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

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

#include <sys/time.h>

static int dry_run, quiet;
static const char unpack_usage[] = "git-unpack-objects < pack-file";

/* We always read in 4kB chunks. */
static unsigned char buffer[4096];
static unsigned long offset, len, eof;
static SHA_CTX ctx;

/*
 * Make sure at least "min" bytes are available in the buffer, and
 * return the pointer to the buffer.
 */
static void * fill(int min)
{
	if (min <= len)
		return buffer + offset;
	if (eof)
		die("unable to fill input");
	if (min > sizeof(buffer))
		die("cannot fill %d bytes", min);
	if (offset) {
		SHA1_Update(&ctx, buffer, offset);
		memcpy(buffer, buffer + offset, len);
		offset = 0;
	}
	do {
		int ret = read(0, buffer + len, sizeof(buffer) - len);
		if (ret <= 0) {
			if (!ret)
				die("early EOF");
			if (errno == EAGAIN || errno == EINTR)
				continue;
			die("read error on input: %s", strerror(errno));
		}
		len += ret;
	} while (len < min);
	return buffer;
}

static void use(int bytes)
{
	if (bytes > len)
		die("used more bytes than were available");
	len -= bytes;
	offset += bytes;
}

static void *get_data(unsigned long size)
{
	z_stream stream;
	void *buf = xmalloc(size);

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

	stream.next_out = buf;
	stream.avail_out = size;
	stream.next_in = fill(1);
	stream.avail_in = len;
	inflateInit(&stream);

	for (;;) {
		int ret = inflate(&stream, 0);
		use(len - stream.avail_in);
		if (stream.total_out == size && ret == Z_STREAM_END)
			break;
		if (ret != Z_OK)
			die("inflate returned %d\n", ret);
		stream.next_in = fill(1);
		stream.avail_in = len;
	}
	return buf;
}

struct delta_info {
	unsigned char base_sha1[20];
	unsigned long size;
	void *delta;
	struct delta_info *next;
};

static struct delta_info *delta_list;

static void add_delta_to_list(unsigned char *base_sha1, void *delta, unsigned long size)
{
	struct delta_info *info = xmalloc(sizeof(*info));

	memcpy(info->base_sha1, base_sha1, 20);
	info->size = size;
	info->delta = delta;
	info->next = delta_list;
	delta_list = info;
}

static void added_object(unsigned char *sha1, const char *type, void *data, unsigned long size);

static void write_object(void *buf, unsigned long size, const char *type)
{
	unsigned char sha1[20];
	if (write_sha1_file(buf, size, type, sha1) < 0)
		die("failed to write object");
	added_object(sha1, type, buf, size);
}

static int resolve_delta(const char *type,
	void *base, unsigned long base_size, 
	void *delta, unsigned long delta_size)
{
	void *result;
	unsigned long result_size;

	result = patch_delta(base, base_size,
			     delta, delta_size,
			     &result_size);
	if (!result)
		die("failed to apply delta");
	free(delta);
	write_object(result, result_size, type);
	free(result);
	return 0;
}

static void added_object(unsigned char *sha1, const char *type, void *data, unsigned long size)
{
	struct delta_info **p = &delta_list;
	struct delta_info *info;

	while ((info = *p) != NULL) {
		if (!memcmp(info->base_sha1, sha1, 20)) {
			*p = info->next;
			p = &delta_list;
			resolve_delta(type, data, size, info->delta, info->size);
			free(info);
			continue;
		}
		p = &info->next;
	}
}

static int unpack_non_delta_entry(enum object_type kind, unsigned long size)
{
	void *buf = get_data(size);
	const char *type;

	switch (kind) {
	case OBJ_COMMIT: type = "commit"; break;
	case OBJ_TREE:   type = "tree"; break;
	case OBJ_BLOB:   type = "blob"; break;
	case OBJ_TAG:    type = "tag"; break;
	default: die("bad type %d", kind);
	}
	if (!dry_run)
		write_object(buf, size, type);
	free(buf);
	return 0;
}

static int unpack_delta_entry(unsigned long delta_size)
{
	void *delta_data, *base;
	unsigned long base_size;
	char type[20];
	unsigned char base_sha1[20];

	memcpy(base_sha1, fill(20), 20);
	use(20);

	delta_data = get_data(delta_size);
	if (dry_run) {
		free(delta_data);
		return 0;
	}

	if (!has_sha1_file(base_sha1)) {
		add_delta_to_list(base_sha1, delta_data, delta_size);
		return 0;
	}
	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));
	return resolve_delta(type, base, base_size, delta_data, delta_size);
}

static void unpack_one(unsigned nr, unsigned total)
{
	unsigned shift;
	unsigned char *pack, c;
	unsigned long size;
	enum object_type type;

	pack = fill(1);
	c = *pack;
	use(1);
	type = (c >> 4) & 7;
	size = (c & 15);
	shift = 4;
	while (c & 0x80) {
		pack = fill(1);
		c = *pack++;
		use(1);
		size += (c & 0x7f) << shift;
		shift += 7;
	}
	if (!quiet) {
		static unsigned long last_sec;
		static unsigned last_percent;
		struct timeval now;
		unsigned percentage = (nr * 100) / total;

		gettimeofday(&now, NULL);
		if (percentage != last_percent || now.tv_sec != last_sec) {
			last_sec = now.tv_sec;
			last_percent = percentage;
			fprintf(stderr, "%4u%% (%u/%u) done\r", percentage, nr, total);
		}
	}
	switch (type) {
	case OBJ_COMMIT:
	case OBJ_TREE:
	case OBJ_BLOB:
	case OBJ_TAG:
		unpack_non_delta_entry(type, size);
		return;
	case OBJ_DELTA:
		unpack_delta_entry(size);
		return;
	default:
		die("bad object type %d", type);
	}
}

/*
 * 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;
	struct pack_header *hdr = fill(sizeof(struct pack_header));
	unsigned version = ntohl(hdr->hdr_version);
	unsigned nr_objects = ntohl(hdr->hdr_entries);

	if (ntohl(hdr->hdr_signature) != PACK_SIGNATURE)
		die("bad pack file");
	if (version != PACK_VERSION)
		die("unable to handle pack file version %d", version);
	fprintf(stderr, "Unpacking %d objects\n", nr_objects);

	use(sizeof(struct pack_header));
	for (i = 0; i < nr_objects; i++)
		unpack_one(i+1, nr_objects);
	if (delta_list)
		die("unresolved deltas left after unpacking");
}

int main(int argc, char **argv)
{
	int i;
	unsigned char sha1[20];

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

		if (*arg == '-') {
			if (!strcmp(arg, "-n")) {
				dry_run = 1;
				continue;
			}
			if (!strcmp(arg, "-q")) {
				quiet = 1;
				continue;
			}
			usage(unpack_usage);
		}

		/* We don't take any non-flag arguments now.. Maybe some day */
		usage(unpack_usage);
	}
	SHA1_Init(&ctx);
	unpack_all();
	SHA1_Update(&ctx, buffer, offset);
	SHA1_Final(sha1, &ctx);
	if (memcmp(fill(20), sha1, 20))
		die("final sha1 did not match");
	use(20);

	/* Write the last part of the buffer to stdout */
	while (len) {
		int ret = write(1, buffer + offset, len);
		if (!ret)
			break;
		if (ret < 0) {
			if (errno == EAGAIN || errno == EINTR)
				continue;
			break;
		}
		len -= ret;
		offset += ret;
	}

	/* All done */
	if (!quiet)
		fprintf(stderr, "\n");
	return 0;
}
Commit	Line	Data
bad50dc8	1	#include "cache.h"
74536958	2	#include "object.h"
8ee378a0	3	#include "delta.h"
a733cb60	4	#include "pack.h"
bad50dc8	5
d36f7b80 LT	6	#include <sys/time.h>
	7
	8	static int dry_run, quiet;
67e5a5ec	9	static const char unpack_usage[] = "git-unpack-objects < pack-file";
bad50dc8	10
67e5a5ec LT	11	/* We always read in 4kB chunks. */
	12	static unsigned char buffer[4096];
	13	static unsigned long offset, len, eof;
	14	static SHA_CTX ctx;
bad50dc8	15
67e5a5ec LT	16	/*
	17	* Make sure at least "min" bytes are available in the buffer, and
	18	* return the pointer to the buffer.
	19	*/
	20	static void * fill(int min)
	21	{
	22	if (min <= len)
	23	return buffer + offset;
	24	if (eof)
	25	die("unable to fill input");
	26	if (min > sizeof(buffer))
	27	die("cannot fill %d bytes", min);
	28	if (offset) {
	29	SHA1_Update(&ctx, buffer, offset);
	30	memcpy(buffer, buffer + offset, len);
	31	offset = 0;
	32	}
	33	do {
	34	int ret = read(0, buffer + len, sizeof(buffer) - len);
	35	if (ret <= 0) {
	36	if (!ret)
	37	die("early EOF");
	38	if (errno == EAGAIN \|\| errno == EINTR)
	39	continue;
	40	die("read error on input: %s", strerror(errno));
	41	}
	42	len += ret;
	43	} while (len < min);
	44	return buffer;
	45	}
74536958	46
67e5a5ec LT	47	static void use(int bytes)
	48	{
	49	if (bytes > len)
	50	die("used more bytes than were available");
	51	len -= bytes;
	52	offset += bytes;
	53	}
bad50dc8	54
67e5a5ec	55	static void *get_data(unsigned long size)
bad50dc8	56	{
67e5a5ec LT	57	z_stream stream;
	58	void *buf = xmalloc(size);
	59
67e5a5ec LT	60	memset(&stream, 0, sizeof(stream));
	61
	62	stream.next_out = buf;
	63	stream.avail_out = size;
	64	stream.next_in = fill(1);
	65	stream.avail_in = len;
	66	inflateInit(&stream);
	67
	68	for (;;) {
	69	int ret = inflate(&stream, 0);
	70	use(len - stream.avail_in);
	71	if (stream.total_out == size && ret == Z_STREAM_END)
	72	break;
	73	if (ret != Z_OK)
	74	die("inflate returned %d\n", ret);
	75	stream.next_in = fill(1);
	76	stream.avail_in = len;
	77	}
	78	return buf;
bad50dc8 LT	79	}
bad50dc8 LT	80
67e5a5ec LT	81	struct delta_info {
	82	unsigned char base_sha1[20];
	83	unsigned long size;
	84	void *delta;
	85	struct delta_info *next;
	86	};
	87
	88	static struct delta_info *delta_list;
	89
	90	static void add_delta_to_list(unsigned char base_sha1, void delta, unsigned long size)
bad50dc8	91	{
67e5a5ec	92	struct delta_info info = xmalloc(sizeof(info));
bad50dc8	93
67e5a5ec LT	94	memcpy(info->base_sha1, base_sha1, 20);
	95	info->size = size;
	96	info->delta = delta;
	97	info->next = delta_list;
	98	delta_list = info;
bad50dc8 LT	99	}
bad50dc8 LT	100
cca7081a	101	static void added_object(unsigned char sha1, const char type, void *data, unsigned long size);
67e5a5ec	102
cca7081a LT	103	static void write_object(void buf, unsigned long size, const char type)
	104	{
	105	unsigned char sha1[20];
	106	if (write_sha1_file(buf, size, type, sha1) < 0)
	107	die("failed to write object");
	108	added_object(sha1, type, buf, size);
	109	}
	110
	111	static int resolve_delta(const char *type,
67e5a5ec LT	112	void *base, unsigned long base_size,
67e5a5ec LT	113	void *delta, unsigned long delta_size)
bad50dc8	114	{
67e5a5ec LT	115	void *result;
67e5a5ec LT	116	unsigned long result_size;
bad50dc8	117
67e5a5ec LT	118	result = patch_delta(base, base_size,
	119	delta, delta_size,
	120	&result_size);
	121	if (!result)
	122	die("failed to apply delta");
	123	free(delta);
cca7081a	124	write_object(result, result_size, type);
67e5a5ec	125	free(result);
bad50dc8 LT	126	return 0;
	127	}
	128
cca7081a	129	static void added_object(unsigned char sha1, const char type, void *data, unsigned long size)
74536958	130	{
67e5a5ec LT	131	struct delta_info **p = &delta_list;
	132	struct delta_info *info;
	133
	134	while ((info = *p) != NULL) {
	135	if (!memcmp(info->base_sha1, sha1, 20)) {
	136	*p = info->next;
	137	p = &delta_list;
	138	resolve_delta(type, data, size, info->delta, info->size);
	139	free(info);
	140	continue;
	141	}
	142	p = &info->next;
	143	}
	144	}
	145
	146	static int unpack_non_delta_entry(enum object_type kind, unsigned long size)
	147	{
	148	void *buf = get_data(size);
cca7081a	149	const char *type;
8ee378a0	150
8ee378a0	151	switch (kind) {
a733cb60 LT	152	case OBJ_COMMIT: type = "commit"; break;
	153	case OBJ_TREE: type = "tree"; break;
	154	case OBJ_BLOB: type = "blob"; break;
	155	case OBJ_TAG: type = "tag"; break;
67e5a5ec	156	default: die("bad type %d", kind);
8ee378a0	157	}
dddafffe LT	158	if (!dry_run)
dddafffe LT	159	write_object(buf, size, type);
67e5a5ec	160	free(buf);
8ee378a0 JH	161	return 0;
	162	}
	163
67e5a5ec	164	static int unpack_delta_entry(unsigned long delta_size)
8ee378a0	165	{
67e5a5ec LT	166	void delta_data, base;
67e5a5ec LT	167	unsigned long base_size;
8ee378a0	168	char type[20];
67e5a5ec	169	unsigned char base_sha1[20];
8ee378a0	170
67e5a5ec LT	171	memcpy(base_sha1, fill(20), 20);
67e5a5ec LT	172	use(20);
c4fb06c0	173
67e5a5ec	174	delta_data = get_data(delta_size);
dddafffe LT	175	if (dry_run) {
	176	free(delta_data);
	177	return 0;
	178	}
8ee378a0	179
c4fb06c0	180	if (!has_sha1_file(base_sha1)) {
67e5a5ec LT	181	add_delta_to_list(base_sha1, delta_data, delta_size);
67e5a5ec LT	182	return 0;
8ee378a0	183	}
c4fb06c0	184	base = read_sha1_file(base_sha1, type, &base_size);
8ee378a0	185	if (!base)
c4fb06c0	186	die("failed to read delta-pack base object %s", sha1_to_hex(base_sha1));
67e5a5ec	187	return resolve_delta(type, base, base_size, delta_data, delta_size);
8ee378a0 JH	188	}
8ee378a0 JH	189
d36f7b80	190	static void unpack_one(unsigned nr, unsigned total)
8ee378a0	191	{
01247d87	192	unsigned shift;
a733cb60	193	unsigned char *pack, c;
67e5a5ec LT	194	unsigned long size;
67e5a5ec LT	195	enum object_type type;
74536958	196
67e5a5ec LT	197	pack = fill(1);
	198	c = *pack;
	199	use(1);
a733cb60 LT	200	type = (c >> 4) & 7;
a733cb60 LT	201	size = (c & 15);
01247d87	202	shift = 4;
a733cb60	203	while (c & 0x80) {
67e5a5ec	204	pack = fill(1);
a733cb60	205	c = *pack++;
67e5a5ec	206	use(1);
01247d87 LT	207	size += (c & 0x7f) << shift;
01247d87 LT	208	shift += 7;
a733cb60	209	}
d36f7b80 LT	210	if (!quiet) {
	211	static unsigned long last_sec;
	212	static unsigned last_percent;
	213	struct timeval now;
cf219196	214	unsigned percentage = (nr * 100) / total;
d36f7b80 LT	215
	216	gettimeofday(&now, NULL);
	217	if (percentage != last_percent \|\| now.tv_sec != last_sec) {
	218	last_sec = now.tv_sec;
	219	last_percent = percentage;
	220	fprintf(stderr, "%4u%% (%u/%u) done\r", percentage, nr, total);
	221	}
	222	}
a733cb60 LT	223	switch (type) {
	224	case OBJ_COMMIT:
	225	case OBJ_TREE:
	226	case OBJ_BLOB:
	227	case OBJ_TAG:
67e5a5ec	228	unpack_non_delta_entry(type, size);
a733cb60 LT	229	return;
a733cb60 LT	230	case OBJ_DELTA:
67e5a5ec	231	unpack_delta_entry(size);
a733cb60	232	return;
67e5a5ec LT	233	default:
67e5a5ec LT	234	die("bad object type %d", type);
74536958 LT	235	}
	236	}
	237
	238	/*
	239	* We unpack from the end, older files first. Now, usually
	240	* there are deltas etc, so we'll not actually write the
	241	* objects in that order, but we might as well try..
	242	*/
	243	static void unpack_all(void)
	244	{
67e5a5ec LT	245	int i;
	246	struct pack_header *hdr = fill(sizeof(struct pack_header));
	247	unsigned version = ntohl(hdr->hdr_version);
	248	unsigned nr_objects = ntohl(hdr->hdr_entries);
	249
	250	if (ntohl(hdr->hdr_signature) != PACK_SIGNATURE)
	251	die("bad pack file");
01247d87	252	if (version != PACK_VERSION)
67e5a5ec LT	253	die("unable to handle pack file version %d", version);
	254	fprintf(stderr, "Unpacking %d objects\n", nr_objects);
	255
	256	use(sizeof(struct pack_header));
	257	for (i = 0; i < nr_objects; i++)
cf219196	258	unpack_one(i+1, nr_objects);
67e5a5ec LT	259	if (delta_list)
67e5a5ec LT	260	die("unresolved deltas left after unpacking");
74536958 LT	261	}
74536958 LT	262
bad50dc8 LT	263	int main(int argc, char **argv)
	264	{
	265	int i;
67e5a5ec	266	unsigned char sha1[20];
bad50dc8 LT	267
	268	for (i = 1 ; i < argc; i++) {
	269	const char *arg = argv[i];
	270
	271	if (*arg == '-') {
74536958 LT	272	if (!strcmp(arg, "-n")) {
	273	dry_run = 1;
	274	continue;
	275	}
d36f7b80 LT	276	if (!strcmp(arg, "-q")) {
	277	quiet = 1;
	278	continue;
	279	}
bad50dc8 LT	280	usage(unpack_usage);
bad50dc8 LT	281	}
67e5a5ec LT	282
67e5a5ec LT	283	/* We don't take any non-flag arguments now.. Maybe some day */
bad50dc8	284	usage(unpack_usage);
67e5a5ec LT	285	}
67e5a5ec LT	286	SHA1_Init(&ctx);
74536958	287	unpack_all();
67e5a5ec LT	288	SHA1_Update(&ctx, buffer, offset);
	289	SHA1_Final(sha1, &ctx);
	290	if (memcmp(fill(20), sha1, 20))
	291	die("final sha1 did not match");
	292	use(20);
	293
	294	/* Write the last part of the buffer to stdout */
	295	while (len) {
	296	int ret = write(1, buffer + offset, len);
	297	if (!ret)
	298	break;
	299	if (ret < 0) {
	300	if (errno == EAGAIN \|\| errno == EINTR)
	301	continue;
	302	break;
	303	}
	304	len -= ret;
	305	offset += ret;
	306	}
	307
	308	/* All done */
d36f7b80 LT	309	if (!quiet)
d36f7b80 LT	310	fprintf(stderr, "\n");
bad50dc8 LT	311	return 0;
bad50dc8 LT	312	}