[thirdparty/git.git] / match-trees.c

#include "cache.h"
#include "tree.h"
#include "tree-walk.h"

static int score_missing(unsigned mode, const char *path)
{
	int score;

	if (S_ISDIR(mode))
		score = -1000;
	else if (S_ISLNK(mode))
		score = -500;
	else
		score = -50;
	return score;
}

static int score_differs(unsigned mode1, unsigned mode2, const char *path)
{
	int score;

	if (S_ISDIR(mode1) != S_ISDIR(mode2))
		score = -100;
	else if (S_ISLNK(mode1) != S_ISLNK(mode2))
		score = -50;
	else
		score = -5;
	return score;
}

static int score_matches(unsigned mode1, unsigned mode2, const char *path)
{
	int score;

	/* Heh, we found SHA-1 collisions between different kind of objects */
	if (S_ISDIR(mode1) != S_ISDIR(mode2))
		score = -100;
	else if (S_ISLNK(mode1) != S_ISLNK(mode2))
		score = -50;

	else if (S_ISDIR(mode1))
		score = 1000;
	else if (S_ISLNK(mode1))
		score = 500;
	else
		score = 250;
	return score;
}

static void *fill_tree_desc_strict(struct tree_desc *desc,
				   const struct object_id *hash)
{
	void *buffer;
	enum object_type type;
	unsigned long size;

	buffer = read_sha1_file(hash->hash, &type, &size);
	if (!buffer)
		die("unable to read tree (%s)", oid_to_hex(hash));
	if (type != OBJ_TREE)
		die("%s is not a tree", oid_to_hex(hash));
	init_tree_desc(desc, buffer, size);
	return buffer;
}

static int base_name_entries_compare(const struct name_entry *a,
				     const struct name_entry *b)
{
	return base_name_compare(a->path, tree_entry_len(a), a->mode,
				 b->path, tree_entry_len(b), b->mode);
}

/*
 * Inspect two trees, and give a score that tells how similar they are.
 */
static int score_trees(const struct object_id *hash1, const struct object_id *hash2)
{
	struct tree_desc one;
	struct tree_desc two;
	void *one_buf = fill_tree_desc_strict(&one, hash1);
	void *two_buf = fill_tree_desc_strict(&two, hash2);
	int score = 0;

	for (;;) {
		struct name_entry e1, e2;
		int got_entry_from_one = tree_entry(&one, &e1);
		int got_entry_from_two = tree_entry(&two, &e2);
		int cmp;

		if (got_entry_from_one && got_entry_from_two)
			cmp = base_name_entries_compare(&e1, &e2);
		else if (got_entry_from_one)
			/* two lacks this entry */
			cmp = -1;
		else if (got_entry_from_two)
			/* two has more entries */
			cmp = 1;
		else
			break;

		if (cmp < 0)
			/* path1 does not appear in two */
			score += score_missing(e1.mode, e1.path);
		else if (cmp > 0)
			/* path2 does not appear in one */
			score += score_missing(e2.mode, e2.path);
		else if (oidcmp(e1.oid, e2.oid))
			/* they are different */
			score += score_differs(e1.mode, e2.mode, e1.path);
		else
			/* same subtree or blob */
			score += score_matches(e1.mode, e2.mode, e1.path);
	}
	free(one_buf);
	free(two_buf);
	return score;
}

/*
 * Match one itself and its subtrees with two and pick the best match.
 */
static void match_trees(const struct object_id *hash1,
			const struct object_id *hash2,
			int *best_score,
			char **best_match,
			const char *base,
			int recurse_limit)
{
	struct tree_desc one;
	void *one_buf = fill_tree_desc_strict(&one, hash1);

	while (one.size) {
		const char *path;
		const struct object_id *elem;
		unsigned mode;
		int score;

		elem = tree_entry_extract(&one, &path, &mode);
		if (!S_ISDIR(mode))
			goto next;
		score = score_trees(elem, hash2);
		if (*best_score < score) {
			free(*best_match);
			*best_match = xstrfmt("%s%s", base, path);
			*best_score = score;
		}
		if (recurse_limit) {
			char *newbase = xstrfmt("%s%s/", base, path);
			match_trees(elem, hash2, best_score, best_match,
				    newbase, recurse_limit - 1);
			free(newbase);
		}

	next:
		update_tree_entry(&one);
	}
	free(one_buf);
}

/*
 * A tree "oid1" has a subdirectory at "prefix".  Come up with a tree object by
 * replacing it with another tree "oid2".
 */
static int splice_tree(const struct object_id *oid1, const char *prefix,
		       const struct object_id *oid2, struct object_id *result)
{
	char *subpath;
	int toplen;
	char *buf;
	unsigned long sz;
	struct tree_desc desc;
	struct object_id *rewrite_here;
	const struct object_id *rewrite_with;
	struct object_id subtree;
	enum object_type type;
	int status;

	subpath = strchrnul(prefix, '/');
	toplen = subpath - prefix;
	if (*subpath)
		subpath++;

	buf = read_sha1_file(oid1->hash, &type, &sz);
	if (!buf)
		die("cannot read tree %s", oid_to_hex(oid1));
	init_tree_desc(&desc, buf, sz);

	rewrite_here = NULL;
	while (desc.size) {
		const char *name;
		unsigned mode;
		const struct object_id *oid;

		oid = tree_entry_extract(&desc, &name, &mode);
		if (strlen(name) == toplen &&
		    !memcmp(name, prefix, toplen)) {
			if (!S_ISDIR(mode))
				die("entry %s in tree %s is not a tree", name,
				    oid_to_hex(oid1));
			rewrite_here = (struct object_id *)oid;
			break;
		}
		update_tree_entry(&desc);
	}
	if (!rewrite_here)
		die("entry %.*s not found in tree %s", toplen, prefix,
		    oid_to_hex(oid1));
	if (*subpath) {
		status = splice_tree(rewrite_here, subpath, oid2, &subtree);
		if (status)
			return status;
		rewrite_with = &subtree;
	} else {
		rewrite_with = oid2;
	}
	oidcpy(rewrite_here, rewrite_with);
	status = write_object_file(buf, sz, tree_type, result);
	free(buf);
	return status;
}

/*
 * We are trying to come up with a merge between one and two that
 * results in a tree shape similar to one.  The tree two might
 * correspond to a subtree of one, in which case it needs to be
 * shifted down by prefixing otherwise empty directories.  On the
 * other hand, it could cover tree one and we might need to pick a
 * subtree of it.
 */
void shift_tree(const struct object_id *hash1,
		const struct object_id *hash2,
		struct object_id *shifted,
		int depth_limit)
{
	char *add_prefix;
	char *del_prefix;
	int add_score, del_score;

	/*
	 * NEEDSWORK: this limits the recursion depth to hardcoded
	 * value '2' to avoid excessive overhead.
	 */
	if (!depth_limit)
		depth_limit = 2;

	add_score = del_score = score_trees(hash1, hash2);
	add_prefix = xcalloc(1, 1);
	del_prefix = xcalloc(1, 1);

	/*
	 * See if one's subtree resembles two; if so we need to prefix
	 * two with a few fake trees to match the prefix.
	 */
	match_trees(hash1, hash2, &add_score, &add_prefix, "", depth_limit);

	/*
	 * See if two's subtree resembles one; if so we need to
	 * pick only subtree of two.
	 */
	match_trees(hash2, hash1, &del_score, &del_prefix, "", depth_limit);

	/* Assume we do not have to do any shifting */
	oidcpy(shifted, hash2);

	if (add_score < del_score) {
		/* We need to pick a subtree of two */
		unsigned mode;

		if (!*del_prefix)
			return;

		if (get_tree_entry(hash2->hash, del_prefix, shifted->hash, &mode))
			die("cannot find path %s in tree %s",
			    del_prefix, oid_to_hex(hash2));
		return;
	}

	if (!*add_prefix)
		return;

	splice_tree(hash1, add_prefix, hash2, shifted);
}

/*
 * The user says the trees will be shifted by this much.
 * Unfortunately we cannot fundamentally tell which one to
 * be prefixed, as recursive merge can work in either direction.
 */
void shift_tree_by(const struct object_id *hash1,
		   const struct object_id *hash2,
		   struct object_id *shifted,
		   const char *shift_prefix)
{
	struct object_id sub1, sub2;
	unsigned mode1, mode2;
	unsigned candidate = 0;

	/* Can hash2 be a tree at shift_prefix in tree hash1? */
	if (!get_tree_entry(hash1->hash, shift_prefix, sub1.hash, &mode1) &&
	    S_ISDIR(mode1))
		candidate |= 1;

	/* Can hash1 be a tree at shift_prefix in tree hash2? */
	if (!get_tree_entry(hash2->hash, shift_prefix, sub2.hash, &mode2) &&
	    S_ISDIR(mode2))
		candidate |= 2;

	if (candidate == 3) {
		/* Both are plausible -- we need to evaluate the score */
		int best_score = score_trees(hash1, hash2);
		int score;

		candidate = 0;
		score = score_trees(&sub1, hash2);
		if (score > best_score) {
			candidate = 1;
			best_score = score;
		}
		score = score_trees(&sub2, hash1);
		if (score > best_score)
			candidate = 2;
	}

	if (!candidate) {
		/* Neither is plausible -- do not shift */
		oidcpy(shifted, hash2);
		return;
	}

	if (candidate == 1)
		/*
		 * shift tree2 down by adding shift_prefix above it
		 * to match tree1.
		 */
		splice_tree(hash1, shift_prefix, hash2, shifted);
	else
		/*
		 * shift tree2 up by removing shift_prefix from it
		 * to match tree1.
		 */
		oidcpy(shifted, &sub2);
}
Commit	Line	Data
68faf689 JH	1	#include "cache.h"
	2	#include "tree.h"
	3	#include "tree-walk.h"
	4
	5	static int score_missing(unsigned mode, const char *path)
	6	{
	7	int score;
	8
	9	if (S_ISDIR(mode))
	10	score = -1000;
	11	else if (S_ISLNK(mode))
	12	score = -500;
	13	else
	14	score = -50;
	15	return score;
	16	}
	17
	18	static int score_differs(unsigned mode1, unsigned mode2, const char *path)
	19	{
	20	int score;
	21
	22	if (S_ISDIR(mode1) != S_ISDIR(mode2))
	23	score = -100;
	24	else if (S_ISLNK(mode1) != S_ISLNK(mode2))
	25	score = -50;
	26	else
	27	score = -5;
	28	return score;
	29	}
	30
	31	static int score_matches(unsigned mode1, unsigned mode2, const char *path)
	32	{
	33	int score;
	34
	35	/* Heh, we found SHA-1 collisions between different kind of objects */
	36	if (S_ISDIR(mode1) != S_ISDIR(mode2))
	37	score = -100;
	38	else if (S_ISLNK(mode1) != S_ISLNK(mode2))
	39	score = -50;
	40
	41	else if (S_ISDIR(mode1))
	42	score = 1000;
	43	else if (S_ISLNK(mode1))
	44	score = 500;
	45	else
	46	score = 250;
	47	return score;
	48	}
	49
10a3fb00	50	static void fill_tree_desc_strict(struct tree_desc desc,
b6aec868	51	const struct object_id *hash)
10a3fb00 RS	52	{
	53	void *buffer;
	54	enum object_type type;
	55	unsigned long size;
	56
b6aec868	57	buffer = read_sha1_file(hash->hash, &type, &size);
10a3fb00	58	if (!buffer)
b6aec868	59	die("unable to read tree (%s)", oid_to_hex(hash));
10a3fb00	60	if (type != OBJ_TREE)
b6aec868	61	die("%s is not a tree", oid_to_hex(hash));
10a3fb00 RS	62	init_tree_desc(desc, buffer, size);
	63	return buffer;
	64	}
	65
d8febde3 RS	66	static int base_name_entries_compare(const struct name_entry *a,
	67	const struct name_entry *b)
	68	{
	69	return base_name_compare(a->path, tree_entry_len(a), a->mode,
	70	b->path, tree_entry_len(b), b->mode);
	71	}
	72
68faf689 JH	73	/*
	74	* Inspect two trees, and give a score that tells how similar they are.
	75	*/
b6aec868	76	static int score_trees(const struct object_id hash1, const struct object_id hash2)
68faf689 JH	77	{
	78	struct tree_desc one;
	79	struct tree_desc two;
10a3fb00 RS	80	void *one_buf = fill_tree_desc_strict(&one, hash1);
10a3fb00 RS	81	void *two_buf = fill_tree_desc_strict(&two, hash2);
68faf689	82	int score = 0;
68faf689	83
d8febde3 RS	84	for (;;) {
	85	struct name_entry e1, e2;
	86	int got_entry_from_one = tree_entry(&one, &e1);
	87	int got_entry_from_two = tree_entry(&two, &e2);
68faf689 JH	88	int cmp;
68faf689 JH	89
d8febde3 RS	90	if (got_entry_from_one && got_entry_from_two)
	91	cmp = base_name_entries_compare(&e1, &e2);
	92	else if (got_entry_from_one)
68faf689	93	/* two lacks this entry */
d8febde3 RS	94	cmp = -1;
	95	else if (got_entry_from_two)
	96	/* two has more entries */
	97	cmp = 1;
	98	else
	99	break;
	100
	101	if (cmp < 0)
68faf689	102	/* path1 does not appear in two */
d8febde3 RS	103	score += score_missing(e1.mode, e1.path);
d8febde3 RS	104	else if (cmp > 0)
68faf689	105	/* path2 does not appear in one */
d8febde3	106	score += score_missing(e2.mode, e2.path);
7d924c91	107	else if (oidcmp(e1.oid, e2.oid))
68faf689	108	/* they are different */
d8febde3	109	score += score_differs(e1.mode, e2.mode, e1.path);
68faf689 JH	110	else
68faf689 JH	111	/* same subtree or blob */
d8febde3	112	score += score_matches(e1.mode, e2.mode, e1.path);
68faf689 JH	113	}
	114	free(one_buf);
	115	free(two_buf);
	116	return score;
	117	}
	118
	119	/*
	120	* Match one itself and its subtrees with two and pick the best match.
	121	*/
b6aec868	122	static void match_trees(const struct object_id *hash1,
b6aec868	123	const struct object_id *hash2,
68faf689 JH	124	int *best_score,
68faf689 JH	125	char **best_match,
538dfe73	126	const char *base,
68faf689 JH	127	int recurse_limit)
	128	{
	129	struct tree_desc one;
10a3fb00	130	void *one_buf = fill_tree_desc_strict(&one, hash1);
68faf689 JH	131
	132	while (one.size) {
	133	const char *path;
ce6663a9	134	const struct object_id *elem;
68faf689 JH	135	unsigned mode;
	136	int score;
	137
	138	elem = tree_entry_extract(&one, &path, &mode);
	139	if (!S_ISDIR(mode))
	140	goto next;
b6aec868	141	score = score_trees(elem, hash2);
68faf689	142	if (*best_score < score) {
68faf689	143	free(*best_match);
28310186	144	*best_match = xstrfmt("%s%s", base, path);
68faf689 JH	145	*best_score = score;
	146	}
	147	if (recurse_limit) {
28310186	148	char *newbase = xstrfmt("%s%s/", base, path);
b6aec868	149	match_trees(elem, hash2, best_score, best_match,
68faf689 JH	150	newbase, recurse_limit - 1);
	151	free(newbase);
	152	}
	153
	154	next:
	155	update_tree_entry(&one);
	156	}
	157	free(one_buf);
	158	}
	159
	160	/*
3b34934d PO	161	* A tree "oid1" has a subdirectory at "prefix". Come up with a tree object by
3b34934d PO	162	* replacing it with another tree "oid2".
68faf689	163	*/
3b34934d PO	164	static int splice_tree(const struct object_id oid1, const char prefix,
3b34934d PO	165	const struct object_id oid2, struct object_id result)
68faf689 JH	166	{
	167	char *subpath;
	168	int toplen;
	169	char *buf;
	170	unsigned long sz;
	171	struct tree_desc desc;
3b34934d PO	172	struct object_id *rewrite_here;
	173	const struct object_id *rewrite_with;
	174	struct object_id subtree;
68faf689 JH	175	enum object_type type;
	176	int status;
	177
2c5495f7 RM	178	subpath = strchrnul(prefix, '/');
	179	toplen = subpath - prefix;
	180	if (*subpath)
68faf689	181	subpath++;
68faf689	182
3b34934d	183	buf = read_sha1_file(oid1->hash, &type, &sz);
68faf689	184	if (!buf)
3b34934d	185	die("cannot read tree %s", oid_to_hex(oid1));
68faf689 JH	186	init_tree_desc(&desc, buf, sz);
	187
	188	rewrite_here = NULL;
	189	while (desc.size) {
	190	const char *name;
	191	unsigned mode;
ce6663a9	192	const struct object_id *oid;
68faf689	193
ce6663a9	194	oid = tree_entry_extract(&desc, &name, &mode);
68faf689 JH	195	if (strlen(name) == toplen &&
	196	!memcmp(name, prefix, toplen)) {
	197	if (!S_ISDIR(mode))
3b34934d PO	198	die("entry %s in tree %s is not a tree", name,
	199	oid_to_hex(oid1));
	200	rewrite_here = (struct object_id *)oid;
68faf689 JH	201	break;
	202	}
	203	update_tree_entry(&desc);
	204	}
	205	if (!rewrite_here)
3b34934d PO	206	die("entry %.*s not found in tree %s", toplen, prefix,
3b34934d PO	207	oid_to_hex(oid1));
2c5495f7	208	if (*subpath) {
3b34934d	209	status = splice_tree(rewrite_here, subpath, oid2, &subtree);
68faf689 JH	210	if (status)
68faf689 JH	211	return status;
3b34934d PO	212	rewrite_with = &subtree;
	213	} else {
	214	rewrite_with = oid2;
68faf689	215	}
3b34934d	216	oidcpy(rewrite_here, rewrite_with);
a09c985e	217	status = write_object_file(buf, sz, tree_type, result);
68faf689 JH	218	free(buf);
	219	return status;
	220	}
	221
	222	/*
	223	* We are trying to come up with a merge between one and two that
	224	* results in a tree shape similar to one. The tree two might
	225	* correspond to a subtree of one, in which case it needs to be
	226	* shifted down by prefixing otherwise empty directories. On the
	227	* other hand, it could cover tree one and we might need to pick a
	228	* subtree of it.
	229	*/
82db3d44	230	void shift_tree(const struct object_id *hash1,
	231	const struct object_id *hash2,
	232	struct object_id *shifted,
68faf689 JH	233	int depth_limit)
	234	{
	235	char *add_prefix;
	236	char *del_prefix;
	237	int add_score, del_score;
	238
85e51b78 JH	239	/*
	240	* NEEDSWORK: this limits the recursion depth to hardcoded
	241	* value '2' to avoid excessive overhead.
	242	*/
	243	if (!depth_limit)
	244	depth_limit = 2;
	245
b6aec868	246	add_score = del_score = score_trees(hash1, hash2);
68faf689 JH	247	add_prefix = xcalloc(1, 1);
	248	del_prefix = xcalloc(1, 1);
	249
	250	/*
	251	* See if one's subtree resembles two; if so we need to prefix
	252	* two with a few fake trees to match the prefix.
	253	*/
b6aec868	254	match_trees(hash1, hash2, &add_score, &add_prefix, "", depth_limit);
68faf689 JH	255
	256	/*
	257	* See if two's subtree resembles one; if so we need to
	258	* pick only subtree of two.
	259	*/
b6aec868	260	match_trees(hash2, hash1, &del_score, &del_prefix, "", depth_limit);
68faf689 JH	261
68faf689 JH	262	/* Assume we do not have to do any shifting */
82db3d44	263	oidcpy(shifted, hash2);
68faf689 JH	264
	265	if (add_score < del_score) {
	266	/* We need to pick a subtree of two */
	267	unsigned mode;
	268
	269	if (!*del_prefix)
	270	return;
	271
82db3d44	272	if (get_tree_entry(hash2->hash, del_prefix, shifted->hash, &mode))
68faf689	273	die("cannot find path %s in tree %s",
82db3d44	274	del_prefix, oid_to_hex(hash2));
68faf689 JH	275	return;
	276	}
	277
	278	if (!*add_prefix)
	279	return;
	280
3b34934d	281	splice_tree(hash1, add_prefix, hash2, shifted);
68faf689	282	}
85e51b78 JH	283
	284	/*
	285	* The user says the trees will be shifted by this much.
	286	* Unfortunately we cannot fundamentally tell which one to
	287	* be prefixed, as recursive merge can work in either direction.
	288	*/
82db3d44	289	void shift_tree_by(const struct object_id *hash1,
	290	const struct object_id *hash2,
	291	struct object_id *shifted,
85e51b78 JH	292	const char *shift_prefix)
85e51b78 JH	293	{
82db3d44	294	struct object_id sub1, sub2;
85e51b78 JH	295	unsigned mode1, mode2;
	296	unsigned candidate = 0;
	297
	298	/* Can hash2 be a tree at shift_prefix in tree hash1? */
82db3d44	299	if (!get_tree_entry(hash1->hash, shift_prefix, sub1.hash, &mode1) &&
85e51b78 JH	300	S_ISDIR(mode1))
	301	candidate \|= 1;
	302
	303	/* Can hash1 be a tree at shift_prefix in tree hash2? */
82db3d44	304	if (!get_tree_entry(hash2->hash, shift_prefix, sub2.hash, &mode2) &&
85e51b78 JH	305	S_ISDIR(mode2))
	306	candidate \|= 2;
	307
	308	if (candidate == 3) {
	309	/* Both are plausible -- we need to evaluate the score */
b6aec868	310	int best_score = score_trees(hash1, hash2);
85e51b78 JH	311	int score;
	312
	313	candidate = 0;
b6aec868	314	score = score_trees(&sub1, hash2);
85e51b78 JH	315	if (score > best_score) {
	316	candidate = 1;
	317	best_score = score;
	318	}
b6aec868	319	score = score_trees(&sub2, hash1);
85e51b78 JH	320	if (score > best_score)
	321	candidate = 2;
	322	}
	323
	324	if (!candidate) {
	325	/* Neither is plausible -- do not shift */
82db3d44	326	oidcpy(shifted, hash2);
85e51b78 JH	327	return;
	328	}
	329
	330	if (candidate == 1)
	331	/*
	332	* shift tree2 down by adding shift_prefix above it
	333	* to match tree1.
	334	*/
3b34934d	335	splice_tree(hash1, shift_prefix, hash2, shifted);
85e51b78 JH	336	else
	337	/*
	338	* shift tree2 up by removing shift_prefix from it
	339	* to match tree1.
	340	*/
82db3d44	341	oidcpy(shifted, &sub2);
85e51b78	342	}