[thirdparty/git.git] / fsck-cache.c

#include <sys/types.h>
#include <dirent.h>

#include "cache.h"
#include "commit.h"
#include "tree.h"
#include "blob.h"
#include "tag.h"

#define REACHABLE 0x0001

static int show_root = 0;
static int show_tags = 0;
static int show_unreachable = 0;
static int keep_cache_objects = 0; 
static unsigned char head_sha1[20];

static void check_connectivity(void)
{
	int i;

	/* Look up all the requirements, warn about missing objects.. */
	for (i = 0; i < nr_objs; i++) {
		struct object *obj = objs[i];
		struct object_list *refs;

		if (!obj->parsed) {
			printf("missing %s %s\n", obj->type, sha1_to_hex(obj->sha1));
			continue;
		}

		for (refs = obj->refs; refs; refs = refs->next) {
			if (refs->item->parsed)
				continue;
			printf("broken link from %7s %s\n",
			       obj->type, sha1_to_hex(obj->sha1));
			printf("              to %7s %s\n",
			       refs->item->type, sha1_to_hex(refs->item->sha1));
		}

		/* Don't bother with tag reachability. */
		if (obj->type == tag_type)
			continue;

		if (show_unreachable && !(obj->flags & REACHABLE)) {
			printf("unreachable %s %s\n", obj->type, sha1_to_hex(obj->sha1));
			continue;
		}

		if (!obj->used) {
			printf("dangling %s %s\n", obj->type, 
			       sha1_to_hex(obj->sha1));
		}
	}
}

/*
 * The entries in a tree are ordered in the _path_ order,
 * which means that a directory entry is ordered by adding
 * a slash to the end of it.
 *
 * So a directory called "a" is ordered _after_ a file
 * called "a.c", because "a/" sorts after "a.c".
 */
#define TREE_UNORDERED (-1)
#define TREE_HAS_DUPS  (-2)

static int verify_ordered(struct tree_entry_list *a, struct tree_entry_list *b)
{
	int len1 = strlen(a->name);
	int len2 = strlen(b->name);
	int len = len1 < len2 ? len1 : len2;
	unsigned char c1, c2;
	int cmp;

	cmp = memcmp(a->name, b->name, len);
	if (cmp < 0)
		return 0;
	if (cmp > 0)
		return TREE_UNORDERED;

	/*
	 * Ok, the first <len> characters are the same.
	 * Now we need to order the next one, but turn
	 * a '\0' into a '/' for a directory entry.
	 */
	c1 = a->name[len];
	c2 = b->name[len];
	if (!c1 && !c2)
		/*
		 * git-write-tree used to write out a nonsense tree that has
		 * entries with the same name, one blob and one tree.  Make
		 * sure we do not have duplicate entries.
		 */
		return TREE_HAS_DUPS;
	if (!c1 && a->directory)
		c1 = '/';
	if (!c2 && b->directory)
		c2 = '/';
	return c1 < c2 ? 0 : TREE_UNORDERED;
}

static int fsck_tree(struct tree *item)
{
	int has_full_path = 0;
	struct tree_entry_list *entry, *last;

	last = NULL;
	for (entry = item->entries; entry; entry = entry->next) {
		if (strchr(entry->name, '/'))
			has_full_path = 1;

		switch (entry->mode) {
		/*
		 * Standard modes.. 
		 */
		case S_IFREG | 0755:
		case S_IFREG | 0644:
		case S_IFLNK:
		case S_IFDIR:
			break;
		/*
		 * This is nonstandard, but we had a few of these
		 * early on when we honored the full set of mode
		 * bits..
		 */
		case S_IFREG | 0664:
			break;
		default:
			printf("tree %s has entry %o %s\n",
				sha1_to_hex(item->object.sha1),
				entry->mode, entry->name);
		}

		if (last) {
			switch (verify_ordered(last, entry)) {
			case TREE_UNORDERED:
				fprintf(stderr, "tree %s not ordered\n",
					sha1_to_hex(item->object.sha1));
				return -1;
			case TREE_HAS_DUPS:
				fprintf(stderr, "tree %s has duplicate entries for '%s'\n",
					sha1_to_hex(item->object.sha1),
					entry->name);
				return -1;
			default:
				break;
			}
		}

		last = entry;
	}

	if (has_full_path) {
		fprintf(stderr, "warning: fsck-cache: tree %s "
			"has full pathnames in it\n", 
			sha1_to_hex(item->object.sha1));
	}

	return 0;
}

static int fsck_commit(struct commit *commit)
{
	if (!commit->tree)
		return -1;
	if (!commit->parents && show_root)
		printf("root %s\n", sha1_to_hex(commit->object.sha1));
	if (!commit->date)
		printf("bad commit date in %s\n", 
		       sha1_to_hex(commit->object.sha1));
	return 0;
}

static int fsck_tag(struct tag *tag)
{
	struct object *tagged = tag->tagged;

	if (!tagged) {
		printf("bad object in tag %s\n", sha1_to_hex(tag->object.sha1));
		return -1;
	}
	if (!show_tags)
		return 0;

	printf("tagged %s %s", tagged->type, sha1_to_hex(tagged->sha1));
	printf(" (%s) in %s\n", tag->tag, sha1_to_hex(tag->object.sha1));
	return 0;
}

static int fsck_sha1(unsigned char *sha1)
{
	struct object *obj = parse_object(sha1);
	if (!obj)
		return -1;
	if (obj->type == blob_type)
		return 0;
	if (obj->type == tree_type)
		return fsck_tree((struct tree *) obj);
	if (obj->type == commit_type)
		return fsck_commit((struct commit *) obj);
	if (obj->type == tag_type)
		return fsck_tag((struct tag *) obj);
	return -1;
}

/*
 * This is the sorting chunk size: make it reasonably
 * big so that we can sort well..
 */
#define MAX_SHA1_ENTRIES (1024)

struct sha1_entry {
	unsigned long ino;
	unsigned char sha1[20];
};

static struct {
	unsigned long nr;
	struct sha1_entry *entry[MAX_SHA1_ENTRIES];
} sha1_list;

static int ino_compare(const void *_a, const void *_b)
{
	const struct sha1_entry *a = _a, *b = _b;
	unsigned long ino1 = a->ino, ino2 = b->ino;
	return ino1 < ino2 ? -1 : ino1 > ino2 ? 1 : 0;
}

static void fsck_sha1_list(void)
{
	int i, nr = sha1_list.nr;

	qsort(sha1_list.entry, nr, sizeof(struct sha1_entry *), ino_compare);
	for (i = 0; i < nr; i++) {
		struct sha1_entry *entry = sha1_list.entry[i];
		unsigned char *sha1 = entry->sha1;

		sha1_list.entry[i] = NULL;
		if (fsck_sha1(sha1) < 0)
			fprintf(stderr, "bad sha1 entry '%s'\n", sha1_to_hex(sha1));
		free(entry);
	}
	sha1_list.nr = 0;
}

static void add_sha1_list(unsigned char *sha1, unsigned long ino)
{
	struct sha1_entry *entry = xmalloc(sizeof(*entry));
	int nr;

	entry->ino = ino;
	memcpy(entry->sha1, sha1, 20);
	nr = sha1_list.nr;
	if (nr == MAX_SHA1_ENTRIES) {
		fsck_sha1_list();
		nr = 0;
	}
	sha1_list.entry[nr] = entry;
	sha1_list.nr = ++nr;
}

static int fsck_dir(int i, char *path)
{
	DIR *dir = opendir(path);
	struct dirent *de;

	if (!dir) {
		return error("missing sha1 directory '%s'", path);
	}

	while ((de = readdir(dir)) != NULL) {
		char name[100];
		unsigned char sha1[20];
		int len = strlen(de->d_name);

		switch (len) {
		case 2:
			if (de->d_name[1] != '.')
				break;
		case 1:
			if (de->d_name[0] != '.')
				break;
			continue;
		case 38:
			sprintf(name, "%02x", i);
			memcpy(name+2, de->d_name, len+1);
			if (get_sha1_hex(name, sha1) < 0)
				break;
			add_sha1_list(sha1, de->d_ino);
			continue;
		}
		fprintf(stderr, "bad sha1 file: %s/%s\n", path, de->d_name);
	}
	closedir(dir);
	return 0;
}

int main(int argc, char **argv)
{
	int i, heads;
	char *sha1_dir;

	for (i = 1; i < argc; i++) {
		const char *arg = argv[i];

		if (!strcmp(arg, "--unreachable")) {
			show_unreachable = 1;
			continue;
		}
		if (!strcmp(arg, "--tags")) {
			show_tags = 1;
			continue;
		}
		if (!strcmp(arg, "--root")) {
			show_root = 1;
			continue;
		}
		if (!strcmp(arg, "--cache")) {
			keep_cache_objects = 1;
			continue;
		}
		if (*arg == '-')
			usage("fsck-cache [--tags] [[--unreachable] [--cache] <head-sha1>*]");
	}

	sha1_dir = get_object_directory();
	for (i = 0; i < 256; i++) {
		static char dir[4096];
		sprintf(dir, "%s/%02x", sha1_dir, i);
		fsck_dir(i, dir);
	}
	fsck_sha1_list();

	heads = 0;
	for (i = 1; i < argc; i++) {
		const char *arg = argv[i]; 

		if (*arg == '-')
			continue;

		if (!get_sha1(arg, head_sha1)) {
			struct object *obj = lookup_object(head_sha1);

			/* Error is printed by lookup_object(). */
			if (!obj)
				continue;

			obj->used = 1;
			mark_reachable(obj, REACHABLE);
			heads++;
			continue;
		}
		error("expected sha1, got %s", arg);
	}

	if (keep_cache_objects) {
		int i;
		read_cache();
		for (i = 0; i < active_nr; i++) {
			struct blob *blob = lookup_blob(active_cache[i]->sha1);
			struct object *obj;
			if (!blob)
				continue;
			obj = &blob->object;
			obj->used = 1;
			mark_reachable(obj, REACHABLE);
		}
	}

	if (!heads && !keep_cache_objects) {
		if (show_unreachable) {
			fprintf(stderr, "unable to do reachability without a head nor --cache\n");
			show_unreachable = 0; 
		}
		if (!heads)
			fprintf(stderr, "expect dangling commits - potential heads - due to lack of head information\n");
	}

	check_connectivity();
	return 0;
}
Commit	Line	Data
20222118 LT	1	#include <sys/types.h>
	2	#include <dirent.h>
	3
4b182421	4	#include "cache.h"
ff5ebe39 DB	5	#include "commit.h"
	6	#include "tree.h"
	7	#include "blob.h"
c418eda4	8	#include "tag.h"
ff5ebe39 DB	9
ff5ebe39 DB	10	#define REACHABLE 0x0001
d9839e03	11
ab7df187	12	static int show_root = 0;
889262ea	13	static int show_tags = 0;
d9839e03	14	static int show_unreachable = 0;
ae7c0c92	15	static int keep_cache_objects = 0;
d9839e03 LT	16	static unsigned char head_sha1[20];
d9839e03 LT	17
8ba0bbb2 LT	18	static void check_connectivity(void)
	19	{
	20	int i;
	21
8ba0bbb2	22	/* Look up all the requirements, warn about missing objects.. */
ff5ebe39 DB	23	for (i = 0; i < nr_objs; i++) {
ff5ebe39 DB	24	struct object *obj = objs[i];
c418eda4	25	struct object_list *refs;
8ba0bbb2	26
3a6a23e6 LT	27	if (!obj->parsed) {
	28	printf("missing %s %s\n", obj->type, sha1_to_hex(obj->sha1));
	29	continue;
	30	}
	31
	32	for (refs = obj->refs; refs; refs = refs->next) {
	33	if (refs->item->parsed)
	34	continue;
	35	printf("broken link from %7s %s\n",
	36	obj->type, sha1_to_hex(obj->sha1));
	37	printf(" to %7s %s\n",
aa034134	38	refs->item->type, sha1_to_hex(refs->item->sha1));
3a6a23e6 LT	39	}
	40
	41	/* Don't bother with tag reachability. */
	42	if (obj->type == tag_type)
	43	continue;
	44
ff5ebe39	45	if (show_unreachable && !(obj->flags & REACHABLE)) {
f43b8abc	46	printf("unreachable %s %s\n", obj->type, sha1_to_hex(obj->sha1));
8ba0bbb2	47	continue;
d9839e03	48	}
8ba0bbb2	49
ff5ebe39 DB	50	if (!obj->used) {
	51	printf("dangling %s %s\n", obj->type,
	52	sha1_to_hex(obj->sha1));
d9839e03	53	}
8ba0bbb2 LT	54	}
	55	}
	56
85003492 LT	57	/*
	58	* The entries in a tree are ordered in the _path_ order,
	59	* which means that a directory entry is ordered by adding
	60	* a slash to the end of it.
	61	*
	62	* So a directory called "a" is ordered _after_ a file
	63	* called "a.c", because "a/" sorts after "a.c".
	64	*/
a4f35a2d JH	65	#define TREE_UNORDERED (-1)
	66	#define TREE_HAS_DUPS (-2)
	67
85003492 LT	68	static int verify_ordered(struct tree_entry_list a, struct tree_entry_list b)
	69	{
	70	int len1 = strlen(a->name);
	71	int len2 = strlen(b->name);
	72	int len = len1 < len2 ? len1 : len2;
	73	unsigned char c1, c2;
	74	int cmp;
	75
	76	cmp = memcmp(a->name, b->name, len);
	77	if (cmp < 0)
	78	return 0;
	79	if (cmp > 0)
a4f35a2d	80	return TREE_UNORDERED;
85003492 LT	81
	82	/*
	83	* Ok, the first <len> characters are the same.
	84	* Now we need to order the next one, but turn
	85	* a '\0' into a '/' for a directory entry.
	86	*/
	87	c1 = a->name[len];
	88	c2 = b->name[len];
a4f35a2d JH	89	if (!c1 && !c2)
	90	/*
	91	* git-write-tree used to write out a nonsense tree that has
	92	* entries with the same name, one blob and one tree. Make
	93	* sure we do not have duplicate entries.
	94	*/
	95	return TREE_HAS_DUPS;
85003492 LT	96	if (!c1 && a->directory)
	97	c1 = '/';
	98	if (!c2 && b->directory)
	99	c2 = '/';
a4f35a2d	100	return c1 < c2 ? 0 : TREE_UNORDERED;
85003492 LT	101	}
85003492 LT	102
c418eda4	103	static int fsck_tree(struct tree *item)
20222118	104	{
85003492 LT	105	int has_full_path = 0;
	106	struct tree_entry_list entry, last;
	107
	108	last = NULL;
	109	for (entry = item->entries; entry; entry = entry->next) {
	110	if (strchr(entry->name, '/'))
	111	has_full_path = 1;
	112
42ea9cb2 LT	113	switch (entry->mode) {
	114	/*
	115	* Standard modes..
	116	*/
	117	case S_IFREG \| 0755:
	118	case S_IFREG \| 0644:
	119	case S_IFLNK:
	120	case S_IFDIR:
	121	break;
	122	/*
	123	* This is nonstandard, but we had a few of these
	124	* early on when we honored the full set of mode
	125	* bits..
	126	*/
	127	case S_IFREG \| 0664:
	128	break;
	129	default:
	130	printf("tree %s has entry %o %s\n",
	131	sha1_to_hex(item->object.sha1),
	132	entry->mode, entry->name);
	133	}
	134
85003492	135	if (last) {
a4f35a2d JH	136	switch (verify_ordered(last, entry)) {
a4f35a2d JH	137	case TREE_UNORDERED:
85003492 LT	138	fprintf(stderr, "tree %s not ordered\n",
	139	sha1_to_hex(item->object.sha1));
	140	return -1;
a4f35a2d JH	141	case TREE_HAS_DUPS:
	142	fprintf(stderr, "tree %s has duplicate entries for '%s'\n",
	143	sha1_to_hex(item->object.sha1),
	144	entry->name);
	145	return -1;
	146	default:
	147	break;
85003492 LT	148	}
	149	}
	150
	151	last = entry;
	152	}
	153
	154	if (has_full_path) {
ff5ebe39	155	fprintf(stderr, "warning: fsck-cache: tree %s "
c418eda4 DB	156	"has full pathnames in it\n",
c418eda4 DB	157	sha1_to_hex(item->object.sha1));
1ea34e36	158	}
85003492	159
59c1e249	160	return 0;
1ea34e36 LT	161	}
1ea34e36 LT	162
c418eda4	163	static int fsck_commit(struct commit *commit)
1ea34e36	164	{
ff5ebe39	165	if (!commit->tree)
1ea34e36	166	return -1;
ab7df187	167	if (!commit->parents && show_root)
c418eda4	168	printf("root %s\n", sha1_to_hex(commit->object.sha1));
e6948b6d	169	if (!commit->date)
c418eda4 DB	170	printf("bad commit date in %s\n",
c418eda4 DB	171	sha1_to_hex(commit->object.sha1));
4728b861 LT	172	return 0;
	173	}
	174
c418eda4	175	static int fsck_tag(struct tag *tag)
ec4465ad	176	{
92d4c85d LT	177	struct object *tagged = tag->tagged;
	178
	179	if (!tagged) {
	180	printf("bad object in tag %s\n", sha1_to_hex(tag->object.sha1));
	181	return -1;
	182	}
889262ea LT	183	if (!show_tags)
	184	return 0;
	185
92d4c85d LT	186	printf("tagged %s %s", tagged->type, sha1_to_hex(tagged->sha1));
92d4c85d LT	187	printf(" (%s) in %s\n", tag->tag, sha1_to_hex(tag->object.sha1));
ff5ebe39	188	return 0;
20222118 LT	189	}
20222118 LT	190
7e8c174a	191	static int fsck_sha1(unsigned char *sha1)
20222118	192	{
7e8c174a LT	193	struct object *obj = parse_object(sha1);
	194	if (!obj)
	195	return -1;
	196	if (obj->type == blob_type)
	197	return 0;
	198	if (obj->type == tree_type)
	199	return fsck_tree((struct tree *) obj);
	200	if (obj->type == commit_type)
	201	return fsck_commit((struct commit *) obj);
	202	if (obj->type == tag_type)
	203	return fsck_tag((struct tag *) obj);
	204	return -1;
	205	}
	206
	207	/*
	208	* This is the sorting chunk size: make it reasonably
	209	* big so that we can sort well..
	210	*/
	211	#define MAX_SHA1_ENTRIES (1024)
	212
	213	struct sha1_entry {
	214	unsigned long ino;
20222118	215	unsigned char sha1[20];
7e8c174a LT	216	};
	217
	218	static struct {
	219	unsigned long nr;
	220	struct sha1_entry *entry[MAX_SHA1_ENTRIES];
	221	} sha1_list;
	222
	223	static int ino_compare(const void _a, const void _b)
	224	{
	225	const struct sha1_entry a = _a, b = _b;
	226	unsigned long ino1 = a->ino, ino2 = b->ino;
	227	return ino1 < ino2 ? -1 : ino1 > ino2 ? 1 : 0;
	228	}
	229
	230	static void fsck_sha1_list(void)
	231	{
	232	int i, nr = sha1_list.nr;
	233
	234	qsort(sha1_list.entry, nr, sizeof(struct sha1_entry *), ino_compare);
	235	for (i = 0; i < nr; i++) {
	236	struct sha1_entry *entry = sha1_list.entry[i];
	237	unsigned char *sha1 = entry->sha1;
	238
	239	sha1_list.entry[i] = NULL;
	240	if (fsck_sha1(sha1) < 0)
	241	fprintf(stderr, "bad sha1 entry '%s'\n", sha1_to_hex(sha1));
	242	free(entry);
20222118	243	}
7e8c174a LT	244	sha1_list.nr = 0;
	245	}
	246
	247	static void add_sha1_list(unsigned char *sha1, unsigned long ino)
	248	{
	249	struct sha1_entry entry = xmalloc(sizeof(entry));
	250	int nr;
	251
	252	entry->ino = ino;
	253	memcpy(entry->sha1, sha1, 20);
	254	nr = sha1_list.nr;
	255	if (nr == MAX_SHA1_ENTRIES) {
	256	fsck_sha1_list();
	257	nr = 0;
	258	}
	259	sha1_list.entry[nr] = entry;
	260	sha1_list.nr = ++nr;
20222118 LT	261	}
	262
	263	static int fsck_dir(int i, char *path)
	264	{
	265	DIR *dir = opendir(path);
	266	struct dirent *de;
	267
	268	if (!dir) {
2de381f9	269	return error("missing sha1 directory '%s'", path);
20222118 LT	270	}
	271
	272	while ((de = readdir(dir)) != NULL) {
	273	char name[100];
7e8c174a	274	unsigned char sha1[20];
20222118 LT	275	int len = strlen(de->d_name);
	276
	277	switch (len) {
	278	case 2:
	279	if (de->d_name[1] != '.')
	280	break;
	281	case 1:
	282	if (de->d_name[0] != '.')
	283	break;
	284	continue;
	285	case 38:
	286	sprintf(name, "%02x", i);
	287	memcpy(name+2, de->d_name, len+1);
7e8c174a LT	288	if (get_sha1_hex(name, sha1) < 0)
	289	break;
	290	add_sha1_list(sha1, de->d_ino);
	291	continue;
20222118 LT	292	}
	293	fprintf(stderr, "bad sha1 file: %s/%s\n", path, de->d_name);
	294	}
	295	closedir(dir);
	296	return 0;
	297	}
	298
	299	int main(int argc, char **argv)
	300	{
bcee6fd8	301	int i, heads;
20222118 LT	302	char *sha1_dir;
20222118 LT	303
889262ea LT	304	for (i = 1; i < argc; i++) {
	305	const char *arg = argv[i];
	306
	307	if (!strcmp(arg, "--unreachable")) {
	308	show_unreachable = 1;
	309	continue;
	310	}
	311	if (!strcmp(arg, "--tags")) {
	312	show_tags = 1;
	313	continue;
	314	}
ab7df187 LT	315	if (!strcmp(arg, "--root")) {
	316	show_root = 1;
	317	continue;
	318	}
ae7c0c92 JH	319	if (!strcmp(arg, "--cache")) {
	320	keep_cache_objects = 1;
	321	continue;
	322	}
889262ea	323	if (*arg == '-')
ae7c0c92	324	usage("fsck-cache [--tags] [[--unreachable] [--cache] <head-sha1>*]");
889262ea LT	325	}
889262ea LT	326
ace1534d	327	sha1_dir = get_object_directory();
bcee6fd8 LT	328	for (i = 0; i < 256; i++) {
	329	static char dir[4096];
	330	sprintf(dir, "%s/%02x", sha1_dir, i);
	331	fsck_dir(i, dir);
	332	}
7e8c174a	333	fsck_sha1_list();
bcee6fd8 LT	334
bcee6fd8 LT	335	heads = 0;
d9839e03	336	for (i = 1; i < argc; i++) {
889262ea LT	337	const char *arg = argv[i];
	338
	339	if (*arg == '-')
d9839e03	340	continue;
889262ea	341
3c249c95	342	if (!get_sha1(arg, head_sha1)) {
770896e5	343	struct object *obj = lookup_object(head_sha1);
e1a1388d	344
770896e5 LT	345	/* Error is printed by lookup_object(). */
770896e5 LT	346	if (!obj)
e1a1388d JF	347	continue;
e1a1388d JF	348
ff5ebe39 DB	349	obj->used = 1;
ff5ebe39 DB	350	mark_reachable(obj, REACHABLE);
bcee6fd8	351	heads++;
d9839e03 LT	352	continue;
d9839e03 LT	353	}
889262ea	354	error("expected sha1, got %s", arg);
d9839e03	355	}
d9839e03	356
ae7c0c92 JH	357	if (keep_cache_objects) {
	358	int i;
	359	read_cache();
	360	for (i = 0; i < active_nr; i++) {
	361	struct blob *blob = lookup_blob(active_cache[i]->sha1);
	362	struct object *obj;
	363	if (!blob)
	364	continue;
	365	obj = &blob->object;
	366	obj->used = 1;
	367	mark_reachable(obj, REACHABLE);
	368	}
	369	}
	370
	371	if (!heads && !keep_cache_objects) {
bcee6fd8	372	if (show_unreachable) {
ae7c0c92	373	fprintf(stderr, "unable to do reachability without a head nor --cache\n");
bcee6fd8 LT	374	show_unreachable = 0;
bcee6fd8 LT	375	}
ae7c0c92 JH	376	if (!heads)
ae7c0c92 JH	377	fprintf(stderr, "expect dangling commits - potential heads - due to lack of head information\n");
20222118	378	}
bcee6fd8	379
8ba0bbb2	380	check_connectivity();
20222118 LT	381	return 0;
20222118 LT	382	}