[thirdparty/git.git] / unpack-objects.c

#include "cache.h"
#include "object.h"
#include "delta.h"

static int dry_run;
static int nr_entries;
static const char *base_name;
static const char unpack_usage[] = "git-unpack-objects basename";

struct pack_entry {
	unsigned int offset; /* network byte order */
	unsigned char sha1[20];
};

static void *pack_base;
static unsigned long pack_size;
static void *index_base;
static unsigned long index_size;

static struct pack_entry **pack_list;

static void *map_file(const char *suffix, unsigned long *sizep)
{
	static char pathname[PATH_MAX];
	unsigned long len;
	int fd;
	struct stat st;
	void *map;

	len = snprintf(pathname, PATH_MAX, "%s.%s", base_name, suffix);
	if (len >= PATH_MAX)
		die("bad pack base-name");
	fd = open(pathname, O_RDONLY);
	if (fd < 0 || fstat(fd, &st))
		die("unable to open '%s'", pathname);
	len = st.st_size;
	if (!len)
		die("bad pack file '%s'", pathname);
	map = mmap(NULL, len, PROT_READ, MAP_PRIVATE, fd, 0);
	if (-1 == (int)(long)map)
		die("unable to mmap '%s'", pathname);
	close(fd);
	*sizep = len;
	return map;
}

static int sort_by_offset(const void *_a, const void *_b)
{
	struct pack_entry *a = *(struct pack_entry **)_a;
	struct pack_entry *b = *(struct pack_entry **)_b;
	unsigned int o1, o2;

	o1 = ntohl(a->offset);
	o2 = ntohl(b->offset);
	return o1 < o2 ? -1 : 1;
}

static int check_index(void)
{
	unsigned int *array = index_base;
	unsigned int nr;
	int i;

	if (index_size < 4*256 + 20)
		return error("index file too small");
	nr = 0;
	for (i = 0; i < 256; i++) {
		unsigned int n = ntohl(array[i]);
		if (n < nr)
			return error("non-monotonic index");
		nr = n;
	}
	/*
	 * Total size:
	 *  - 256 index entries 4 bytes each
	 *  - 24-byte entries * nr (20-byte sha1 + 4-byte offset)
	 *  - 20-byte SHA1 file checksum
	 */
	if (index_size != 4*256 + nr * 24 + 20)
		return error("wrong index file size");

	nr_entries = nr;
	pack_list = xmalloc(nr * sizeof(struct pack_entry *));
	for (i = 0; i < nr; i++)
		pack_list[i] = index_base + 4*256 + i*24;

	qsort(pack_list, nr, sizeof(*pack_list), sort_by_offset);

	printf("%d entries\n", nr);
	return 0;
}

static int unpack_non_delta_entry(struct pack_entry *entry,
				  int kind,
				  unsigned char *data,
				  unsigned long size,
				  unsigned long left)
{
	int st;
	z_stream stream;
	char *buffer;
	unsigned char sha1[20];
	char *type_s;

	printf("%s %c %lu\n", sha1_to_hex(entry->sha1), kind, size);
	if (dry_run)
		return 0;

	buffer = xmalloc(size + 1);
	buffer[size] = 0;
	memset(&stream, 0, sizeof(stream));
	stream.next_in = data;
	stream.avail_in = left;
	stream.next_out = buffer;
	stream.avail_out = size;

	inflateInit(&stream);
	st = inflate(&stream, Z_FINISH);
	inflateEnd(&stream);
	if ((st != Z_STREAM_END) || stream.total_out != size)
		goto err_finish;
	switch (kind) {
	case 'C': type_s = "commit"; break;
	case 'T': type_s = "tree"; break;
	case 'B': type_s = "blob"; break;
	default: goto err_finish;
	}
	if (write_sha1_file(buffer, size, type_s, sha1) < 0)
		die("failed to write %s (%s)",
		    sha1_to_hex(entry->sha1), type_s);
	printf("%s %s\n", sha1_to_hex(sha1), type_s);
	if (memcmp(sha1, entry->sha1, 20))
		die("resulting %s have wrong SHA1", type_s);

 finish:
	st = 0;
	free(buffer);
	return st;
 err_finish:
	st = -1;
	goto finish;
}

static int find_pack_entry(unsigned char *sha1, struct pack_entry **ent)
{
	int *level1_ofs = index_base;
	int hi = ntohl(level1_ofs[*sha1]);
	int lo = ((*sha1 == 0x0) ? 0 : ntohl(level1_ofs[*sha1 - 1]));
	void *index = index_base + 4*256;

	do {
		int mi = (lo + hi) / 2;
		int cmp = memcmp(index + 24 * mi + 4, sha1, 20);
		if (!cmp) {
			*ent = index + 24 * mi;
			return 1;
		}
		if (cmp > 0)
			hi = mi;
		else
			lo = mi+1;
	} while (lo < hi);
	return 0;
}

/* forward declaration for a mutually recursive function */
static void unpack_entry(struct pack_entry *);

static int unpack_delta_entry(struct pack_entry *entry,
			      unsigned char *base_sha1,
			      unsigned long delta_size,
			      unsigned long left)
{
	void *data, *delta_data, *result, *base;
	unsigned long data_size, result_size, base_size;
	z_stream stream;
	int st;
	char type[20];
	unsigned char sha1[20];

	if (left < 20)
		die("truncated pack file");
	data = base_sha1 + 20;
	data_size = left - 20;
	printf("%s D %lu", sha1_to_hex(entry->sha1), delta_size);
	printf(" %s\n", sha1_to_hex(base_sha1));

	if (dry_run)
		return 0;

	/* pack+5 is the base sha1, unless we have it, we need to
	 * unpack it first.
	 */
	if (!has_sha1_file(base_sha1)) {
		struct pack_entry *base;
		if (!find_pack_entry(base_sha1, &base))
			die("cannot find delta-pack base object");
		unpack_entry(base);
	}
	delta_data = xmalloc(delta_size);

	memset(&stream, 0, sizeof(stream));

	stream.next_in = data;
	stream.avail_in = data_size;
	stream.next_out = delta_data;
	stream.avail_out = delta_size;

	inflateInit(&stream);
	st = inflate(&stream, Z_FINISH);
	inflateEnd(&stream);
	if ((st != Z_STREAM_END) || stream.total_out != delta_size)
		die("delta data unpack failed");

	base = read_sha1_file(base_sha1, type, &base_size);
	if (!base)
		die("failed to read delta-pack base object %s", sha1_to_hex(base_sha1));
	result = patch_delta(base, base_size,
			     delta_data, delta_size,
			     &result_size);
	if (!result)
		die("failed to apply delta");
	free(delta_data);

	if (write_sha1_file(result, result_size, type, sha1) < 0)
		die("failed to write %s (%s)",
		    sha1_to_hex(entry->sha1), type);
	free(result);
	printf("%s %s\n", sha1_to_hex(sha1), type);
	if (memcmp(sha1, entry->sha1, 20))
		die("resulting %s have wrong SHA1", type);
	return 0;
}

static void unpack_entry(struct pack_entry *entry)
{
	unsigned long offset, size, left;
	unsigned char *pack;

	/* Have we done this one already due to deltas based on it? */
	if (lookup_object(entry->sha1))
		return;

	offset = ntohl(entry->offset);
	if (offset > pack_size - 5)
		die("object offset outside of pack file");
	pack = pack_base + offset;
	size = (pack[1] << 24) + (pack[2] << 16) + (pack[3] << 8) + pack[4];
	left = pack_size - offset - 5;
	switch (*pack) {
	case 'C': case 'T': case 'B':
		unpack_non_delta_entry(entry, *pack, pack+5, size, left);
		break;
	case 'D':
		unpack_delta_entry(entry, pack+5, size, left);
		break;
	default:
		die("corrupted pack file");
	}
}

/*
 * We unpack from the end, older files first. Now, usually
 * there are deltas etc, so we'll not actually write the
 * objects in that order, but we might as well try..
 */
static void unpack_all(void)
{
	int i = nr_entries;

	while (--i >= 0) {
		struct pack_entry *entry = pack_list[i];
		unpack_entry(entry);
	}
}

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

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

		if (*arg == '-') {
			if (!strcmp(arg, "-n")) {
				dry_run = 1;
				continue;
			}
			usage(unpack_usage);
		}
		if (base_name)
			usage(unpack_usage);
		base_name = arg;
	}
	if (!base_name)
		usage(unpack_usage);
	index_base = map_file("idx", &index_size);
	pack_base = map_file("pack", &pack_size);
	if (check_index() < 0)
		die("bad index file");
	unpack_all();
	return 0;
}
Commit	Line	Data
bad50dc8	1	#include "cache.h"
74536958	2	#include "object.h"
8ee378a0	3	#include "delta.h"
bad50dc8	4
74536958	5	static int dry_run;
bad50dc8 LT	6	static int nr_entries;
	7	static const char *base_name;
	8	static const char unpack_usage[] = "git-unpack-objects basename";
	9
	10	struct pack_entry {
8ee378a0	11	unsigned int offset; /* network byte order */
bad50dc8 LT	12	unsigned char sha1[20];
	13	};
	14
74536958 LT	15	static void *pack_base;
74536958 LT	16	static unsigned long pack_size;
8ee378a0 JH	17	static void *index_base;
8ee378a0 JH	18	static unsigned long index_size;
74536958	19
bad50dc8 LT	20	static struct pack_entry **pack_list;
	21
	22	static void map_file(const char suffix, unsigned long *sizep)
	23	{
	24	static char pathname[PATH_MAX];
	25	unsigned long len;
	26	int fd;
	27	struct stat st;
	28	void *map;
	29
	30	len = snprintf(pathname, PATH_MAX, "%s.%s", base_name, suffix);
	31	if (len >= PATH_MAX)
	32	die("bad pack base-name");
	33	fd = open(pathname, O_RDONLY);
	34	if (fd < 0 \|\| fstat(fd, &st))
	35	die("unable to open '%s'", pathname);
	36	len = st.st_size;
	37	if (!len)
	38	die("bad pack file '%s'", pathname);
	39	map = mmap(NULL, len, PROT_READ, MAP_PRIVATE, fd, 0);
	40	if (-1 == (int)(long)map)
	41	die("unable to mmap '%s'", pathname);
	42	close(fd);
	43	*sizep = len;
	44	return map;
	45	}
	46
	47	static int sort_by_offset(const void _a, const void _b)
	48	{
	49	struct pack_entry a = (struct pack_entry **)_a;
	50	struct pack_entry b = (struct pack_entry **)_b;
	51	unsigned int o1, o2;
	52
	53	o1 = ntohl(a->offset);
	54	o2 = ntohl(b->offset);
	55	return o1 < o2 ? -1 : 1;
	56	}
	57
8ee378a0	58	static int check_index(void)
bad50dc8	59	{
8ee378a0	60	unsigned int *array = index_base;
bad50dc8 LT	61	unsigned int nr;
	62	int i;
	63
c38138cd	64	if (index_size < 4*256 + 20)
bad50dc8 LT	65	return error("index file too small");
	66	nr = 0;
	67	for (i = 0; i < 256; i++) {
	68	unsigned int n = ntohl(array[i]);
	69	if (n < nr)
	70	return error("non-monotonic index");
	71	nr = n;
	72	}
c38138cd LT	73	/*
	74	* Total size:
	75	* - 256 index entries 4 bytes each
	76	* - 24-byte entries * nr (20-byte sha1 + 4-byte offset)
	77	* - 20-byte SHA1 file checksum
	78	*/
	79	if (index_size != 4256 + nr 24 + 20)
bad50dc8	80	return error("wrong index file size");
bad50dc8 LT	81
	82	nr_entries = nr;
	83	pack_list = xmalloc(nr * sizeof(struct pack_entry *));
	84	for (i = 0; i < nr; i++)
8ee378a0	85	pack_list[i] = index_base + 4256 + i24;
bad50dc8 LT	86
	87	qsort(pack_list, nr, sizeof(*pack_list), sort_by_offset);
	88
	89	printf("%d entries\n", nr);
	90	return 0;
	91	}
	92
8ee378a0	93	static int unpack_non_delta_entry(struct pack_entry *entry,
c4fb06c0 LT	94	int kind,
	95	unsigned char *data,
	96	unsigned long size,
	97	unsigned long left)
74536958	98	{
c4fb06c0	99	int st;
8ee378a0 JH	100	z_stream stream;
	101	char *buffer;
	102	unsigned char sha1[20];
	103	char *type_s;
8ee378a0	104
8ee378a0	105	printf("%s %c %lu\n", sha1_to_hex(entry->sha1), kind, size);
c4fb06c0 LT	106	if (dry_run)
c4fb06c0 LT	107	return 0;
8ee378a0 JH	108
	109	buffer = xmalloc(size + 1);
	110	buffer[size] = 0;
	111	memset(&stream, 0, sizeof(stream));
c4fb06c0 LT	112	stream.next_in = data;
c4fb06c0 LT	113	stream.avail_in = left;
8ee378a0 JH	114	stream.next_out = buffer;
	115	stream.avail_out = size;
	116
	117	inflateInit(&stream);
	118	st = inflate(&stream, Z_FINISH);
	119	inflateEnd(&stream);
	120	if ((st != Z_STREAM_END) \|\| stream.total_out != size)
	121	goto err_finish;
	122	switch (kind) {
	123	case 'C': type_s = "commit"; break;
	124	case 'T': type_s = "tree"; break;
	125	case 'B': type_s = "blob"; break;
	126	default: goto err_finish;
	127	}
	128	if (write_sha1_file(buffer, size, type_s, sha1) < 0)
	129	die("failed to write %s (%s)",
	130	sha1_to_hex(entry->sha1), type_s);
	131	printf("%s %s\n", sha1_to_hex(sha1), type_s);
	132	if (memcmp(sha1, entry->sha1, 20))
	133	die("resulting %s have wrong SHA1", type_s);
	134
	135	finish:
	136	st = 0;
	137	free(buffer);
	138	return st;
	139	err_finish:
	140	st = -1;
	141	goto finish;
	142	}
	143
	144	static int find_pack_entry(unsigned char sha1, struct pack_entry *ent)
	145	{
	146	int *level1_ofs = index_base;
	147	int hi = ntohl(level1_ofs[*sha1]);
	148	int lo = ((sha1 == 0x0) ? 0 : ntohl(level1_ofs[sha1 - 1]));
	149	void index = index_base + 4256;
	150
	151	do {
	152	int mi = (lo + hi) / 2;
	153	int cmp = memcmp(index + 24 * mi + 4, sha1, 20);
	154	if (!cmp) {
	155	ent = index + 24 mi;
	156	return 1;
	157	}
c4fb06c0	158	if (cmp > 0)
8ee378a0 JH	159	hi = mi;
8ee378a0 JH	160	else
c4fb06c0	161	lo = mi+1;
8ee378a0 JH	162	} while (lo < hi);
	163	return 0;
	164	}
	165
	166	/* forward declaration for a mutually recursive function */
	167	static void unpack_entry(struct pack_entry *);
	168
c4fb06c0 LT	169	static int unpack_delta_entry(struct pack_entry *entry,
	170	unsigned char *base_sha1,
	171	unsigned long delta_size,
	172	unsigned long left)
8ee378a0	173	{
c4fb06c0 LT	174	void data, delta_data, result, base;
c4fb06c0 LT	175	unsigned long data_size, result_size, base_size;
8ee378a0 JH	176	z_stream stream;
	177	int st;
	178	char type[20];
	179	unsigned char sha1[20];
	180
c4fb06c0 LT	181	if (left < 20)
	182	die("truncated pack file");
	183	data = base_sha1 + 20;
	184	data_size = left - 20;
	185	printf("%s D %lu", sha1_to_hex(entry->sha1), delta_size);
	186	printf(" %s\n", sha1_to_hex(base_sha1));
	187
	188	if (dry_run)
	189	return 0;
8ee378a0	190
c4fb06c0	191	/* pack+5 is the base sha1, unless we have it, we need to
8ee378a0 JH	192	* unpack it first.
8ee378a0 JH	193	*/
c4fb06c0	194	if (!has_sha1_file(base_sha1)) {
8ee378a0	195	struct pack_entry *base;
c4fb06c0	196	if (!find_pack_entry(base_sha1, &base))
8ee378a0 JH	197	die("cannot find delta-pack base object");
	198	unpack_entry(base);
	199	}
c4fb06c0	200	delta_data = xmalloc(delta_size);
8ee378a0 JH	201
	202	memset(&stream, 0, sizeof(stream));
	203
c4fb06c0 LT	204	stream.next_in = data;
c4fb06c0 LT	205	stream.avail_in = data_size;
8ee378a0	206	stream.next_out = delta_data;
c4fb06c0	207	stream.avail_out = delta_size;
8ee378a0 JH	208
8ee378a0 JH	209	inflateInit(&stream);
c4fb06c0	210	st = inflate(&stream, Z_FINISH);
8ee378a0	211	inflateEnd(&stream);
c4fb06c0 LT	212	if ((st != Z_STREAM_END) \|\| stream.total_out != delta_size)
c4fb06c0 LT	213	die("delta data unpack failed");
8ee378a0	214
c4fb06c0	215	base = read_sha1_file(base_sha1, type, &base_size);
8ee378a0	216	if (!base)
c4fb06c0	217	die("failed to read delta-pack base object %s", sha1_to_hex(base_sha1));
8ee378a0 JH	218	result = patch_delta(base, base_size,
	219	delta_data, delta_size,
	220	&result_size);
	221	if (!result)
	222	die("failed to apply delta");
	223	free(delta_data);
	224
	225	if (write_sha1_file(result, result_size, type, sha1) < 0)
	226	die("failed to write %s (%s)",
	227	sha1_to_hex(entry->sha1), type);
	228	free(result);
	229	printf("%s %s\n", sha1_to_hex(sha1), type);
	230	if (memcmp(sha1, entry->sha1, 20))
	231	die("resulting %s have wrong SHA1", type);
	232	return 0;
	233	}
	234
	235	static void unpack_entry(struct pack_entry *entry)
	236	{
c4fb06c0	237	unsigned long offset, size, left;
74536958 LT	238	unsigned char *pack;
	239
	240	/* Have we done this one already due to deltas based on it? */
	241	if (lookup_object(entry->sha1))
	242	return;
	243
	244	offset = ntohl(entry->offset);
	245	if (offset > pack_size - 5)
	246	die("object offset outside of pack file");
	247	pack = pack_base + offset;
c4fb06c0 LT	248	size = (pack[1] << 24) + (pack[2] << 16) + (pack[3] << 8) + pack[4];
c4fb06c0 LT	249	left = pack_size - offset - 5;
74536958 LT	250	switch (*pack) {
74536958 LT	251	case 'C': case 'T': case 'B':
c4fb06c0	252	unpack_non_delta_entry(entry, *pack, pack+5, size, left);
74536958 LT	253	break;
74536958 LT	254	case 'D':
c4fb06c0	255	unpack_delta_entry(entry, pack+5, size, left);
74536958 LT	256	break;
	257	default:
	258	die("corrupted pack file");
	259	}
	260	}
	261
	262	/*
	263	* We unpack from the end, older files first. Now, usually
	264	* there are deltas etc, so we'll not actually write the
	265	* objects in that order, but we might as well try..
	266	*/
	267	static void unpack_all(void)
	268	{
	269	int i = nr_entries;
	270
	271	while (--i >= 0) {
	272	struct pack_entry *entry = pack_list[i];
	273	unpack_entry(entry);
	274	}
	275	}
	276
bad50dc8 LT	277	int main(int argc, char **argv)
	278	{
	279	int i;
bad50dc8 LT	280
	281	for (i = 1 ; i < argc; i++) {
	282	const char *arg = argv[i];
	283
	284	if (*arg == '-') {
74536958 LT	285	if (!strcmp(arg, "-n")) {
	286	dry_run = 1;
	287	continue;
	288	}
bad50dc8 LT	289	usage(unpack_usage);
	290	}
	291	if (base_name)
	292	usage(unpack_usage);
	293	base_name = arg;
	294	}
	295	if (!base_name)
	296	usage(unpack_usage);
8ee378a0	297	index_base = map_file("idx", &index_size);
74536958	298	pack_base = map_file("pack", &pack_size);
8ee378a0	299	if (check_index() < 0)
bad50dc8	300	die("bad index file");
74536958	301	unpack_all();
bad50dc8 LT	302	return 0;
bad50dc8 LT	303	}