[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"
#include "refs.h"
#include "pack.h"

#define REACHABLE 0x0001

static int show_root = 0;
static int show_tags = 0;
static int show_unreachable = 0;
static int standalone = 0;
static int check_full = 0;
static int check_strict = 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) {
			if (!standalone && has_sha1_file(obj->sha1))
				; /* it is in pack */
			else
				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 ||
			    (!standalone && has_sha1_file(refs->item->sha1)))
				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));
		}

		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 retval;
	int has_full_path = 0;
	int has_zero_pad = 0;
	int has_bad_modes = 0;
	int has_dup_entries = 0;
	int not_properly_sorted = 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;
		has_zero_pad |= entry->zeropad;

		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:
			if (!check_strict)
				break;
		default:
			has_bad_modes = 1;
		}

		if (last) {
			switch (verify_ordered(last, entry)) {
			case TREE_UNORDERED:
				not_properly_sorted = 1;
				break;
			case TREE_HAS_DUPS:
				has_dup_entries = 1;
				break;
			default:
				break;
			}
		}

		last = entry;
	}

	retval = 0;
	if (has_full_path) {
		fprintf(stderr, "warning: git-fsck-cache: tree %s "
			"has full pathnames in it\n", 
			sha1_to_hex(item->object.sha1));
	}
	if (has_zero_pad) {
		fprintf(stderr, "warning: git-fsck-cache: tree %s "
			"has zero-padded file modes in it\n",
			sha1_to_hex(item->object.sha1));
	}
	if (has_bad_modes) {
		fprintf(stderr, "warning: git-fsck-cache: tree %s "
			"has bad file modes in it\n",
			sha1_to_hex(item->object.sha1));
	}
	if (has_dup_entries) {
		fprintf(stderr, "error: git-fsck-cache: tree %s "
			"has duplicate file entries\n",
			sha1_to_hex(item->object.sha1));
		retval = -1;
	}
	if (not_properly_sorted) {
		fprintf(stderr, "error: git-fsck-cache: tree %s "
			"is not properly sorted\n",
			sha1_to_hex(item->object.sha1));
		retval = -1;
	}
	return retval;
}

static int fsck_commit(struct commit *commit)
{
	char *buffer = commit->buffer;
	unsigned char sha1[20];

	if (memcmp(buffer, "tree ", 5))
		return -1;
	if (get_sha1_hex(buffer+5, sha1) || buffer[45] != '\n')
		return -1;
	buffer += 46;
	while (!memcmp(buffer, "parent ", 7)) {
		if (get_sha1_hex(buffer+7, sha1) || buffer[47] != '\n')
			return -1;
		buffer += 48;
	}
	if (memcmp(buffer, "author ", 7))
		return -1;
	free(commit->buffer);
	commit->buffer = NULL;
	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;
}

static int default_refs = 0;

static int fsck_handle_ref(const char *refname, const unsigned char *sha1)
{
	struct object *obj;

	obj = lookup_object(sha1);
	if (!obj) {
		if (!standalone && has_sha1_file(sha1)) {
			default_refs++;
			return 0; /* it is in a pack */
		}
		error("%s: invalid sha1 pointer %s", refname, sha1_to_hex(sha1));
		/* We'll continue with the rest despite the error.. */
		return 0;
	}
	default_refs++;
	obj->used = 1;
	mark_reachable(obj, REACHABLE);
	return 0;
}

static void get_default_heads(void)
{
	for_each_ref(fsck_handle_ref);
	if (!default_refs)
		die("No default references");
}

static void fsck_object_dir(const char *path)
{
	int i;
	for (i = 0; i < 256; i++) {
		static char dir[4096];
		sprintf(dir, "%s/%02x", path, i);
		fsck_dir(i, dir);
	}
	fsck_sha1_list();
}

static int fsck_head_link(void)
{
	int fd, count;
	char hex[40];
	unsigned char sha1[20];
	static char path[PATH_MAX], link[PATH_MAX];
	const char *git_dir = gitenv(GIT_DIR_ENVIRONMENT) ? : DEFAULT_GIT_DIR_ENVIRONMENT;

	snprintf(path, sizeof(path), "%s/HEAD", git_dir);
	if (readlink(path, link, sizeof(link)) < 0)
		return error("HEAD is not a symlink");
	if (strncmp("refs/heads/", link, 11))
		return error("HEAD points to something strange (%s)", link);
	fd = open(path, O_RDONLY);
	if (fd < 0)
		return error("HEAD: %s", strerror(errno));
	count = read(fd, hex, sizeof(hex));
	close(fd);
	if (count < 0)
		return error("HEAD: %s", strerror(errno));
	if (count < 40 || get_sha1_hex(hex, sha1))
		return error("HEAD: not a valid git pointer");
	return 0;
}

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

	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 (!strcmp(arg, "--standalone")) {
			standalone = 1;
			continue;
		}
		if (!strcmp(arg, "--full")) {
			check_full = 1;
			continue;
		}
		if (!strcmp(arg, "--strict")) {
			check_strict = 1;
			continue;
		}
		if (*arg == '-')
			usage("git-fsck-cache [--tags] [--root] [[--unreachable] [--cache] [--standalone | --full] [--strict] <head-sha1>*]");
	}

	if (standalone && check_full)
		die("Only one of --standalone or --full can be used.");
	if (standalone)
		unsetenv("GIT_ALTERNATE_OBJECT_DIRECTORIES");

	fsck_head_link();
	fsck_object_dir(get_object_directory());
	if (check_full) {
		int j;
		struct packed_git *p;
		prepare_alt_odb();
		for (j = 0; alt_odb[j].base; j++) {
			char namebuf[PATH_MAX];
			int namelen = alt_odb[j].name - alt_odb[j].base;
			memcpy(namebuf, alt_odb[j].base, namelen);
			namebuf[namelen - 1] = 0;
			fsck_object_dir(namebuf);
		}
		prepare_packed_git();
		for (p = packed_git; p; p = p->next)
			/* verify gives error messages itself */
			verify_pack(p, 0);

		for (p = packed_git; p; p = p->next) {
			int num = num_packed_objects(p);
			for (i = 0; i < num; i++) {
				unsigned char sha1[20];
				nth_packed_object_sha1(p, i, sha1);
				if (fsck_sha1(sha1) < 0)
					fprintf(stderr, "bad sha1 entry '%s'\n", sha1_to_hex(sha1));

			}
		}
	}

	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 we've not been given any explicit head information, do the
	 * default ones from .git/refs. We also consider the index file
	 * in this case (ie this implies --cache).
	 */
	if (!heads) {
		get_default_heads();
		keep_cache_objects = 1;
	}

	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);
		}
	}

	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"
944d8589	9	#include "refs.h"
f9253394	10	#include "pack.h"
ff5ebe39 DB	11
ff5ebe39 DB	12	#define REACHABLE 0x0001
d9839e03	13
ab7df187	14	static int show_root = 0;
889262ea	15	static int show_tags = 0;
d9839e03	16	static int show_unreachable = 0;
8a498a05 JH	17	static int standalone = 0;
8a498a05 JH	18	static int check_full = 0;
de2eb7f6	19	static int check_strict = 0;
ae7c0c92	20	static int keep_cache_objects = 0;
d9839e03 LT	21	static unsigned char head_sha1[20];
d9839e03 LT	22
8ba0bbb2 LT	23	static void check_connectivity(void)
	24	{
	25	int i;
	26
8ba0bbb2	27	/* Look up all the requirements, warn about missing objects.. */
ff5ebe39 DB	28	for (i = 0; i < nr_objs; i++) {
ff5ebe39 DB	29	struct object *obj = objs[i];
c418eda4	30	struct object_list *refs;
8ba0bbb2	31
3a6a23e6	32	if (!obj->parsed) {
8a498a05 JH	33	if (!standalone && has_sha1_file(obj->sha1))
	34	; /* it is in pack */
	35	else
	36	printf("missing %s %s\n",
	37	obj->type, sha1_to_hex(obj->sha1));
3a6a23e6 LT	38	continue;
	39	}
	40
	41	for (refs = obj->refs; refs; refs = refs->next) {
8a498a05 JH	42	if (refs->item->parsed \|\|
8a498a05 JH	43	(!standalone && has_sha1_file(refs->item->sha1)))
3a6a23e6 LT	44	continue;
	45	printf("broken link from %7s %s\n",
	46	obj->type, sha1_to_hex(obj->sha1));
	47	printf(" to %7s %s\n",
aa034134	48	refs->item->type, sha1_to_hex(refs->item->sha1));
3a6a23e6 LT	49	}
3a6a23e6 LT	50
ff5ebe39	51	if (show_unreachable && !(obj->flags & REACHABLE)) {
c4584ae3 JH	52	printf("unreachable %s %s\n",
c4584ae3 JH	53	obj->type, sha1_to_hex(obj->sha1));
8ba0bbb2	54	continue;
d9839e03	55	}
8ba0bbb2	56
ff5ebe39 DB	57	if (!obj->used) {
	58	printf("dangling %s %s\n", obj->type,
	59	sha1_to_hex(obj->sha1));
d9839e03	60	}
8ba0bbb2 LT	61	}
	62	}
	63
85003492 LT	64	/*
	65	* The entries in a tree are ordered in the _path_ order,
	66	* which means that a directory entry is ordered by adding
	67	* a slash to the end of it.
	68	*
	69	* So a directory called "a" is ordered _after_ a file
	70	* called "a.c", because "a/" sorts after "a.c".
	71	*/
a4f35a2d JH	72	#define TREE_UNORDERED (-1)
	73	#define TREE_HAS_DUPS (-2)
	74
85003492 LT	75	static int verify_ordered(struct tree_entry_list a, struct tree_entry_list b)
	76	{
	77	int len1 = strlen(a->name);
	78	int len2 = strlen(b->name);
	79	int len = len1 < len2 ? len1 : len2;
	80	unsigned char c1, c2;
	81	int cmp;
	82
	83	cmp = memcmp(a->name, b->name, len);
	84	if (cmp < 0)
	85	return 0;
	86	if (cmp > 0)
a4f35a2d	87	return TREE_UNORDERED;
85003492 LT	88
	89	/*
	90	* Ok, the first <len> characters are the same.
	91	* Now we need to order the next one, but turn
	92	* a '\0' into a '/' for a directory entry.
	93	*/
	94	c1 = a->name[len];
	95	c2 = b->name[len];
a4f35a2d JH	96	if (!c1 && !c2)
	97	/*
	98	* git-write-tree used to write out a nonsense tree that has
	99	* entries with the same name, one blob and one tree. Make
	100	* sure we do not have duplicate entries.
	101	*/
	102	return TREE_HAS_DUPS;
85003492 LT	103	if (!c1 && a->directory)
	104	c1 = '/';
	105	if (!c2 && b->directory)
	106	c2 = '/';
a4f35a2d	107	return c1 < c2 ? 0 : TREE_UNORDERED;
85003492 LT	108	}
85003492 LT	109
c418eda4	110	static int fsck_tree(struct tree *item)
20222118	111	{
64071805	112	int retval;
85003492	113	int has_full_path = 0;
64071805 LT	114	int has_zero_pad = 0;
	115	int has_bad_modes = 0;
	116	int has_dup_entries = 0;
	117	int not_properly_sorted = 0;
85003492 LT	118	struct tree_entry_list entry, last;
	119
	120	last = NULL;
	121	for (entry = item->entries; entry; entry = entry->next) {
	122	if (strchr(entry->name, '/'))
	123	has_full_path = 1;
64071805	124	has_zero_pad \|= entry->zeropad;
85003492	125
42ea9cb2 LT	126	switch (entry->mode) {
	127	/*
	128	* Standard modes..
	129	*/
	130	case S_IFREG \| 0755:
	131	case S_IFREG \| 0644:
	132	case S_IFLNK:
	133	case S_IFDIR:
	134	break;
	135	/*
	136	* This is nonstandard, but we had a few of these
	137	* early on when we honored the full set of mode
	138	* bits..
	139	*/
	140	case S_IFREG \| 0664:
de2eb7f6 LT	141	if (!check_strict)
de2eb7f6 LT	142	break;
42ea9cb2	143	default:
64071805	144	has_bad_modes = 1;
42ea9cb2 LT	145	}
42ea9cb2 LT	146
85003492	147	if (last) {
a4f35a2d JH	148	switch (verify_ordered(last, entry)) {
a4f35a2d JH	149	case TREE_UNORDERED:
64071805 LT	150	not_properly_sorted = 1;
64071805 LT	151	break;
a4f35a2d	152	case TREE_HAS_DUPS:
64071805 LT	153	has_dup_entries = 1;
64071805 LT	154	break;
a4f35a2d JH	155	default:
a4f35a2d JH	156	break;
85003492 LT	157	}
	158	}
	159
	160	last = entry;
	161	}
	162
64071805	163	retval = 0;
85003492	164	if (has_full_path) {
667bb59b	165	fprintf(stderr, "warning: git-fsck-cache: tree %s "
c418eda4 DB	166	"has full pathnames in it\n",
c418eda4 DB	167	sha1_to_hex(item->object.sha1));
1ea34e36	168	}
64071805 LT	169	if (has_zero_pad) {
	170	fprintf(stderr, "warning: git-fsck-cache: tree %s "
	171	"has zero-padded file modes in it\n",
	172	sha1_to_hex(item->object.sha1));
	173	}
	174	if (has_bad_modes) {
	175	fprintf(stderr, "warning: git-fsck-cache: tree %s "
	176	"has bad file modes in it\n",
	177	sha1_to_hex(item->object.sha1));
	178	}
	179	if (has_dup_entries) {
	180	fprintf(stderr, "error: git-fsck-cache: tree %s "
	181	"has duplicate file entries\n",
	182	sha1_to_hex(item->object.sha1));
	183	retval = -1;
	184	}
	185	if (not_properly_sorted) {
	186	fprintf(stderr, "error: git-fsck-cache: tree %s "
	187	"is not properly sorted\n",
	188	sha1_to_hex(item->object.sha1));
	189	retval = -1;
	190	}
	191	return retval;
1ea34e36 LT	192	}
1ea34e36 LT	193
c418eda4	194	static int fsck_commit(struct commit *commit)
1ea34e36	195	{
de2eb7f6 LT	196	char *buffer = commit->buffer;
	197	unsigned char sha1[20];
	198
	199	if (memcmp(buffer, "tree ", 5))
	200	return -1;
	201	if (get_sha1_hex(buffer+5, sha1) \|\| buffer[45] != '\n')
	202	return -1;
	203	buffer += 46;
	204	while (!memcmp(buffer, "parent ", 7)) {
	205	if (get_sha1_hex(buffer+7, sha1) \|\| buffer[47] != '\n')
	206	return -1;
	207	buffer += 48;
	208	}
	209	if (memcmp(buffer, "author ", 7))
	210	return -1;
bd1e17e2 LT	211	free(commit->buffer);
bd1e17e2 LT	212	commit->buffer = NULL;
ff5ebe39	213	if (!commit->tree)
1ea34e36	214	return -1;
ab7df187	215	if (!commit->parents && show_root)
c418eda4	216	printf("root %s\n", sha1_to_hex(commit->object.sha1));
e6948b6d	217	if (!commit->date)
c418eda4 DB	218	printf("bad commit date in %s\n",
c418eda4 DB	219	sha1_to_hex(commit->object.sha1));
4728b861 LT	220	return 0;
	221	}
	222
c418eda4	223	static int fsck_tag(struct tag *tag)
ec4465ad	224	{
92d4c85d LT	225	struct object *tagged = tag->tagged;
	226
	227	if (!tagged) {
	228	printf("bad object in tag %s\n", sha1_to_hex(tag->object.sha1));
	229	return -1;
	230	}
889262ea LT	231	if (!show_tags)
	232	return 0;
	233
92d4c85d LT	234	printf("tagged %s %s", tagged->type, sha1_to_hex(tagged->sha1));
92d4c85d LT	235	printf(" (%s) in %s\n", tag->tag, sha1_to_hex(tag->object.sha1));
ff5ebe39	236	return 0;
20222118 LT	237	}
20222118 LT	238
7e8c174a	239	static int fsck_sha1(unsigned char *sha1)
20222118	240	{
7e8c174a LT	241	struct object *obj = parse_object(sha1);
	242	if (!obj)
	243	return -1;
	244	if (obj->type == blob_type)
	245	return 0;
	246	if (obj->type == tree_type)
	247	return fsck_tree((struct tree *) obj);
	248	if (obj->type == commit_type)
	249	return fsck_commit((struct commit *) obj);
	250	if (obj->type == tag_type)
	251	return fsck_tag((struct tag *) obj);
	252	return -1;
	253	}
	254
	255	/*
	256	* This is the sorting chunk size: make it reasonably
	257	* big so that we can sort well..
	258	*/
	259	#define MAX_SHA1_ENTRIES (1024)
	260
	261	struct sha1_entry {
	262	unsigned long ino;
20222118	263	unsigned char sha1[20];
7e8c174a LT	264	};
	265
	266	static struct {
	267	unsigned long nr;
	268	struct sha1_entry *entry[MAX_SHA1_ENTRIES];
	269	} sha1_list;
	270
	271	static int ino_compare(const void _a, const void _b)
	272	{
	273	const struct sha1_entry a = _a, b = _b;
	274	unsigned long ino1 = a->ino, ino2 = b->ino;
	275	return ino1 < ino2 ? -1 : ino1 > ino2 ? 1 : 0;
	276	}
	277
	278	static void fsck_sha1_list(void)
	279	{
	280	int i, nr = sha1_list.nr;
	281
	282	qsort(sha1_list.entry, nr, sizeof(struct sha1_entry *), ino_compare);
	283	for (i = 0; i < nr; i++) {
	284	struct sha1_entry *entry = sha1_list.entry[i];
	285	unsigned char *sha1 = entry->sha1;
	286
	287	sha1_list.entry[i] = NULL;
	288	if (fsck_sha1(sha1) < 0)
	289	fprintf(stderr, "bad sha1 entry '%s'\n", sha1_to_hex(sha1));
	290	free(entry);
20222118	291	}
7e8c174a LT	292	sha1_list.nr = 0;
	293	}
	294
	295	static void add_sha1_list(unsigned char *sha1, unsigned long ino)
	296	{
	297	struct sha1_entry entry = xmalloc(sizeof(entry));
	298	int nr;
	299
	300	entry->ino = ino;
	301	memcpy(entry->sha1, sha1, 20);
	302	nr = sha1_list.nr;
	303	if (nr == MAX_SHA1_ENTRIES) {
	304	fsck_sha1_list();
	305	nr = 0;
	306	}
	307	sha1_list.entry[nr] = entry;
	308	sha1_list.nr = ++nr;
20222118 LT	309	}
	310
	311	static int fsck_dir(int i, char *path)
	312	{
	313	DIR *dir = opendir(path);
	314	struct dirent *de;
	315
	316	if (!dir) {
2de381f9	317	return error("missing sha1 directory '%s'", path);
20222118 LT	318	}
	319
	320	while ((de = readdir(dir)) != NULL) {
	321	char name[100];
7e8c174a	322	unsigned char sha1[20];
20222118 LT	323	int len = strlen(de->d_name);
	324
	325	switch (len) {
	326	case 2:
	327	if (de->d_name[1] != '.')
	328	break;
	329	case 1:
	330	if (de->d_name[0] != '.')
	331	break;
	332	continue;
	333	case 38:
	334	sprintf(name, "%02x", i);
	335	memcpy(name+2, de->d_name, len+1);
7e8c174a LT	336	if (get_sha1_hex(name, sha1) < 0)
	337	break;
	338	add_sha1_list(sha1, de->d_ino);
	339	continue;
20222118 LT	340	}
	341	fprintf(stderr, "bad sha1 file: %s/%s\n", path, de->d_name);
	342	}
	343	closedir(dir);
	344	return 0;
	345	}
	346
944d8589 LT	347	static int default_refs = 0;
	348
	349	static int fsck_handle_ref(const char refname, const unsigned char sha1)
1024932f	350	{
1024932f LT	351	struct object *obj;
1024932f LT	352
1024932f	353	obj = lookup_object(sha1);
8a498a05	354	if (!obj) {
659cacf5 LT	355	if (!standalone && has_sha1_file(sha1)) {
659cacf5 LT	356	default_refs++;
944d8589	357	return 0; /* it is in a pack */
659cacf5	358	}
944d8589 LT	359	error("%s: invalid sha1 pointer %s", refname, sha1_to_hex(sha1));
	360	/* We'll continue with the rest despite the error.. */
	361	return 0;
8a498a05	362	}
944d8589	363	default_refs++;
1024932f LT	364	obj->used = 1;
1024932f LT	365	mark_reachable(obj, REACHABLE);
7c4d07c7	366	return 0;
1024932f LT	367	}
1024932f LT	368
1024932f LT	369	static void get_default_heads(void)
1024932f LT	370	{
944d8589 LT	371	for_each_ref(fsck_handle_ref);
944d8589 LT	372	if (!default_refs)
477606f5	373	die("No default references");
1024932f LT	374	}
1024932f LT	375
8a498a05 JH	376	static void fsck_object_dir(const char *path)
	377	{
	378	int i;
	379	for (i = 0; i < 256; i++) {
	380	static char dir[4096];
	381	sprintf(dir, "%s/%02x", path, i);
	382	fsck_dir(i, dir);
	383	}
	384	fsck_sha1_list();
	385	}
	386
c3330383 LT	387	static int fsck_head_link(void)
	388	{
	389	int fd, count;
	390	char hex[40];
	391	unsigned char sha1[20];
	392	static char path[PATH_MAX], link[PATH_MAX];
	393	const char *git_dir = gitenv(GIT_DIR_ENVIRONMENT) ? : DEFAULT_GIT_DIR_ENVIRONMENT;
	394
	395	snprintf(path, sizeof(path), "%s/HEAD", git_dir);
	396	if (readlink(path, link, sizeof(link)) < 0)
	397	return error("HEAD is not a symlink");
	398	if (strncmp("refs/heads/", link, 11))
	399	return error("HEAD points to something strange (%s)", link);
	400	fd = open(path, O_RDONLY);
	401	if (fd < 0)
	402	return error("HEAD: %s", strerror(errno));
	403	count = read(fd, hex, sizeof(hex));
	404	close(fd);
	405	if (count < 0)
	406	return error("HEAD: %s", strerror(errno));
	407	if (count < 40 \|\| get_sha1_hex(hex, sha1))
	408	return error("HEAD: not a valid git pointer");
	409	return 0;
	410	}
	411
20222118 LT	412	int main(int argc, char **argv)
20222118 LT	413	{
bcee6fd8	414	int i, heads;
20222118	415
889262ea LT	416	for (i = 1; i < argc; i++) {
	417	const char *arg = argv[i];
	418
	419	if (!strcmp(arg, "--unreachable")) {
	420	show_unreachable = 1;
	421	continue;
	422	}
	423	if (!strcmp(arg, "--tags")) {
	424	show_tags = 1;
	425	continue;
	426	}
ab7df187 LT	427	if (!strcmp(arg, "--root")) {
	428	show_root = 1;
	429	continue;
	430	}
ae7c0c92 JH	431	if (!strcmp(arg, "--cache")) {
	432	keep_cache_objects = 1;
	433	continue;
	434	}
8a498a05 JH	435	if (!strcmp(arg, "--standalone")) {
	436	standalone = 1;
	437	continue;
	438	}
	439	if (!strcmp(arg, "--full")) {
	440	check_full = 1;
	441	continue;
	442	}
de2eb7f6 LT	443	if (!strcmp(arg, "--strict")) {
	444	check_strict = 1;
	445	continue;
	446	}
889262ea	447	if (*arg == '-')
fb6a3d86	448	usage("git-fsck-cache [--tags] [--root] [[--unreachable] [--cache] [--standalone \| --full] [--strict] <head-sha1>*]");
889262ea LT	449	}
889262ea LT	450
8a498a05 JH	451	if (standalone && check_full)
	452	die("Only one of --standalone or --full can be used.");
	453	if (standalone)
	454	unsetenv("GIT_ALTERNATE_OBJECT_DIRECTORIES");
	455
c3330383	456	fsck_head_link();
8a498a05 JH	457	fsck_object_dir(get_object_directory());
	458	if (check_full) {
	459	int j;
	460	struct packed_git *p;
	461	prepare_alt_odb();
	462	for (j = 0; alt_odb[j].base; j++) {
a3eb250f JH	463	char namebuf[PATH_MAX];
	464	int namelen = alt_odb[j].name - alt_odb[j].base;
	465	memcpy(namebuf, alt_odb[j].base, namelen);
	466	namebuf[namelen - 1] = 0;
	467	fsck_object_dir(namebuf);
8a498a05 JH	468	}
8a498a05 JH	469	prepare_packed_git();
f9253394 JH	470	for (p = packed_git; p; p = p->next)
f9253394 JH	471	/* verify gives error messages itself */
f3bf9224	472	verify_pack(p, 0);
f9253394	473
8a498a05 JH	474	for (p = packed_git; p; p = p->next) {
	475	int num = num_packed_objects(p);
	476	for (i = 0; i < num; i++) {
	477	unsigned char sha1[20];
	478	nth_packed_object_sha1(p, i, sha1);
	479	if (fsck_sha1(sha1) < 0)
	480	fprintf(stderr, "bad sha1 entry '%s'\n", sha1_to_hex(sha1));
	481
	482	}
	483	}
bcee6fd8 LT	484	}
	485
	486	heads = 0;
d9839e03	487	for (i = 1; i < argc; i++) {
889262ea LT	488	const char *arg = argv[i];
	489
	490	if (*arg == '-')
d9839e03	491	continue;
889262ea	492
3c249c95	493	if (!get_sha1(arg, head_sha1)) {
770896e5	494	struct object *obj = lookup_object(head_sha1);
e1a1388d	495
770896e5 LT	496	/* Error is printed by lookup_object(). */
770896e5 LT	497	if (!obj)
e1a1388d JF	498	continue;
e1a1388d JF	499
ff5ebe39 DB	500	obj->used = 1;
ff5ebe39 DB	501	mark_reachable(obj, REACHABLE);
bcee6fd8	502	heads++;
d9839e03 LT	503	continue;
d9839e03 LT	504	}
889262ea	505	error("expected sha1, got %s", arg);
d9839e03	506	}
d9839e03	507
1024932f	508	/*
d1af002d	509	* If we've not been given any explicit head information, do the
e7bd907d LT	510	* default ones from .git/refs. We also consider the index file
e7bd907d LT	511	* in this case (ie this implies --cache).
1024932f	512	*/
e7bd907d	513	if (!heads) {
1024932f LT	514	get_default_heads();
	515	keep_cache_objects = 1;
	516	}
	517
ae7c0c92 JH	518	if (keep_cache_objects) {
	519	int i;
	520	read_cache();
	521	for (i = 0; i < active_nr; i++) {
	522	struct blob *blob = lookup_blob(active_cache[i]->sha1);
	523	struct object *obj;
	524	if (!blob)
	525	continue;
	526	obj = &blob->object;
	527	obj->used = 1;
	528	mark_reachable(obj, REACHABLE);
	529	}
	530	}
	531
8ba0bbb2	532	check_connectivity();
20222118 LT	533	return 0;
20222118 LT	534	}