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

#include "cache.h"
#include "object.h"
#include "delta.h"
#include "pack.h"
#include "blob.h"
#include "commit.h"
#include "tag.h"
#include "tree.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_type; break;
	case OBJ_TREE:   type = tree_type; break;
	case OBJ_BLOB:   type = blob_type; break;
	case OBJ_TAG:    type = tag_type; 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);
	}
}

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"
8e440259 PE	5	#include "blob.h"
	6	#include "commit.h"
	7	#include "tag.h"
	8	#include "tree.h"
bad50dc8	9
d36f7b80 LT	10	#include <sys/time.h>
	11
	12	static int dry_run, quiet;
b2309b70	13	static const char unpack_usage[] = "git-unpack-objects [-n] [-q] < pack-file";
bad50dc8	14
67e5a5ec LT	15	/* We always read in 4kB chunks. */
	16	static unsigned char buffer[4096];
	17	static unsigned long offset, len, eof;
	18	static SHA_CTX ctx;
bad50dc8	19
67e5a5ec LT	20	/*
	21	* Make sure at least "min" bytes are available in the buffer, and
	22	* return the pointer to the buffer.
	23	*/
	24	static void * fill(int min)
	25	{
	26	if (min <= len)
	27	return buffer + offset;
	28	if (eof)
	29	die("unable to fill input");
	30	if (min > sizeof(buffer))
	31	die("cannot fill %d bytes", min);
	32	if (offset) {
	33	SHA1_Update(&ctx, buffer, offset);
	34	memcpy(buffer, buffer + offset, len);
	35	offset = 0;
	36	}
	37	do {
1c15afb9	38	int ret = xread(0, buffer + len, sizeof(buffer) - len);
67e5a5ec LT	39	if (ret <= 0) {
	40	if (!ret)
	41	die("early EOF");
67e5a5ec LT	42	die("read error on input: %s", strerror(errno));
	43	}
	44	len += ret;
	45	} while (len < min);
	46	return buffer;
	47	}
74536958	48
67e5a5ec LT	49	static void use(int bytes)
	50	{
	51	if (bytes > len)
	52	die("used more bytes than were available");
	53	len -= bytes;
	54	offset += bytes;
	55	}
bad50dc8	56
67e5a5ec	57	static void *get_data(unsigned long size)
bad50dc8	58	{
67e5a5ec LT	59	z_stream stream;
	60	void *buf = xmalloc(size);
	61
67e5a5ec LT	62	memset(&stream, 0, sizeof(stream));
	63
	64	stream.next_out = buf;
	65	stream.avail_out = size;
	66	stream.next_in = fill(1);
	67	stream.avail_in = len;
	68	inflateInit(&stream);
	69
	70	for (;;) {
	71	int ret = inflate(&stream, 0);
	72	use(len - stream.avail_in);
	73	if (stream.total_out == size && ret == Z_STREAM_END)
	74	break;
	75	if (ret != Z_OK)
	76	die("inflate returned %d\n", ret);
	77	stream.next_in = fill(1);
	78	stream.avail_in = len;
	79	}
ee639140	80	inflateEnd(&stream);
67e5a5ec	81	return buf;
bad50dc8 LT	82	}
bad50dc8 LT	83
67e5a5ec LT	84	struct delta_info {
	85	unsigned char base_sha1[20];
	86	unsigned long size;
	87	void *delta;
	88	struct delta_info *next;
	89	};
	90
	91	static struct delta_info *delta_list;
	92
	93	static void add_delta_to_list(unsigned char base_sha1, void delta, unsigned long size)
bad50dc8	94	{
67e5a5ec	95	struct delta_info info = xmalloc(sizeof(info));
bad50dc8	96
67e5a5ec LT	97	memcpy(info->base_sha1, base_sha1, 20);
	98	info->size = size;
	99	info->delta = delta;
	100	info->next = delta_list;
	101	delta_list = info;
bad50dc8 LT	102	}
bad50dc8 LT	103
cca7081a	104	static void added_object(unsigned char sha1, const char type, void *data, unsigned long size);
67e5a5ec	105
cca7081a LT	106	static void write_object(void buf, unsigned long size, const char type)
	107	{
	108	unsigned char sha1[20];
	109	if (write_sha1_file(buf, size, type, sha1) < 0)
	110	die("failed to write object");
	111	added_object(sha1, type, buf, size);
	112	}
	113
	114	static int resolve_delta(const char *type,
67e5a5ec LT	115	void *base, unsigned long base_size,
67e5a5ec LT	116	void *delta, unsigned long delta_size)
bad50dc8	117	{
67e5a5ec LT	118	void *result;
67e5a5ec LT	119	unsigned long result_size;
bad50dc8	120
67e5a5ec LT	121	result = patch_delta(base, base_size,
	122	delta, delta_size,
	123	&result_size);
	124	if (!result)
	125	die("failed to apply delta");
	126	free(delta);
cca7081a	127	write_object(result, result_size, type);
67e5a5ec	128	free(result);
bad50dc8 LT	129	return 0;
	130	}
	131
cca7081a	132	static void added_object(unsigned char sha1, const char type, void *data, unsigned long size)
74536958	133	{
67e5a5ec LT	134	struct delta_info **p = &delta_list;
	135	struct delta_info *info;
	136
	137	while ((info = *p) != NULL) {
	138	if (!memcmp(info->base_sha1, sha1, 20)) {
	139	*p = info->next;
	140	p = &delta_list;
	141	resolve_delta(type, data, size, info->delta, info->size);
	142	free(info);
	143	continue;
	144	}
	145	p = &info->next;
	146	}
	147	}
	148
	149	static int unpack_non_delta_entry(enum object_type kind, unsigned long size)
	150	{
	151	void *buf = get_data(size);
cca7081a	152	const char *type;
8ee378a0	153
8ee378a0	154	switch (kind) {
8e440259 PE	155	case OBJ_COMMIT: type = commit_type; break;
	156	case OBJ_TREE: type = tree_type; break;
	157	case OBJ_BLOB: type = blob_type; break;
	158	case OBJ_TAG: type = tag_type; break;
67e5a5ec	159	default: die("bad type %d", kind);
8ee378a0	160	}
dddafffe LT	161	if (!dry_run)
dddafffe LT	162	write_object(buf, size, type);
67e5a5ec	163	free(buf);
8ee378a0 JH	164	return 0;
	165	}
	166
67e5a5ec	167	static int unpack_delta_entry(unsigned long delta_size)
8ee378a0	168	{
67e5a5ec LT	169	void delta_data, base;
67e5a5ec LT	170	unsigned long base_size;
8ee378a0	171	char type[20];
67e5a5ec	172	unsigned char base_sha1[20];
ee639140	173	int result;
8ee378a0	174
67e5a5ec LT	175	memcpy(base_sha1, fill(20), 20);
67e5a5ec LT	176	use(20);
c4fb06c0	177
67e5a5ec	178	delta_data = get_data(delta_size);
dddafffe LT	179	if (dry_run) {
	180	free(delta_data);
	181	return 0;
	182	}
8ee378a0	183
c4fb06c0	184	if (!has_sha1_file(base_sha1)) {
67e5a5ec LT	185	add_delta_to_list(base_sha1, delta_data, delta_size);
67e5a5ec LT	186	return 0;
8ee378a0	187	}
c4fb06c0	188	base = read_sha1_file(base_sha1, type, &base_size);
8ee378a0	189	if (!base)
c4fb06c0	190	die("failed to read delta-pack base object %s", sha1_to_hex(base_sha1));
ee639140 SV	191	result = resolve_delta(type, base, base_size, delta_data, delta_size);
	192	free(base);
	193	return result;
8ee378a0 JH	194	}
8ee378a0 JH	195
d36f7b80	196	static void unpack_one(unsigned nr, unsigned total)
8ee378a0	197	{
01247d87	198	unsigned shift;
a733cb60	199	unsigned char *pack, c;
67e5a5ec LT	200	unsigned long size;
67e5a5ec LT	201	enum object_type type;
74536958	202
67e5a5ec LT	203	pack = fill(1);
	204	c = *pack;
	205	use(1);
a733cb60 LT	206	type = (c >> 4) & 7;
a733cb60 LT	207	size = (c & 15);
01247d87	208	shift = 4;
a733cb60	209	while (c & 0x80) {
67e5a5ec	210	pack = fill(1);
a733cb60	211	c = *pack++;
67e5a5ec	212	use(1);
01247d87 LT	213	size += (c & 0x7f) << shift;
01247d87 LT	214	shift += 7;
a733cb60	215	}
d36f7b80 LT	216	if (!quiet) {
	217	static unsigned long last_sec;
	218	static unsigned last_percent;
	219	struct timeval now;
cf219196	220	unsigned percentage = (nr * 100) / total;
d36f7b80 LT	221
	222	gettimeofday(&now, NULL);
	223	if (percentage != last_percent \|\| now.tv_sec != last_sec) {
	224	last_sec = now.tv_sec;
	225	last_percent = percentage;
	226	fprintf(stderr, "%4u%% (%u/%u) done\r", percentage, nr, total);
	227	}
	228	}
a733cb60 LT	229	switch (type) {
	230	case OBJ_COMMIT:
	231	case OBJ_TREE:
	232	case OBJ_BLOB:
	233	case OBJ_TAG:
67e5a5ec	234	unpack_non_delta_entry(type, size);
a733cb60 LT	235	return;
a733cb60 LT	236	case OBJ_DELTA:
67e5a5ec	237	unpack_delta_entry(size);
a733cb60	238	return;
67e5a5ec LT	239	default:
67e5a5ec LT	240	die("bad object type %d", type);
74536958 LT	241	}
	242	}
	243
74536958 LT	244	static void unpack_all(void)
74536958 LT	245	{
67e5a5ec LT	246	int i;
67e5a5ec LT	247	struct pack_header *hdr = fill(sizeof(struct pack_header));
67e5a5ec LT	248	unsigned nr_objects = ntohl(hdr->hdr_entries);
	249
	250	if (ntohl(hdr->hdr_signature) != PACK_SIGNATURE)
	251	die("bad pack file");
d60fc1c8 NP	252	if (!pack_version_ok(hdr->hdr_version))
d60fc1c8 NP	253	die("unknown pack file version %d", ntohl(hdr->hdr_version));
67e5a5ec LT	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	267
53228a5f JH	268	setup_git_directory();
53228a5f JH	269
476e8011 JH	270	quiet = !isatty(2);
476e8011 JH	271
bad50dc8 LT	272	for (i = 1 ; i < argc; i++) {
	273	const char *arg = argv[i];
	274
	275	if (*arg == '-') {
74536958 LT	276	if (!strcmp(arg, "-n")) {
	277	dry_run = 1;
	278	continue;
	279	}
d36f7b80 LT	280	if (!strcmp(arg, "-q")) {
	281	quiet = 1;
	282	continue;
	283	}
bad50dc8 LT	284	usage(unpack_usage);
bad50dc8 LT	285	}
67e5a5ec LT	286
67e5a5ec LT	287	/* We don't take any non-flag arguments now.. Maybe some day */
bad50dc8	288	usage(unpack_usage);
67e5a5ec LT	289	}
67e5a5ec LT	290	SHA1_Init(&ctx);
74536958	291	unpack_all();
67e5a5ec LT	292	SHA1_Update(&ctx, buffer, offset);
	293	SHA1_Final(sha1, &ctx);
	294	if (memcmp(fill(20), sha1, 20))
	295	die("final sha1 did not match");
	296	use(20);
	297
	298	/* Write the last part of the buffer to stdout */
	299	while (len) {
1c15afb9 JH	300	int ret = xwrite(1, buffer + offset, len);
1c15afb9 JH	301	if (ret <= 0)
67e5a5ec	302	break;
67e5a5ec LT	303	len -= ret;
	304	offset += ret;
	305	}
	306
	307	/* All done */
d36f7b80 LT	308	if (!quiet)
d36f7b80 LT	309	fprintf(stderr, "\n");
bad50dc8 LT	310	return 0;
bad50dc8 LT	311	}