development/git
development/git
1
0
Fork 0
mirror of https://github.com/git/git synced 2024-08-28 12:09:19 +00:00
Code Issues Packages Projects Releases Wiki Activity

Fix object packing/unpacking.

Browse source 
This actually successfully packed and unpacked a git archive down to
1.3MB (17MB unpacked).

Right now unpacking is way too noisy, lots of debug messages left.
This commit is contained in:
Linus Torvalds 2005-06-26 08:40:08 -07:00
parent 8ee378a0f0
commit c4fb06c0d0
2 changed files with 49 additions and 62 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

10
pack-objects.c
View file

@ -96,7 +96,7 @@ static unsigned long write_object(FILE *f, struct object_entry *entry)
unsigned long size; unsigned long size;
char type[10]; char type[10];
void *buf = read_sha1_file(entry->sha1, type, &size); void *buf = read_sha1_file(entry->sha1, type, &size);
char header[21]; char header[25];
unsigned hdrlen, datalen; unsigned hdrlen, datalen;
if (!buf) if (!buf)
@ -110,16 +110,16 @@ static unsigned long write_object(FILE *f, struct object_entry *entry)
* instead. * instead.
*/ */
header[0] = ".CTB"[entry->type]; header[0] = ".CTB"[entry->type];
datalen = htonl(size);
memcpy(header+1, &datalen, 4);
hdrlen = 5; hdrlen = 5;
if (entry->delta) { if (entry->delta) {
header[0] = 'D'; header[0] = 'D';
memcpy(header+1, entry->delta, 20); memcpy(header+5, entry->delta, 20);
buf = delta_against(buf, size, entry); buf = delta_against(buf, size, entry);
size = entry->delta_size; size = entry->delta_size;
hdrlen = 21; hdrlen = 25;
} }
datalen = htonl(size);
memcpy(header+1, &datalen, 4);
fwrite(header, hdrlen, 1, f); fwrite(header, hdrlen, 1, f);
datalen = fwrite_compressed(buf, size, f); datalen = fwrite_compressed(buf, size, f);
free(buf); free(buf);

101
unpack-objects.c
View file

@ -88,26 +88,26 @@ static int check_index(void)
} }
static int unpack_non_delta_entry(struct pack_entry *entry, static int unpack_non_delta_entry(struct pack_entry *entry,
unsigned char *pack) int kind,
unsigned char *data,
unsigned long size,
unsigned long left)
{ {
int st, kind; int st;
unsigned long size;
z_stream stream; z_stream stream;
char *buffer; char *buffer;
unsigned char sha1[20]; unsigned char sha1[20];
char *type_s; char *type_s;
unsigned long offset = ntohl(entry->offset);
kind = pack[0];
size = (pack[1] << 24) + (pack[2] << 16) + (pack[3] << 8) + pack[4];
printf("%s %c %lu\n", sha1_to_hex(entry->sha1), kind, size); printf("%s %c %lu\n", sha1_to_hex(entry->sha1), kind, size);
pack += 5; if (dry_run)
return 0;
buffer = xmalloc(size + 1); buffer = xmalloc(size + 1);
buffer[size] = 0; buffer[size] = 0;
memset(&stream, 0, sizeof(stream)); memset(&stream, 0, sizeof(stream));
stream.next_in = pack; stream.next_in = data;
stream.avail_in = pack_size - offset; /* sheesh. */ stream.avail_in = left;
stream.next_out = buffer; stream.next_out = buffer;
stream.avail_out = size; stream.avail_out = size;
@ -148,14 +148,15 @@ static int find_pack_entry(unsigned char *sha1, struct pack_entry **ent)
do { do {
int mi = (lo + hi) / 2; int mi = (lo + hi) / 2;
int cmp = memcmp(index + 24 * mi + 4, sha1, 20); int cmp = memcmp(index + 24 * mi + 4, sha1, 20);
printf("lo=%d mi=%d hi=%d cmp=%d\n", lo, mi, hi, cmp);
if (!cmp) { if (!cmp) {
*ent = index + 24 * mi; *ent = index + 24 * mi;
return 1; return 1;
} }
if (cmp < 0) if (cmp > 0)
hi = mi; hi = mi;
else else
lo = mi; lo = mi+1;
} while (lo < hi); } while (lo < hi);
return 0; return 0;
} }
@ -163,70 +164,55 @@ static int find_pack_entry(unsigned char *sha1, struct pack_entry **ent)
/* forward declaration for a mutually recursive function */ /* forward declaration for a mutually recursive function */
static void unpack_entry(struct pack_entry *); static void unpack_entry(struct pack_entry *);
static int unpack_delta_entry(struct pack_entry *entry, unsigned char *pack) static int unpack_delta_entry(struct pack_entry *entry,
unsigned char *base_sha1,
unsigned long delta_size,
unsigned long left)
{ {
void *delta_data, *result, *base; void *data, *delta_data, *result, *base;
unsigned long delta_alloc, delta_size, result_size, base_size; unsigned long data_size, result_size, base_size;
z_stream stream; z_stream stream;
int st; int st;
char type[20]; char type[20];
unsigned char sha1[20]; unsigned char sha1[20];
printf("%s D", sha1_to_hex(entry->sha1)); if (left < 20)
printf(" %s\n", sha1_to_hex(pack+1)); 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));
/* pack+1 is the base sha1, unless we have it, we need to if (dry_run)
return 0;
/* pack+5 is the base sha1, unless we have it, we need to
* unpack it first. * unpack it first.
*/ */
if (!has_sha1_file(pack+1)) { if (!has_sha1_file(base_sha1)) {
struct pack_entry *base; struct pack_entry *base;
if (!find_pack_entry(pack+1, &base)) if (!find_pack_entry(base_sha1, &base))
die("cannot find delta-pack base object"); die("cannot find delta-pack base object");
unpack_entry(base); unpack_entry(base);
} }
delta_data = xmalloc(delta_size);
/* pack+1 thru pack+20 is the base sha1 and
* pack+21 thru unknown number is the delta data.
* we do not even have size of the delta data uncompressed.
* sheesh!
*/
delta_alloc = 1024;
delta_data = xmalloc(delta_alloc);
memset(&stream, 0, sizeof(stream)); memset(&stream, 0, sizeof(stream));
stream.next_in = pack + 21; stream.next_in = data;
stream.avail_in = pack_size - ntohl(entry->offset); /* sheesh. */ stream.avail_in = data_size;
stream.next_out = delta_data; stream.next_out = delta_data;
stream.avail_out = delta_alloc; stream.avail_out = delta_size;
delta_size = 0;
inflateInit(&stream); inflateInit(&stream);
while (1) { st = inflate(&stream, Z_FINISH);
st = inflate(&stream, Z_FINISH);
if (st == Z_STREAM_END) {
delta_size = stream.total_out;
break;
}
if (st < 0)
break;
if (delta_alloc <= stream.total_out) {
delta_alloc = (delta_alloc +1024) * 3 / 2;
delta_data = xrealloc(delta_data, delta_alloc);
stream.next_out = delta_data + stream.total_out;
stream.avail_out = delta_alloc - stream.total_out;
}
}
inflateEnd(&stream); inflateEnd(&stream);
if (st != Z_STREAM_END) { if ((st != Z_STREAM_END) || stream.total_out != delta_size)
free(delta_data); die("delta data unpack failed");
return -1;
}
base = read_sha1_file(pack+1, type, &base_size); base = read_sha1_file(base_sha1, type, &base_size);
if (!base) if (!base)
die("failed to read delta-pack base object"); die("failed to read delta-pack base object %s", sha1_to_hex(base_sha1));
result = patch_delta(base, base_size, result = patch_delta(base, base_size,
delta_data, delta_size, delta_data, delta_size,
&result_size); &result_size);
@ -246,7 +232,7 @@ static int unpack_delta_entry(struct pack_entry *entry, unsigned char *pack)
static void unpack_entry(struct pack_entry *entry) static void unpack_entry(struct pack_entry *entry)
{ {
unsigned long offset; unsigned long offset, size, left;
unsigned char *pack; unsigned char *pack;
/* Have we done this one already due to deltas based on it? */ /* Have we done this one already due to deltas based on it? */
@ -257,13 +243,14 @@ static void unpack_entry(struct pack_entry *entry)
if (offset > pack_size - 5) if (offset > pack_size - 5)
die("object offset outside of pack file"); die("object offset outside of pack file");
pack = pack_base + offset; pack = pack_base + offset;
offset = pack_size - offset; size = (pack[1] << 24) + (pack[2] << 16) + (pack[3] << 8) + pack[4];
left = pack_size - offset - 5;
switch (*pack) { switch (*pack) {
case 'C': case 'T': case 'B': case 'C': case 'T': case 'B':
unpack_non_delta_entry(entry, pack); unpack_non_delta_entry(entry, *pack, pack+5, size, left);
break; break;
case 'D': case 'D':
unpack_delta_entry(entry, pack); unpack_delta_entry(entry, pack+5, size, left);
break; break;
default: default:
die("corrupted pack file"); die("corrupted pack file");