83	a = xmalloc(sizeof(*a) + len + 1);
	84	memcpy(a->name, name, len);
	85	a->name[len] = 0;
	86	a->h = hval;
	87	a->next = git_attr_hash[pos];
f48fd688	88	a->attr_nr = attr_nr++;
d0bfd026	89	git_attr_hash[pos] = a;
f48fd688 JH	90
	91	check_all_attr = xrealloc(check_all_attr,
	92	sizeof(check_all_attr) attr_nr);
	93	check_all_attr[a->attr_nr].attr = a;
515106fa	94	check_all_attr[a->attr_nr].value = ATTR__UNKNOWN;
d0bfd026 JH	95	return a;
	96	}
	97
	98	/*
	99	* .gitattributes file is one line per record, each of which is
	100	*
	101	* (1) glob pattern.
	102	* (2) whitespace
	103	* (3) whitespace separated list of attribute names, each of which
515106fa JH	104	* could be prefixed with '-' to mean "set to false", '!' to mean
515106fa JH	105	* "unset".
d0bfd026 JH	106	*/
d0bfd026 JH	107
515106fa	108	/* What does a matched pattern decide? */
d0bfd026	109	struct attr_state {
d0bfd026	110	struct git_attr *attr;
a5e92abd	111	const char *setto;
d0bfd026 JH	112	};
	113
	114	struct match_attr {
f48fd688 JH	115	union {
	116	char *pattern;
	117	struct git_attr *attr;
	118	} u;
	119	char is_macro;
d0bfd026 JH	120	unsigned num_attr;
	121	struct attr_state state[FLEX_ARRAY];
	122	};
	123
	124	static const char blank[] = " \t\r\n";
	125
515106fa JH	126	static const char parse_attr(const char src, int lineno, const char *cp,
	127	int num_attr, struct match_attr res)
	128	{
	129	const char ep, equals;
	130	int len;
	131
	132	ep = cp + strcspn(cp, blank);
	133	equals = strchr(cp, '=');
	134	if (equals && ep < equals)
	135	equals = NULL;
	136	if (equals)
	137	len = equals - cp;
	138	else
	139	len = ep - cp;
	140	if (!res) {
	141	if (cp == '-' \|\| cp == '!') {
	142	cp++;
	143	len--;
	144	}
	145	if (invalid_attr_name(cp, len)) {
	146	fprintf(stderr,
	147	"%.*s is not a valid attribute name: %s:%d\n",
	148	len, cp, src, lineno);
	149	return NULL;
	150	}
	151	} else {
	152	struct attr_state *e;
	153
	154	e = &(res->state[*num_attr]);
	155	if (cp == '-' \|\| cp == '!') {
	156	e->setto = (*cp == '-') ? ATTR__FALSE : ATTR__UNSET;
	157	cp++;
	158	len--;
	159	}
	160	else if (!equals)
	161	e->setto = ATTR__TRUE;
	162	else {
	163	char *value;
	164	int vallen = ep - equals;
	165	value = xmalloc(vallen);
	166	memcpy(value, equals+1, vallen-1);
	167	value[vallen-1] = 0;
	168	e->setto = value;
	169	}
	170	e->attr = git_attr(cp, len);
	171	}
	172	(*num_attr)++;
	173	return ep + strspn(ep, blank);
	174	}
	175
f48fd688 JH	176	static struct match_attr parse_attr_line(const char line, const char *src,
f48fd688 JH	177	int lineno, int macro_ok)
d0bfd026 JH	178	{
	179	int namelen;
	180	int num_attr;
	181	const char cp, name;
515106fa	182	struct match_attr *res = NULL;
d0bfd026	183	int pass;
f48fd688	184	int is_macro;
d0bfd026 JH	185
	186	cp = line + strspn(line, blank);
	187	if (!cp \|\| cp == '#')
	188	return NULL;
	189	name = cp;
	190	namelen = strcspn(name, blank);
f48fd688 JH	191	if (strlen(ATTRIBUTE_MACRO_PREFIX) < namelen &&
	192	!prefixcmp(name, ATTRIBUTE_MACRO_PREFIX)) {
	193	if (!macro_ok) {
	194	fprintf(stderr, "%s not allowed: %s:%d\n",
	195	name, src, lineno);
	196	return NULL;
	197	}
	198	is_macro = 1;
	199	name += strlen(ATTRIBUTE_MACRO_PREFIX);
	200	name += strspn(name, blank);
	201	namelen = strcspn(name, blank);
e4aee10a JH	202	if (invalid_attr_name(name, namelen)) {
	203	fprintf(stderr,
	204	"%.*s is not a valid attribute name: %s:%d\n",
	205	namelen, name, src, lineno);
	206	return NULL;
	207	}
f48fd688 JH	208	}
	209	else
	210	is_macro = 0;
d0bfd026 JH	211
	212	for (pass = 0; pass < 2; pass++) {
	213	/* pass 0 counts and allocates, pass 1 fills */
	214	num_attr = 0;
	215	cp = name + namelen;
	216	cp = cp + strspn(cp, blank);
515106fa JH	217	while (*cp)
515106fa JH	218	cp = parse_attr(src, lineno, cp, &num_attr, res);
d0bfd026 JH	219	if (pass)
	220	break;
	221	res = xcalloc(1,
	222	sizeof(*res) +
	223	sizeof(struct attr_state) * num_attr +
f48fd688	224	(is_macro ? 0 : namelen + 1));
515106fa	225	if (is_macro)
f48fd688 JH	226	res->u.attr = git_attr(name, namelen);
	227	else {
	228	res->u.pattern = (char*)&(res->state[num_attr]);
	229	memcpy(res->u.pattern, name, namelen);
	230	res->u.pattern[namelen] = 0;
	231	}
	232	res->is_macro = is_macro;
d0bfd026 JH	233	res->num_attr = num_attr;
	234	}
	235	return res;
	236	}
	237
	238	/*
	239	* Like info/exclude and .gitignore, the attribute information can
	240	* come from many places.
	241	*
	242	* (1) .gitattribute file of the same directory;
515106fa JH	243	* (2) .gitattribute file of the parent directory if (1) does not have
	244	* any match; this goes recursively upwards, just like .gitignore.
	245	* (3) $GIT_DIR/info/attributes, which overrides both of the above.
d0bfd026 JH	246	*
	247	* In the same file, later entries override the earlier match, so in the
	248	* global list, we would have entries from info/attributes the earliest
	249	* (reading the file from top to bottom), .gitattribute of the root
	250	* directory (again, reading the file from top to bottom) down to the
	251	* current directory, and then scan the list backwards to find the first match.
	252	* This is exactly the same as what excluded() does in dir.c to deal with
	253	* .gitignore
	254	*/
	255
	256	static struct attr_stack {
	257	struct attr_stack *prev;
	258	char *origin;
	259	unsigned num_matches;
a4413118	260	unsigned alloc;
d0bfd026 JH	261	struct match_attr **attrs;
	262	} *attr_stack;
	263
	264	static void free_attr_elem(struct attr_stack *e)
	265	{
	266	int i;
	267	free(e->origin);
515106fa JH	268	for (i = 0; i < e->num_matches; i++) {
	269	struct match_attr *a = e->attrs[i];
	270	int j;
	271	for (j = 0; j < a->num_attr; j++) {
a5e92abd	272	const char *setto = a->state[j].setto;
515106fa JH	273	if (setto == ATTR__TRUE \|\|
	274	setto == ATTR__FALSE \|\|
	275	setto == ATTR__UNSET \|\|
	276	setto == ATTR__UNKNOWN)
	277	;
	278	else
a5e92abd	279	free((char*) setto);
515106fa JH	280	}
	281	free(a);
	282	}
d0bfd026 JH	283	free(e);
	284	}
	285
	286	static const char *builtin_attr[] = {
e4aee10a	287	"[attr]binary -diff -crlf",
d0bfd026 JH	288	NULL,
	289	};
	290
a4413118 JH	291	static void handle_attr_line(struct attr_stack *res,
	292	const char *line,
	293	const char *src,
	294	int lineno,
	295	int macro_ok)
	296	{
	297	struct match_attr *a;
	298
	299	a = parse_attr_line(line, src, lineno, macro_ok);
	300	if (!a)
	301	return;
	302	if (res->alloc <= res->num_matches) {
	303	res->alloc = alloc_nr(res->num_matches);
	304	res->attrs = xrealloc(res->attrs,
	305	sizeof(struct match_attr )
	306	res->alloc);
	307	}
	308	res->attrs[res->num_matches++] = a;
	309	}
	310
d0bfd026 JH	311	static struct attr_stack read_attr_from_array(const char *list)
	312	{
	313	struct attr_stack *res;
	314	const char *line;
f48fd688	315	int lineno = 0;
d0bfd026 JH	316
d0bfd026 JH	317	res = xcalloc(1, sizeof(*res));
a4413118 JH	318	while ((line = *(list++)) != NULL)
a4413118 JH	319	handle_attr_line(res, line, "[builtin]", ++lineno, 1);
d0bfd026 JH	320	return res;
	321	}
	322
f48fd688	323	static struct attr_stack read_attr_from_file(const char path, int macro_ok)
d0bfd026	324	{
a4413118	325	FILE *fp = fopen(path, "r");
d0bfd026 JH	326	struct attr_stack *res;
d0bfd026 JH	327	char buf[2048];
f48fd688	328	int lineno = 0;
d0bfd026	329
d0bfd026	330	if (!fp)
a4413118 JH	331	return NULL;
	332	res = xcalloc(1, sizeof(*res));
	333	while (fgets(buf, sizeof(buf), fp))
	334	handle_attr_line(res, buf, path, ++lineno, macro_ok);
	335	fclose(fp);
	336	return res;
	337	}
d0bfd026	338
1a9d7e9b JH	339	static void read_index_data(const char path)
	340	{
	341	int pos, len;
	342	unsigned long sz;
	343	enum object_type type;
	344	void *data;
	345
	346	len = strlen(path);
	347	pos = cache_name_pos(path, len);
	348	if (pos < 0) {
	349	/*
	350	* We might be in the middle of a merge, in which
	351	* case we would read stage #2 (ours).
	352	*/
	353	int i;
	354	for (i = -pos - 1;
	355	(pos < 0 && i < active_nr &&
	356	!strcmp(active_cache[i]->name, path));
	357	i++)
	358	if (ce_stage(active_cache[i]) == 2)
	359	pos = i;
	360	}
	361	if (pos < 0)
	362	return NULL;
	363	data = read_sha1_file(active_cache[pos]->sha1, &type, &sz);
	364	if (!data \|\| type != OBJ_BLOB) {
	365	free(data);
	366	return NULL;
	367	}
	368	return data;
	369	}
	370
a4413118 JH	371	static struct attr_stack read_attr(const char path, int macro_ok)
	372	{
	373	struct attr_stack *res;
1a9d7e9b JH	374	char buf, sp;
1a9d7e9b JH	375	int lineno = 0;
f48fd688	376
a4413118	377	res = read_attr_from_file(path, macro_ok);
1a9d7e9b JH	378	if (res)
	379	return res;
	380
	381	res = xcalloc(1, sizeof(*res));
	382
	383	/*
	384	* There is no checked out .gitattributes file there, but
	385	* we might have it in the index. We allow operation in a
	386	* sparsely checked out work tree, so read from it.
	387	*/
	388	buf = read_index_data(path);
	389	if (!buf)
	390	return res;
	391
	392	for (sp = buf; *sp; ) {
	393	char *ep;
	394	int more;
	395	for (ep = sp; ep && ep != '\n'; ep++)
	396	;
	397	more = (*ep == '\n');
	398	*ep = '\0';
	399	handle_attr_line(res, sp, path, ++lineno, macro_ok);
	400	sp = ep + more;
	401	}
	402	free(buf);
d0bfd026 JH	403	return res;
	404	}
	405
	406	#if DEBUG_ATTR
	407	static void debug_info(const char what, struct attr_stack elem)
	408	{
	409	fprintf(stderr, "%s: %s\n", what, elem->origin ? elem->origin : "()");
	410	}
515106fa	411	static void debug_set(const char what, const char match, struct git_attr attr, void v)
f48fd688	412	{
515106fa JH	413	const char *value = v;
	414
	415	if (ATTR_TRUE(value))
	416	value = "set";
	417	else if (ATTR_FALSE(value))
	418	value = "unset";
	419	else if (ATTR_UNSET(value))
	420	value = "unspecified";
	421
	422	fprintf(stderr, "%s: %s => %s (%s)\n",
	423	what, attr->name, (char *) value, match);
f48fd688	424	}
d0bfd026 JH	425	#define debug_push(a) debug_info("push", (a))
	426	#define debug_pop(a) debug_info("pop", (a))
	427	#else
	428	#define debug_push(a) do { ; } while (0)
	429	#define debug_pop(a) do { ; } while (0)
f48fd688	430	#define debug_set(a,b,c,d) do { ; } while (0)
d0bfd026 JH	431	#endif
d0bfd026 JH	432
f48fd688 JH	433	static void bootstrap_attr_stack(void)
	434	{
	435	if (!attr_stack) {
	436	struct attr_stack *elem;
	437
	438	elem = read_attr_from_array(builtin_attr);
	439	elem->origin = NULL;
	440	elem->prev = attr_stack;
	441	attr_stack = elem;
	442
a4413118	443	elem = read_attr(GITATTRIBUTES_FILE, 1);
f48fd688 JH	444	elem->origin = strdup("");
	445	elem->prev = attr_stack;
	446	attr_stack = elem;
	447	debug_push(elem);
	448
	449	elem = read_attr_from_file(git_path(INFOATTRIBUTES_FILE), 1);
a4413118 JH	450	if (!elem)
a4413118 JH	451	elem = xcalloc(1, sizeof(*elem));
f48fd688 JH	452	elem->origin = NULL;
	453	elem->prev = attr_stack;
	454	attr_stack = elem;
	455	}
	456	}
	457
d0bfd026 JH	458	static void prepare_attr_stack(const char *path, int dirlen)
	459	{
	460	struct attr_stack elem, info;
	461	int len;
	462	char pathbuf[PATH_MAX];
	463
	464	/*
	465	* At the bottom of the attribute stack is the built-in
	466	* set of attribute definitions. Then, contents from
	467	* .gitattribute files from directories closer to the
	468	* root to the ones in deeper directories are pushed
	469	* to the stack. Finally, at the very top of the stack
	470	* we always keep the contents of $GIT_DIR/info/attributes.
	471	*
	472	* When checking, we use entries from near the top of the
	473	* stack, preferring $GIT_DIR/info/attributes, then
	474	* .gitattributes in deeper directories to shallower ones,
	475	* and finally use the built-in set as the default.
	476	*/
f48fd688 JH	477	if (!attr_stack)
f48fd688 JH	478	bootstrap_attr_stack();
d0bfd026 JH	479
	480	/*
	481	* Pop the "info" one that is always at the top of the stack.
	482	*/
	483	info = attr_stack;
	484	attr_stack = info->prev;
	485
	486	/*
	487	* Pop the ones from directories that are not the prefix of
	488	* the path we are checking.
	489	*/
	490	while (attr_stack && attr_stack->origin) {
	491	int namelen = strlen(attr_stack->origin);
	492
	493	elem = attr_stack;
	494	if (namelen <= dirlen &&
	495	!strncmp(elem->origin, path, namelen))
	496	break;
	497
	498	debug_pop(elem);
	499	attr_stack = elem->prev;
	500	free_attr_elem(elem);
	501	}
	502
	503	/*
	504	* Read from parent directories and push them down
	505	*/
	506	while (1) {
	507	char *cp;
	508
	509	len = strlen(attr_stack->origin);
	510	if (dirlen <= len)
	511	break;
	512	memcpy(pathbuf, path, dirlen);
	513	memcpy(pathbuf + dirlen, "/", 2);
	514	cp = strchr(pathbuf + len + 1, '/');
	515	strcpy(cp + 1, GITATTRIBUTES_FILE);
a4413118	516	elem = read_attr(pathbuf, 0);
d0bfd026 JH	517	*cp = '\0';
	518	elem->origin = strdup(pathbuf);
	519	elem->prev = attr_stack;
	520	attr_stack = elem;
	521	debug_push(elem);
	522	}
	523
	524	/*
	525	* Finally push the "info" one at the top of the stack.
	526	*/
	527	info->prev = attr_stack;
	528	attr_stack = info;
	529	}
	530
	531	static int path_matches(const char *pathname, int pathlen,
	532	const char *pattern,
	533	const char *base, int baselen)
	534	{
	535	if (!strchr(pattern, '/')) {
	536	/* match basename */
	537	const char *basename = strrchr(pathname, '/');
	538	basename = basename ? basename + 1 : pathname;
	539	return (fnmatch(pattern, basename, 0) == 0);
	540	}
	541	/*
	542	* match with FNM_PATHNAME; the pattern has base implicitly
	543	* in front of it.
	544	*/
	545	if (*pattern == '/')
	546	pattern++;
	547	if (pathlen < baselen \|\|
	548	(baselen && pathname[baselen - 1] != '/') \|\|
	549	strncmp(pathname, base, baselen))
	550	return 0;
	551	return fnmatch(pattern, pathname + baselen, FNM_PATHNAME) == 0;
	552	}
	553
515106fa JH	554	static int fill_one(const char what, struct match_attr a, int rem)
	555	{
	556	struct git_attr_check *check = check_all_attr;
	557	int i;
	558
	559	for (i = 0; 0 < rem && i < a->num_attr; i++) {
	560	struct git_attr *attr = a->state[i].attr;
a5e92abd JH	561	const char **n = &(check[attr->attr_nr].value);
a5e92abd JH	562	const char *v = a->state[i].setto;
515106fa JH	563
	564	if (*n == ATTR__UNKNOWN) {
	565	debug_set(what, a->u.pattern, attr, v);
	566	*n = v;
	567	rem--;
	568	}
	569	}
	570	return rem;
	571	}
	572
f48fd688	573	static int fill(const char path, int pathlen, struct attr_stack stk, int rem)
d0bfd026	574	{
515106fa	575	int i;
d0bfd026 JH	576	const char *base = stk->origin ? stk->origin : "";
	577
	578	for (i = stk->num_matches - 1; 0 < rem && 0 <= i; i--) {
	579	struct match_attr *a = stk->attrs[i];
f48fd688 JH	580	if (a->is_macro)
f48fd688 JH	581	continue;
d0bfd026	582	if (path_matches(path, pathlen,
515106fa JH	583	a->u.pattern, base, strlen(base)))
515106fa JH	584	rem = fill_one("fill", a, rem);
d0bfd026 JH	585	}
	586	return rem;
	587	}
	588
f48fd688 JH	589	static int macroexpand(struct attr_stack *stk, int rem)
f48fd688 JH	590	{
515106fa	591	int i;
f48fd688 JH	592	struct git_attr_check *check = check_all_attr;
	593
	594	for (i = stk->num_matches - 1; 0 < rem && 0 <= i; i--) {
	595	struct match_attr *a = stk->attrs[i];
	596	if (!a->is_macro)
	597	continue;
515106fa	598	if (check[a->u.attr->attr_nr].value != ATTR__TRUE)
f48fd688	599	continue;
515106fa	600	rem = fill_one("expand", a, rem);
f48fd688 JH	601	}
	602	return rem;
	603	}
	604
d0bfd026 JH	605	int git_checkattr(const char path, int num, struct git_attr_check check)
	606	{
	607	struct attr_stack *stk;
	608	const char *cp;
	609	int dirlen, pathlen, i, rem;
	610
f48fd688 JH	611	bootstrap_attr_stack();
f48fd688 JH	612	for (i = 0; i < attr_nr; i++)
515106fa	613	check_all_attr[i].value = ATTR__UNKNOWN;
d0bfd026 JH	614
	615	pathlen = strlen(path);
	616	cp = strrchr(path, '/');
	617	if (!cp)
	618	dirlen = 0;
	619	else
	620	dirlen = cp - path;
	621	prepare_attr_stack(path, dirlen);
f48fd688 JH	622	rem = attr_nr;
	623	for (stk = attr_stack; 0 < rem && stk; stk = stk->prev)
	624	rem = fill(path, pathlen, stk, rem);
	625
d0bfd026	626	for (stk = attr_stack; 0 < rem && stk; stk = stk->prev)
f48fd688 JH	627	rem = macroexpand(stk, rem);
f48fd688 JH	628
515106fa	629	for (i = 0; i < num; i++) {
a5e92abd	630	const char *value = check_all_attr[check[i].attr->attr_nr].value;
515106fa JH	631	if (value == ATTR__UNKNOWN)
	632	value = ATTR__UNSET;
	633	check[i].value = value;
	634	}
f48fd688	635
d0bfd026 JH	636	return 0;
d0bfd026 JH	637	}