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

#include "builtin.h"
#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));

	hashcpy(info->base_sha1, base_sha1);
	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 (!hashcmp(info->base_sha1, sha1)) {
			*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;

	hashcpy(base_sha1, fill(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 cmd_unpack_objects(int argc, const char **argv, const char *prefix)
{
	int i;
	unsigned char sha1[20];

	git_config(git_default_config);

	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 (hashcmp(fill(20), sha1))
		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
64413630	1	#include "builtin.h"
bad50dc8	2	#include "cache.h"
74536958	3	#include "object.h"
8ee378a0	4	#include "delta.h"
a733cb60	5	#include "pack.h"
8e440259 PE	6	#include "blob.h"
	7	#include "commit.h"
	8	#include "tag.h"
	9	#include "tree.h"
bad50dc8	10
d36f7b80 LT	11	#include <sys/time.h>
	12
	13	static int dry_run, quiet;
b2309b70	14	static const char unpack_usage[] = "git-unpack-objects [-n] [-q] < pack-file";
bad50dc8	15
67e5a5ec LT	16	/* We always read in 4kB chunks. */
	17	static unsigned char buffer[4096];
	18	static unsigned long offset, len, eof;
	19	static SHA_CTX ctx;
bad50dc8	20
67e5a5ec LT	21	/*
	22	* Make sure at least "min" bytes are available in the buffer, and
	23	* return the pointer to the buffer.
	24	*/
	25	static void * fill(int min)
	26	{
	27	if (min <= len)
	28	return buffer + offset;
	29	if (eof)
	30	die("unable to fill input");
	31	if (min > sizeof(buffer))
	32	die("cannot fill %d bytes", min);
	33	if (offset) {
	34	SHA1_Update(&ctx, buffer, offset);
	35	memcpy(buffer, buffer + offset, len);
	36	offset = 0;
	37	}
	38	do {
1c15afb9	39	int ret = xread(0, buffer + len, sizeof(buffer) - len);
67e5a5ec LT	40	if (ret <= 0) {
	41	if (!ret)
	42	die("early EOF");
67e5a5ec LT	43	die("read error on input: %s", strerror(errno));
	44	}
	45	len += ret;
	46	} while (len < min);
	47	return buffer;
	48	}
74536958	49
67e5a5ec LT	50	static void use(int bytes)
	51	{
	52	if (bytes > len)
	53	die("used more bytes than were available");
	54	len -= bytes;
	55	offset += bytes;
	56	}
bad50dc8	57
67e5a5ec	58	static void *get_data(unsigned long size)
bad50dc8	59	{
67e5a5ec LT	60	z_stream stream;
	61	void *buf = xmalloc(size);
	62
67e5a5ec LT	63	memset(&stream, 0, sizeof(stream));
	64
	65	stream.next_out = buf;
	66	stream.avail_out = size;
	67	stream.next_in = fill(1);
	68	stream.avail_in = len;
	69	inflateInit(&stream);
	70
	71	for (;;) {
	72	int ret = inflate(&stream, 0);
	73	use(len - stream.avail_in);
	74	if (stream.total_out == size && ret == Z_STREAM_END)
	75	break;
	76	if (ret != Z_OK)
	77	die("inflate returned %d\n", ret);
	78	stream.next_in = fill(1);
	79	stream.avail_in = len;
	80	}
ee639140	81	inflateEnd(&stream);
67e5a5ec	82	return buf;
bad50dc8 LT	83	}
bad50dc8 LT	84
67e5a5ec LT	85	struct delta_info {
	86	unsigned char base_sha1[20];
	87	unsigned long size;
	88	void *delta;
	89	struct delta_info *next;
	90	};
	91
	92	static struct delta_info *delta_list;
	93
	94	static void add_delta_to_list(unsigned char base_sha1, void delta, unsigned long size)
bad50dc8	95	{
67e5a5ec	96	struct delta_info info = xmalloc(sizeof(info));
bad50dc8	97
e702496e	98	hashcpy(info->base_sha1, base_sha1);
67e5a5ec LT	99	info->size = size;
	100	info->delta = delta;
	101	info->next = delta_list;
	102	delta_list = info;
bad50dc8 LT	103	}
bad50dc8 LT	104
cca7081a	105	static void added_object(unsigned char sha1, const char type, void *data, unsigned long size);
67e5a5ec	106
cca7081a LT	107	static void write_object(void buf, unsigned long size, const char type)
	108	{
	109	unsigned char sha1[20];
	110	if (write_sha1_file(buf, size, type, sha1) < 0)
	111	die("failed to write object");
	112	added_object(sha1, type, buf, size);
	113	}
	114
	115	static int resolve_delta(const char *type,
64413630	116	void *base, unsigned long base_size,
67e5a5ec	117	void *delta, unsigned long delta_size)
bad50dc8	118	{
67e5a5ec LT	119	void *result;
67e5a5ec LT	120	unsigned long result_size;
bad50dc8	121
67e5a5ec LT	122	result = patch_delta(base, base_size,
	123	delta, delta_size,
	124	&result_size);
	125	if (!result)
	126	die("failed to apply delta");
	127	free(delta);
cca7081a	128	write_object(result, result_size, type);
67e5a5ec	129	free(result);
bad50dc8 LT	130	return 0;
	131	}
	132
cca7081a	133	static void added_object(unsigned char sha1, const char type, void *data, unsigned long size)
74536958	134	{
67e5a5ec LT	135	struct delta_info **p = &delta_list;
	136	struct delta_info *info;
	137
	138	while ((info = *p) != NULL) {
a89fccd2	139	if (!hashcmp(info->base_sha1, sha1)) {
67e5a5ec LT	140	*p = info->next;
	141	p = &delta_list;
	142	resolve_delta(type, data, size, info->delta, info->size);
	143	free(info);
	144	continue;
	145	}
	146	p = &info->next;
	147	}
	148	}
	149
	150	static int unpack_non_delta_entry(enum object_type kind, unsigned long size)
	151	{
	152	void *buf = get_data(size);
cca7081a	153	const char *type;
8ee378a0	154
8ee378a0	155	switch (kind) {
8e440259 PE	156	case OBJ_COMMIT: type = commit_type; break;
	157	case OBJ_TREE: type = tree_type; break;
	158	case OBJ_BLOB: type = blob_type; break;
	159	case OBJ_TAG: type = tag_type; break;
67e5a5ec	160	default: die("bad type %d", kind);
8ee378a0	161	}
dddafffe LT	162	if (!dry_run)
dddafffe LT	163	write_object(buf, size, type);
67e5a5ec	164	free(buf);
8ee378a0 JH	165	return 0;
	166	}
	167
67e5a5ec	168	static int unpack_delta_entry(unsigned long delta_size)
8ee378a0	169	{
67e5a5ec LT	170	void delta_data, base;
67e5a5ec LT	171	unsigned long base_size;
8ee378a0	172	char type[20];
67e5a5ec	173	unsigned char base_sha1[20];
ee639140	174	int result;
8ee378a0	175
e702496e	176	hashcpy(base_sha1, fill(20));
67e5a5ec	177	use(20);
c4fb06c0	178
67e5a5ec	179	delta_data = get_data(delta_size);
dddafffe LT	180	if (dry_run) {
	181	free(delta_data);
	182	return 0;
	183	}
8ee378a0	184
c4fb06c0	185	if (!has_sha1_file(base_sha1)) {
67e5a5ec LT	186	add_delta_to_list(base_sha1, delta_data, delta_size);
67e5a5ec LT	187	return 0;
8ee378a0	188	}
c4fb06c0	189	base = read_sha1_file(base_sha1, type, &base_size);
8ee378a0	190	if (!base)
c4fb06c0	191	die("failed to read delta-pack base object %s", sha1_to_hex(base_sha1));
ee639140 SV	192	result = resolve_delta(type, base, base_size, delta_data, delta_size);
	193	free(base);
	194	return result;
8ee378a0 JH	195	}
8ee378a0 JH	196
d36f7b80	197	static void unpack_one(unsigned nr, unsigned total)
8ee378a0	198	{
01247d87	199	unsigned shift;
a733cb60	200	unsigned char *pack, c;
67e5a5ec LT	201	unsigned long size;
67e5a5ec LT	202	enum object_type type;
74536958	203
67e5a5ec LT	204	pack = fill(1);
	205	c = *pack;
	206	use(1);
a733cb60 LT	207	type = (c >> 4) & 7;
a733cb60 LT	208	size = (c & 15);
01247d87	209	shift = 4;
a733cb60	210	while (c & 0x80) {
67e5a5ec	211	pack = fill(1);
a733cb60	212	c = *pack++;
67e5a5ec	213	use(1);
01247d87 LT	214	size += (c & 0x7f) << shift;
01247d87 LT	215	shift += 7;
a733cb60	216	}
d36f7b80 LT	217	if (!quiet) {
	218	static unsigned long last_sec;
	219	static unsigned last_percent;
	220	struct timeval now;
cf219196	221	unsigned percentage = (nr * 100) / total;
d36f7b80 LT	222
	223	gettimeofday(&now, NULL);
	224	if (percentage != last_percent \|\| now.tv_sec != last_sec) {
	225	last_sec = now.tv_sec;
	226	last_percent = percentage;
	227	fprintf(stderr, "%4u%% (%u/%u) done\r", percentage, nr, total);
	228	}
	229	}
a733cb60 LT	230	switch (type) {
	231	case OBJ_COMMIT:
	232	case OBJ_TREE:
	233	case OBJ_BLOB:
	234	case OBJ_TAG:
67e5a5ec	235	unpack_non_delta_entry(type, size);
a733cb60 LT	236	return;
a733cb60 LT	237	case OBJ_DELTA:
67e5a5ec	238	unpack_delta_entry(size);
a733cb60	239	return;
67e5a5ec LT	240	default:
67e5a5ec LT	241	die("bad object type %d", type);
74536958 LT	242	}
	243	}
	244
74536958 LT	245	static void unpack_all(void)
74536958 LT	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
64413630	264	int cmd_unpack_objects(int argc, const char *argv, const char prefix)
bad50dc8 LT	265	{
bad50dc8 LT	266	int i;
67e5a5ec	267	unsigned char sha1[20];
bad50dc8	268
8e273641	269	git_config(git_default_config);
53228a5f	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);
67e5a5ec LT	294	SHA1_Final(sha1, &ctx);
a89fccd2	295	if (hashcmp(fill(20), sha1))
67e5a5ec LT	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	}