[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 [-n] [-q] < 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 = xread(0, buffer + len, sizeof(buffer) - len);
		if (ret <= 0) {
			if (!ret)
				die("early EOF");
			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;
	}
	inflateEnd(&stream);
	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];
	int result;

	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));
	result = resolve_delta(type, base, base_size, delta_data, delta_size);
	free(base);
	return result;
}

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 nr_objects = ntohl(hdr->hdr_entries);

	if (ntohl(hdr->hdr_signature) != PACK_SIGNATURE)
		die("bad pack file");
	if (!pack_version_ok(hdr->hdr_version))
		die("unknown pack file version %d", ntohl(hdr->hdr_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];

	setup_git_directory();

	quiet = !isatty(2);

	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 = xwrite(1, buffer + offset, len);
		if (ret <= 0)
			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;
b2309b70	9	static const char unpack_usage[] = "git-unpack-objects [-n] [-q] < 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 {
1c15afb9	34	int ret = xread(0, buffer + len, sizeof(buffer) - len);
67e5a5ec LT	35	if (ret <= 0) {
	36	if (!ret)
	37	die("early EOF");
67e5a5ec LT	38	die("read error on input: %s", strerror(errno));
	39	}
	40	len += ret;
	41	} while (len < min);
	42	return buffer;
	43	}
74536958	44
67e5a5ec LT	45	static void use(int bytes)
	46	{
	47	if (bytes > len)
	48	die("used more bytes than were available");
	49	len -= bytes;
	50	offset += bytes;
	51	}
bad50dc8	52
67e5a5ec	53	static void *get_data(unsigned long size)
bad50dc8	54	{
67e5a5ec LT	55	z_stream stream;
	56	void *buf = xmalloc(size);
	57
67e5a5ec LT	58	memset(&stream, 0, sizeof(stream));
	59
	60	stream.next_out = buf;
	61	stream.avail_out = size;
	62	stream.next_in = fill(1);
	63	stream.avail_in = len;
	64	inflateInit(&stream);
	65
	66	for (;;) {
	67	int ret = inflate(&stream, 0);
	68	use(len - stream.avail_in);
	69	if (stream.total_out == size && ret == Z_STREAM_END)
	70	break;
	71	if (ret != Z_OK)
	72	die("inflate returned %d\n", ret);
	73	stream.next_in = fill(1);
	74	stream.avail_in = len;
	75	}
ee639140	76	inflateEnd(&stream);
67e5a5ec	77	return buf;
bad50dc8 LT	78	}
bad50dc8 LT	79
67e5a5ec LT	80	struct delta_info {
	81	unsigned char base_sha1[20];
	82	unsigned long size;
	83	void *delta;
	84	struct delta_info *next;
	85	};
	86
	87	static struct delta_info *delta_list;
	88
	89	static void add_delta_to_list(unsigned char base_sha1, void delta, unsigned long size)
bad50dc8	90	{
67e5a5ec	91	struct delta_info info = xmalloc(sizeof(info));
bad50dc8	92
67e5a5ec LT	93	memcpy(info->base_sha1, base_sha1, 20);
	94	info->size = size;
	95	info->delta = delta;
	96	info->next = delta_list;
	97	delta_list = info;
bad50dc8 LT	98	}
bad50dc8 LT	99
cca7081a	100	static void added_object(unsigned char sha1, const char type, void *data, unsigned long size);
67e5a5ec	101
cca7081a LT	102	static void write_object(void buf, unsigned long size, const char type)
	103	{
	104	unsigned char sha1[20];
	105	if (write_sha1_file(buf, size, type, sha1) < 0)
	106	die("failed to write object");
	107	added_object(sha1, type, buf, size);
	108	}
	109
	110	static int resolve_delta(const char *type,
67e5a5ec LT	111	void *base, unsigned long base_size,
67e5a5ec LT	112	void *delta, unsigned long delta_size)
bad50dc8	113	{
67e5a5ec LT	114	void *result;
67e5a5ec LT	115	unsigned long result_size;
bad50dc8	116
67e5a5ec LT	117	result = patch_delta(base, base_size,
	118	delta, delta_size,
	119	&result_size);
	120	if (!result)
	121	die("failed to apply delta");
	122	free(delta);
cca7081a	123	write_object(result, result_size, type);
67e5a5ec	124	free(result);
bad50dc8 LT	125	return 0;
	126	}
	127
cca7081a	128	static void added_object(unsigned char sha1, const char type, void *data, unsigned long size)
74536958	129	{
67e5a5ec LT	130	struct delta_info **p = &delta_list;
	131	struct delta_info *info;
	132
	133	while ((info = *p) != NULL) {
	134	if (!memcmp(info->base_sha1, sha1, 20)) {
	135	*p = info->next;
	136	p = &delta_list;
	137	resolve_delta(type, data, size, info->delta, info->size);
	138	free(info);
	139	continue;
	140	}
	141	p = &info->next;
	142	}
	143	}
	144
	145	static int unpack_non_delta_entry(enum object_type kind, unsigned long size)
	146	{
	147	void *buf = get_data(size);
cca7081a	148	const char *type;
8ee378a0	149
8ee378a0	150	switch (kind) {
a733cb60 LT	151	case OBJ_COMMIT: type = "commit"; break;
	152	case OBJ_TREE: type = "tree"; break;
	153	case OBJ_BLOB: type = "blob"; break;
	154	case OBJ_TAG: type = "tag"; break;
67e5a5ec	155	default: die("bad type %d", kind);
8ee378a0	156	}
dddafffe LT	157	if (!dry_run)
dddafffe LT	158	write_object(buf, size, type);
67e5a5ec	159	free(buf);
8ee378a0 JH	160	return 0;
	161	}
	162
67e5a5ec	163	static int unpack_delta_entry(unsigned long delta_size)
8ee378a0	164	{
67e5a5ec LT	165	void delta_data, base;
67e5a5ec LT	166	unsigned long base_size;
8ee378a0	167	char type[20];
67e5a5ec	168	unsigned char base_sha1[20];
ee639140	169	int result;
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));
ee639140 SV	187	result = resolve_delta(type, base, base_size, delta_data, delta_size);
	188	free(base);
	189	return result;
8ee378a0 JH	190	}
8ee378a0 JH	191
d36f7b80	192	static void unpack_one(unsigned nr, unsigned total)
8ee378a0	193	{
01247d87	194	unsigned shift;
a733cb60	195	unsigned char *pack, c;
67e5a5ec LT	196	unsigned long size;
67e5a5ec LT	197	enum object_type type;
74536958	198
67e5a5ec LT	199	pack = fill(1);
	200	c = *pack;
	201	use(1);
a733cb60 LT	202	type = (c >> 4) & 7;
a733cb60 LT	203	size = (c & 15);
01247d87	204	shift = 4;
a733cb60	205	while (c & 0x80) {
67e5a5ec	206	pack = fill(1);
a733cb60	207	c = *pack++;
67e5a5ec	208	use(1);
01247d87 LT	209	size += (c & 0x7f) << shift;
01247d87 LT	210	shift += 7;
a733cb60	211	}
d36f7b80 LT	212	if (!quiet) {
	213	static unsigned long last_sec;
	214	static unsigned last_percent;
	215	struct timeval now;
cf219196	216	unsigned percentage = (nr * 100) / total;
d36f7b80 LT	217
	218	gettimeofday(&now, NULL);
	219	if (percentage != last_percent \|\| now.tv_sec != last_sec) {
	220	last_sec = now.tv_sec;
	221	last_percent = percentage;
	222	fprintf(stderr, "%4u%% (%u/%u) done\r", percentage, nr, total);
	223	}
	224	}
a733cb60 LT	225	switch (type) {
	226	case OBJ_COMMIT:
	227	case OBJ_TREE:
	228	case OBJ_BLOB:
	229	case OBJ_TAG:
67e5a5ec	230	unpack_non_delta_entry(type, size);
a733cb60 LT	231	return;
a733cb60 LT	232	case OBJ_DELTA:
67e5a5ec	233	unpack_delta_entry(size);
a733cb60	234	return;
67e5a5ec LT	235	default:
67e5a5ec LT	236	die("bad object type %d", type);
74536958 LT	237	}
	238	}
	239
	240	/*
	241	* We unpack from the end, older files first. Now, usually
	242	* there are deltas etc, so we'll not actually write the
	243	* objects in that order, but we might as well try..
	244	*/
	245	static void unpack_all(void)
	246	{
67e5a5ec LT	247	int i;
67e5a5ec LT	248	struct pack_header *hdr = fill(sizeof(struct pack_header));
67e5a5ec LT	249	unsigned nr_objects = ntohl(hdr->hdr_entries);
	250
	251	if (ntohl(hdr->hdr_signature) != PACK_SIGNATURE)
	252	die("bad pack file");
d60fc1c8 NP	253	if (!pack_version_ok(hdr->hdr_version))
d60fc1c8 NP	254	die("unknown pack file version %d", ntohl(hdr->hdr_version));
67e5a5ec LT	255	fprintf(stderr, "Unpacking %d objects\n", nr_objects);
	256
	257	use(sizeof(struct pack_header));
	258	for (i = 0; i < nr_objects; i++)
cf219196	259	unpack_one(i+1, nr_objects);
67e5a5ec LT	260	if (delta_list)
67e5a5ec LT	261	die("unresolved deltas left after unpacking");
74536958 LT	262	}
74536958 LT	263
bad50dc8 LT	264	int main(int argc, char **argv)
	265	{
	266	int i;
67e5a5ec	267	unsigned char sha1[20];
bad50dc8	268
53228a5f JH	269	setup_git_directory();
53228a5f JH	270
476e8011 JH	271	quiet = !isatty(2);
476e8011 JH	272
bad50dc8 LT	273	for (i = 1 ; i < argc; i++) {
	274	const char *arg = argv[i];
	275
	276	if (*arg == '-') {
74536958 LT	277	if (!strcmp(arg, "-n")) {
	278	dry_run = 1;
	279	continue;
	280	}
d36f7b80 LT	281	if (!strcmp(arg, "-q")) {
	282	quiet = 1;
	283	continue;
	284	}
bad50dc8 LT	285	usage(unpack_usage);
bad50dc8 LT	286	}
67e5a5ec LT	287
67e5a5ec LT	288	/* We don't take any non-flag arguments now.. Maybe some day */
bad50dc8	289	usage(unpack_usage);
67e5a5ec LT	290	}
67e5a5ec LT	291	SHA1_Init(&ctx);
74536958	292	unpack_all();
67e5a5ec LT	293	SHA1_Update(&ctx, buffer, offset);
	294	SHA1_Final(sha1, &ctx);
	295	if (memcmp(fill(20), sha1, 20))
	296	die("final sha1 did not match");
	297	use(20);
	298
	299	/* Write the last part of the buffer to stdout */
	300	while (len) {
1c15afb9 JH	301	int ret = xwrite(1, buffer + offset, len);
1c15afb9 JH	302	if (ret <= 0)
67e5a5ec	303	break;
67e5a5ec LT	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	}