summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorLibravatar Junio C Hamano <junkio@cox.net>2006-10-29 13:44:54 -0800
committerLibravatar Junio C Hamano <junkio@cox.net>2006-10-29 13:44:54 -0800
commit01cff2d2ca77dc050973e1fc4702119a715cc919 (patch)
tree794f24dc8da73f5e976a10469d313cb96b59ff28
parentOnly repack active packs by skipping over kept packs. (diff)
parentindex-pack: minor fixes to comment and function name (diff)
downloadtgif-01cff2d2ca77dc050973e1fc4702119a715cc919.tar.xz
Merge early part of branch 'np/index-pack' into sp/keep-pack
-rw-r--r--Documentation/git-index-pack.txt23
-rw-r--r--fetch-clone.c219
-rw-r--r--fetch-pack.c2
-rw-r--r--index-pack.c476
4 files changed, 436 insertions, 284 deletions
diff --git a/Documentation/git-index-pack.txt b/Documentation/git-index-pack.txt
index 71ce557276..9fa4847d56 100644
--- a/Documentation/git-index-pack.txt
+++ b/Documentation/git-index-pack.txt
@@ -8,7 +8,8 @@ git-index-pack - Build pack index file for an existing packed archive
SYNOPSIS
--------
-'git-index-pack' [-o <index-file>] <pack-file>
+'git-index-pack' [-v] [-o <index-file>] <pack-file>
+'git-index-pack' --stdin [--fix-thin] [-v] [-o <index-file>] [<pack-file>]
DESCRIPTION
@@ -21,6 +22,9 @@ objects/pack/ directory of a git repository.
OPTIONS
-------
+-v::
+ Be verbose about what is going on, including progress status.
+
-o <index-file>::
Write the generated pack index into the specified
file. Without this option the name of pack index
@@ -29,6 +33,23 @@ OPTIONS
fails if the name of packed archive does not end
with .pack).
+--stdin::
+ When this flag is provided, the pack is read from stdin
+ instead and a copy is then written to <pack-file>. If
+ <pack-file> is not specified, the pack is written to
+ objects/pack/ directory of the current git repository with
+ a default name determined from the pack content.
+
+--fix-thin::
+ It is possible for gitlink:git-pack-objects[1] to build
+ "thin" pack, which records objects in deltified form based on
+ objects not included in the pack to reduce network traffic.
+ Those objects are expected to be present on the receiving end
+ and they must be included in the pack for that pack to be self
+ contained and indexable. Without this option any attempt to
+ index a thin pack will fail. This option only makes sense in
+ conjonction with --stdin.
+
Author
------
diff --git a/fetch-clone.c b/fetch-clone.c
index 76b99afcdb..96cdab43c9 100644
--- a/fetch-clone.c
+++ b/fetch-clone.c
@@ -3,97 +3,6 @@
#include "pkt-line.h"
#include "sideband.h"
#include <sys/wait.h>
-#include <sys/time.h>
-
-static int finish_pack(const char *pack_tmp_name, const char *me)
-{
- int pipe_fd[2];
- pid_t pid;
- char idx[PATH_MAX];
- char final[PATH_MAX];
- char hash[41];
- unsigned char sha1[20];
- char *cp;
- int err = 0;
-
- if (pipe(pipe_fd) < 0)
- die("%s: unable to set up pipe", me);
-
- strcpy(idx, pack_tmp_name); /* ".git/objects/pack-XXXXXX" */
- cp = strrchr(idx, '/');
- memcpy(cp, "/pidx", 5);
-
- pid = fork();
- if (pid < 0)
- die("%s: unable to fork off git-index-pack", me);
- if (!pid) {
- close(0);
- dup2(pipe_fd[1], 1);
- close(pipe_fd[0]);
- close(pipe_fd[1]);
- execl_git_cmd("index-pack", "-o", idx, pack_tmp_name, NULL);
- error("cannot exec git-index-pack <%s> <%s>",
- idx, pack_tmp_name);
- exit(1);
- }
- close(pipe_fd[1]);
- if (read(pipe_fd[0], hash, 40) != 40) {
- error("%s: unable to read from git-index-pack", me);
- err = 1;
- }
- close(pipe_fd[0]);
-
- for (;;) {
- int status, code;
-
- if (waitpid(pid, &status, 0) < 0) {
- if (errno == EINTR)
- continue;
- error("waitpid failed (%s)", strerror(errno));
- goto error_die;
- }
- if (WIFSIGNALED(status)) {
- int sig = WTERMSIG(status);
- error("git-index-pack died of signal %d", sig);
- goto error_die;
- }
- if (!WIFEXITED(status)) {
- error("git-index-pack died of unnatural causes %d",
- status);
- goto error_die;
- }
- code = WEXITSTATUS(status);
- if (code) {
- error("git-index-pack died with error code %d", code);
- goto error_die;
- }
- if (err)
- goto error_die;
- break;
- }
- hash[40] = 0;
- if (get_sha1_hex(hash, sha1)) {
- error("git-index-pack reported nonsense '%s'", hash);
- goto error_die;
- }
- /* Now we have pack in pack_tmp_name[], and
- * idx in idx[]; rename them to their final names.
- */
- snprintf(final, sizeof(final),
- "%s/pack/pack-%s.pack", get_object_directory(), hash);
- move_temp_to_file(pack_tmp_name, final);
- chmod(final, 0444);
- snprintf(final, sizeof(final),
- "%s/pack/pack-%s.idx", get_object_directory(), hash);
- move_temp_to_file(idx, final);
- chmod(final, 0444);
- return 0;
-
- error_die:
- unlink(idx);
- unlink(pack_tmp_name);
- exit(1);
-}
static pid_t setup_sideband(int sideband, const char *me, int fd[2], int xd[2])
{
@@ -128,7 +37,7 @@ static pid_t setup_sideband(int sideband, const char *me, int fd[2], int xd[2])
return side_pid;
}
-int receive_unpack_pack(int xd[2], const char *me, int quiet, int sideband)
+static int get_pack(int xd[2], const char *me, int sideband, const char **argv)
{
int status;
pid_t pid, side_pid;
@@ -142,135 +51,37 @@ int receive_unpack_pack(int xd[2], const char *me, int quiet, int sideband)
dup2(fd[0], 0);
close(fd[0]);
close(fd[1]);
- execl_git_cmd("unpack-objects", quiet ? "-q" : NULL, NULL);
- die("git-unpack-objects exec failed");
+ execv_git_cmd(argv);
+ die("%s exec failed", argv[0]);
}
close(fd[0]);
close(fd[1]);
while (waitpid(pid, &status, 0) < 0) {
if (errno != EINTR)
- die("waiting for git-unpack-objects: %s",
- strerror(errno));
+ die("waiting for %s: %s", argv[0], strerror(errno));
}
if (WIFEXITED(status)) {
int code = WEXITSTATUS(status);
if (code)
- die("git-unpack-objects died with error code %d",
- code);
+ die("%s died with error code %d", argv[0], code);
return 0;
}
if (WIFSIGNALED(status)) {
int sig = WTERMSIG(status);
- die("git-unpack-objects died of signal %d", sig);
+ die("%s died of signal %d", argv[0], sig);
}
- die("git-unpack-objects died of unnatural causes %d", status);
+ die("%s died of unnatural causes %d", argv[0], status);
}
-/*
- * We average out the download speed over this many "events", where
- * an event is a minimum of about half a second. That way, we get
- * a reasonably stable number.
- */
-#define NR_AVERAGE (4)
-
-/*
- * A "binary msec" is a power-of-two-msec, aka 1/1024th of a second.
- * Keeping the time in that format means that "bytes / msecs" means
- * the same as kB/s (modulo rounding).
- *
- * 1000512 is a magic number (usecs in a second, rounded up by half
- * of 1024, to make "rounding" come out right ;)
- */
-#define usec_to_binarymsec(x) ((int)(x) / (1000512 >> 10))
+int receive_unpack_pack(int xd[2], const char *me, int quiet, int sideband)
+{
+ const char *argv[3] = { "unpack-objects", quiet ? "-q" : NULL, NULL };
+ return get_pack(xd, me, sideband, argv);
+}
int receive_keep_pack(int xd[2], const char *me, int quiet, int sideband)
{
- char tmpfile[PATH_MAX];
- int ofd, ifd, fd[2];
- unsigned long total;
- static struct timeval prev_tv;
- struct average {
- unsigned long bytes;
- unsigned long time;
- } download[NR_AVERAGE] = { {0, 0}, };
- unsigned long avg_bytes, avg_time;
- int idx = 0;
-
- setup_sideband(sideband, me, fd, xd);
-
- ifd = fd[0];
- snprintf(tmpfile, sizeof(tmpfile),
- "%s/pack/tmp-XXXXXX", get_object_directory());
- ofd = mkstemp(tmpfile);
- if (ofd < 0)
- return error("unable to create temporary file %s", tmpfile);
-
- gettimeofday(&prev_tv, NULL);
- total = 0;
- avg_bytes = 0;
- avg_time = 0;
- while (1) {
- char buf[8192];
- ssize_t sz, wsz, pos;
- sz = read(ifd, buf, sizeof(buf));
- if (sz == 0)
- break;
- if (sz < 0) {
- if (errno != EINTR && errno != EAGAIN) {
- error("error reading pack (%s)", strerror(errno));
- close(ofd);
- unlink(tmpfile);
- return -1;
- }
- sz = 0;
- }
- pos = 0;
- while (pos < sz) {
- wsz = write(ofd, buf + pos, sz - pos);
- if (wsz < 0) {
- error("error writing pack (%s)",
- strerror(errno));
- close(ofd);
- unlink(tmpfile);
- return -1;
- }
- pos += wsz;
- }
- total += sz;
- if (!quiet) {
- static unsigned long last;
- struct timeval tv;
- unsigned long diff = total - last;
- /* not really "msecs", but a power-of-two millisec (1/1024th of a sec) */
- unsigned long msecs;
-
- gettimeofday(&tv, NULL);
- msecs = tv.tv_sec - prev_tv.tv_sec;
- msecs <<= 10;
- msecs += usec_to_binarymsec(tv.tv_usec - prev_tv.tv_usec);
-
- if (msecs > 500) {
- prev_tv = tv;
- last = total;
-
- /* Update averages ..*/
- avg_bytes += diff;
- avg_time += msecs;
- avg_bytes -= download[idx].bytes;
- avg_time -= download[idx].time;
- download[idx].bytes = diff;
- download[idx].time = msecs;
- idx++;
- if (idx >= NR_AVERAGE)
- idx = 0;
-
- fprintf(stderr, "%4lu.%03luMB (%lu kB/s) \r",
- total >> 20,
- 1000*((total >> 10) & 1023)>>10,
- avg_bytes / avg_time );
- }
- }
- }
- close(ofd);
- return finish_pack(tmpfile, me);
+ const char *argv[5] = { "index-pack", "--stdin", "--fix-thin",
+ quiet ? NULL : "-v", NULL };
+ return get_pack(xd, me, sideband, argv);
}
diff --git a/fetch-pack.c b/fetch-pack.c
index 474d54520e..8720ed42e9 100644
--- a/fetch-pack.c
+++ b/fetch-pack.c
@@ -518,8 +518,6 @@ int main(int argc, char **argv)
}
if (!dest)
usage(fetch_pack_usage);
- if (keep_pack)
- use_thin_pack = 0;
pid = git_connect(fd, dest, exec);
if (pid < 0)
return 1;
diff --git a/index-pack.c b/index-pack.c
index e33f60524f..866a054056 100644
--- a/index-pack.c
+++ b/index-pack.c
@@ -6,9 +6,11 @@
#include "commit.h"
#include "tag.h"
#include "tree.h"
+#include <sys/time.h>
+#include <signal.h>
static const char index_pack_usage[] =
-"git-index-pack [-o index-file] pack-file";
+"git-index-pack [-v] [-o <index-file>] { <pack-file> | --stdin [--fix-thin] [<pack-file>] }";
struct object_entry
{
@@ -33,37 +35,83 @@ union delta_base {
struct delta_entry
{
- struct object_entry *obj;
union delta_base base;
+ int obj_no;
};
-static const char *pack_name;
static struct object_entry *objects;
static struct delta_entry *deltas;
static int nr_objects;
static int nr_deltas;
+static int nr_resolved_deltas;
+
+static int from_stdin;
+static int verbose;
+
+static volatile sig_atomic_t progress_update;
+
+static void progress_interval(int signum)
+{
+ progress_update = 1;
+}
+
+static void setup_progress_signal(void)
+{
+ struct sigaction sa;
+ struct itimerval v;
+
+ memset(&sa, 0, sizeof(sa));
+ sa.sa_handler = progress_interval;
+ sigemptyset(&sa.sa_mask);
+ sa.sa_flags = SA_RESTART;
+ sigaction(SIGALRM, &sa, NULL);
+
+ v.it_interval.tv_sec = 1;
+ v.it_interval.tv_usec = 0;
+ v.it_value = v.it_interval;
+ setitimer(ITIMER_REAL, &v, NULL);
+
+}
+
+static unsigned display_progress(unsigned n, unsigned total, unsigned last_pc)
+{
+ unsigned percent = n * 100 / total;
+ if (percent != last_pc || progress_update) {
+ fprintf(stderr, "%4u%% (%u/%u) done\r", percent, n, total);
+ progress_update = 0;
+ }
+ return percent;
+}
/* We always read in 4kB chunks. */
static unsigned char input_buffer[4096];
static unsigned long input_offset, input_len, consumed_bytes;
static SHA_CTX input_ctx;
-static int input_fd;
+static int input_fd, output_fd, mmap_fd;
+
+/* Discard current buffer used content. */
+static void flush()
+{
+ if (input_offset) {
+ if (output_fd >= 0)
+ write_or_die(output_fd, input_buffer, input_offset);
+ SHA1_Update(&input_ctx, input_buffer, input_offset);
+ memcpy(input_buffer, input_buffer + input_offset, input_len);
+ input_offset = 0;
+ }
+}
/*
* Make sure at least "min" bytes are available in the buffer, and
* return the pointer to the buffer.
*/
-static void * fill(int min)
+static void *fill(int min)
{
if (min <= input_len)
return input_buffer + input_offset;
if (min > sizeof(input_buffer))
die("cannot fill %d bytes", min);
- if (input_offset) {
- SHA1_Update(&input_ctx, input_buffer, input_offset);
- memcpy(input_buffer, input_buffer + input_offset, input_len);
- input_offset = 0;
- }
+ flush();
do {
int ret = xread(input_fd, input_buffer + input_len,
sizeof(input_buffer) - input_len);
@@ -86,13 +134,31 @@ static void use(int bytes)
consumed_bytes += bytes;
}
-static void open_pack_file(void)
+static const char *open_pack_file(const char *pack_name)
{
- input_fd = open(pack_name, O_RDONLY);
- if (input_fd < 0)
- die("cannot open packfile '%s': %s", pack_name,
- strerror(errno));
+ if (from_stdin) {
+ input_fd = 0;
+ if (!pack_name) {
+ static char tmpfile[PATH_MAX];
+ snprintf(tmpfile, sizeof(tmpfile),
+ "%s/pack_XXXXXX", get_object_directory());
+ output_fd = mkstemp(tmpfile);
+ pack_name = xstrdup(tmpfile);
+ } else
+ output_fd = open(pack_name, O_CREAT|O_EXCL|O_RDWR, 0600);
+ if (output_fd < 0)
+ die("unable to create %s: %s\n", pack_name, strerror(errno));
+ mmap_fd = output_fd;
+ } else {
+ input_fd = open(pack_name, O_RDONLY);
+ if (input_fd < 0)
+ die("cannot open packfile '%s': %s",
+ pack_name, strerror(errno));
+ output_fd = -1;
+ mmap_fd = input_fd;
+ }
SHA1_Init(&input_ctx);
+ return pack_name;
}
static void parse_pack_header(void)
@@ -101,14 +167,12 @@ static void parse_pack_header(void)
/* Header consistency check */
if (hdr->hdr_signature != htonl(PACK_SIGNATURE))
- die("packfile '%s' signature mismatch", pack_name);
+ die("pack signature mismatch");
if (!pack_version_ok(hdr->hdr_version))
- die("packfile '%s' version %d unsupported",
- pack_name, ntohl(hdr->hdr_version));
+ die("pack version %d unsupported", ntohl(hdr->hdr_version));
nr_objects = ntohl(hdr->hdr_entries);
use(sizeof(struct pack_header));
- /*fprintf(stderr, "Indexing %d objects\n", nr_objects);*/
}
static void bad_object(unsigned long offset, const char *format,
@@ -122,8 +186,7 @@ static void bad_object(unsigned long offset, const char *format, ...)
va_start(params, format);
vsnprintf(buf, sizeof(buf), format, params);
va_end(params);
- die("packfile '%s': bad object at offset %lu: %s",
- pack_name, offset, buf);
+ die("pack has bad object at offset %lu: %s", offset, buf);
}
static void *unpack_entry_data(unsigned long offset, unsigned long size)
@@ -212,7 +275,7 @@ static void *unpack_raw_entry(struct object_entry *obj, union delta_base *delta_
return unpack_entry_data(obj->offset, obj->size);
}
-static void * get_data_from_pack(struct object_entry *obj)
+static void *get_data_from_pack(struct object_entry *obj)
{
unsigned long from = obj[0].offset + obj[0].hdr_size;
unsigned long len = obj[1].offset - from;
@@ -222,9 +285,9 @@ static void * get_data_from_pack(struct object_entry *obj)
int st;
map = mmap(NULL, len + pg_offset, PROT_READ, MAP_PRIVATE,
- input_fd, from - pg_offset);
+ mmap_fd, from - pg_offset);
if (map == MAP_FAILED)
- die("cannot mmap packfile '%s': %s", pack_name, strerror(errno));
+ die("cannot mmap pack file: %s", strerror(errno));
data = xmalloc(obj->size);
memset(&stream, 0, sizeof(stream));
stream.next_out = data;
@@ -261,8 +324,8 @@ static int find_delta(const union delta_base *base)
return -first-1;
}
-static int find_delta_childs(const union delta_base *base,
- int *first_index, int *last_index)
+static int find_delta_children(const union delta_base *base,
+ int *first_index, int *last_index)
{
int first = find_delta(base);
int last = first;
@@ -304,10 +367,9 @@ static void sha1_object(const void *data, unsigned long size,
SHA1_Final(sha1, &ctx);
}
-static void resolve_delta(struct delta_entry *delta, void *base_data,
+static void resolve_delta(struct object_entry *delta_obj, void *base_data,
unsigned long base_size, enum object_type type)
{
- struct object_entry *obj = delta->obj;
void *delta_data;
unsigned long delta_size;
void *result;
@@ -315,29 +377,34 @@ static void resolve_delta(struct delta_entry *delta, void *base_data,
union delta_base delta_base;
int j, first, last;
- obj->real_type = type;
- delta_data = get_data_from_pack(obj);
- delta_size = obj->size;
+ delta_obj->real_type = type;
+ delta_data = get_data_from_pack(delta_obj);
+ delta_size = delta_obj->size;
result = patch_delta(base_data, base_size, delta_data, delta_size,
&result_size);
free(delta_data);
if (!result)
- bad_object(obj->offset, "failed to apply delta");
- sha1_object(result, result_size, type, obj->sha1);
-
- hashcpy(delta_base.sha1, obj->sha1);
- if (!find_delta_childs(&delta_base, &first, &last)) {
- for (j = first; j <= last; j++)
- if (deltas[j].obj->type == OBJ_REF_DELTA)
- resolve_delta(&deltas[j], result, result_size, type);
+ bad_object(delta_obj->offset, "failed to apply delta");
+ sha1_object(result, result_size, type, delta_obj->sha1);
+ nr_resolved_deltas++;
+
+ hashcpy(delta_base.sha1, delta_obj->sha1);
+ if (!find_delta_children(&delta_base, &first, &last)) {
+ for (j = first; j <= last; j++) {
+ struct object_entry *child = objects + deltas[j].obj_no;
+ if (child->real_type == OBJ_REF_DELTA)
+ resolve_delta(child, result, result_size, type);
+ }
}
memset(&delta_base, 0, sizeof(delta_base));
- delta_base.offset = obj->offset;
- if (!find_delta_childs(&delta_base, &first, &last)) {
- for (j = first; j <= last; j++)
- if (deltas[j].obj->type == OBJ_OFS_DELTA)
- resolve_delta(&deltas[j], result, result_size, type);
+ delta_base.offset = delta_obj->offset;
+ if (!find_delta_children(&delta_base, &first, &last)) {
+ for (j = first; j <= last; j++) {
+ struct object_entry *child = objects + deltas[j].obj_no;
+ if (child->real_type == OBJ_OFS_DELTA)
+ resolve_delta(child, result, result_size, type);
+ }
}
free(result);
@@ -353,7 +420,7 @@ static int compare_delta_entry(const void *a, const void *b)
/* Parse all objects and return the pack content SHA1 hash */
static void parse_pack_objects(unsigned char *sha1)
{
- int i;
+ int i, percent = -1;
struct delta_entry *delta = deltas;
void *data;
struct stat st;
@@ -362,34 +429,43 @@ static void parse_pack_objects(unsigned char *sha1)
* First pass:
* - find locations of all objects;
* - calculate SHA1 of all non-delta objects;
- * - remember base SHA1 for all deltas.
+ * - remember base (SHA1 or offset) for all deltas.
*/
+ if (verbose)
+ fprintf(stderr, "Indexing %d objects.\n", nr_objects);
for (i = 0; i < nr_objects; i++) {
struct object_entry *obj = &objects[i];
data = unpack_raw_entry(obj, &delta->base);
obj->real_type = obj->type;
if (obj->type == OBJ_REF_DELTA || obj->type == OBJ_OFS_DELTA) {
nr_deltas++;
- delta->obj = obj;
+ delta->obj_no = i;
delta++;
} else
sha1_object(data, obj->size, obj->type, obj->sha1);
free(data);
+ if (verbose)
+ percent = display_progress(i+1, nr_objects, percent);
}
objects[i].offset = consumed_bytes;
+ if (verbose)
+ fputc('\n', stderr);
/* Check pack integrity */
- SHA1_Update(&input_ctx, input_buffer, input_offset);
+ flush();
SHA1_Final(sha1, &input_ctx);
if (hashcmp(fill(20), sha1))
- die("packfile '%s' SHA1 mismatch", pack_name);
+ die("pack is corrupted (SHA1 mismatch)");
use(20);
/* If input_fd is a file, we should have reached its end now. */
if (fstat(input_fd, &st))
- die("cannot fstat packfile '%s': %s", pack_name, strerror(errno));
+ die("cannot fstat packfile: %s", strerror(errno));
if (S_ISREG(st.st_mode) && st.st_size != consumed_bytes)
- die("packfile '%s' has junk at the end", pack_name);
+ die("pack has junk at the end");
+
+ if (!nr_deltas)
+ return;
/* Sort deltas by base SHA1/offset for fast searching */
qsort(deltas, nr_deltas, sizeof(struct delta_entry),
@@ -403,6 +479,8 @@ static void parse_pack_objects(unsigned char *sha1)
* recursively checking if the resulting object is used as a base
* for some more deltas.
*/
+ if (verbose)
+ fprintf(stderr, "Resolving %d deltas.\n", nr_deltas);
for (i = 0; i < nr_objects; i++) {
struct object_entry *obj = &objects[i];
union delta_base base;
@@ -411,32 +489,179 @@ static void parse_pack_objects(unsigned char *sha1)
if (obj->type == OBJ_REF_DELTA || obj->type == OBJ_OFS_DELTA)
continue;
hashcpy(base.sha1, obj->sha1);
- ref = !find_delta_childs(&base, &ref_first, &ref_last);
+ ref = !find_delta_children(&base, &ref_first, &ref_last);
memset(&base, 0, sizeof(base));
base.offset = obj->offset;
- ofs = !find_delta_childs(&base, &ofs_first, &ofs_last);
+ ofs = !find_delta_children(&base, &ofs_first, &ofs_last);
if (!ref && !ofs)
continue;
data = get_data_from_pack(obj);
if (ref)
- for (j = ref_first; j <= ref_last; j++)
- if (deltas[j].obj->type == OBJ_REF_DELTA)
- resolve_delta(&deltas[j], data,
+ for (j = ref_first; j <= ref_last; j++) {
+ struct object_entry *child = objects + deltas[j].obj_no;
+ if (child->real_type == OBJ_REF_DELTA)
+ resolve_delta(child, data,
obj->size, obj->type);
+ }
if (ofs)
- for (j = ofs_first; j <= ofs_last; j++)
- if (deltas[j].obj->type == OBJ_OFS_DELTA)
- resolve_delta(&deltas[j], data,
+ for (j = ofs_first; j <= ofs_last; j++) {
+ struct object_entry *child = objects + deltas[j].obj_no;
+ if (child->real_type == OBJ_OFS_DELTA)
+ resolve_delta(child, data,
obj->size, obj->type);
+ }
free(data);
+ if (verbose)
+ percent = display_progress(nr_resolved_deltas,
+ nr_deltas, percent);
+ }
+ if (verbose && nr_resolved_deltas == nr_deltas)
+ fputc('\n', stderr);
+}
+
+static int write_compressed(int fd, void *in, unsigned int size)
+{
+ z_stream stream;
+ unsigned long maxsize;
+ void *out;
+
+ memset(&stream, 0, sizeof(stream));
+ deflateInit(&stream, zlib_compression_level);
+ maxsize = deflateBound(&stream, size);
+ out = xmalloc(maxsize);
+
+ /* Compress it */
+ stream.next_in = in;
+ stream.avail_in = size;
+ stream.next_out = out;
+ stream.avail_out = maxsize;
+ while (deflate(&stream, Z_FINISH) == Z_OK);
+ deflateEnd(&stream);
+
+ size = stream.total_out;
+ write_or_die(fd, out, size);
+ free(out);
+ return size;
+}
+
+static void append_obj_to_pack(void *buf,
+ unsigned long size, enum object_type type)
+{
+ struct object_entry *obj = &objects[nr_objects++];
+ unsigned char header[10];
+ unsigned long s = size;
+ int n = 0;
+ unsigned char c = (type << 4) | (s & 15);
+ s >>= 4;
+ while (s) {
+ header[n++] = c | 0x80;
+ c = s & 0x7f;
+ s >>= 7;
}
+ header[n++] = c;
+ write_or_die(output_fd, header, n);
+ obj[1].offset = obj[0].offset + n;
+ obj[1].offset += write_compressed(output_fd, buf, size);
+ sha1_object(buf, size, type, obj->sha1);
+}
+
+static int delta_pos_compare(const void *_a, const void *_b)
+{
+ struct delta_entry *a = *(struct delta_entry **)_a;
+ struct delta_entry *b = *(struct delta_entry **)_b;
+ return a->obj_no - b->obj_no;
+}
- /* Check for unresolved deltas */
+static void fix_unresolved_deltas(int nr_unresolved)
+{
+ struct delta_entry **sorted_by_pos;
+ int i, n = 0, percent = -1;
+
+ /*
+ * Since many unresolved deltas may well be themselves base objects
+ * for more unresolved deltas, we really want to include the
+ * smallest number of base objects that would cover as much delta
+ * as possible by picking the
+ * trunc deltas first, allowing for other deltas to resolve without
+ * additional base objects. Since most base objects are to be found
+ * before deltas depending on them, a good heuristic is to start
+ * resolving deltas in the same order as their position in the pack.
+ */
+ sorted_by_pos = xmalloc(nr_unresolved * sizeof(*sorted_by_pos));
for (i = 0; i < nr_deltas; i++) {
- if (deltas[i].obj->real_type == OBJ_REF_DELTA ||
- deltas[i].obj->real_type == OBJ_OFS_DELTA)
- die("packfile '%s' has unresolved deltas", pack_name);
+ if (objects[deltas[i].obj_no].real_type != OBJ_REF_DELTA)
+ continue;
+ sorted_by_pos[n++] = &deltas[i];
}
+ qsort(sorted_by_pos, n, sizeof(*sorted_by_pos), delta_pos_compare);
+
+ for (i = 0; i < n; i++) {
+ struct delta_entry *d = sorted_by_pos[i];
+ void *data;
+ unsigned long size;
+ char type[10];
+ enum object_type obj_type;
+ int j, first, last;
+
+ if (objects[d->obj_no].real_type != OBJ_REF_DELTA)
+ continue;
+ data = read_sha1_file(d->base.sha1, type, &size);
+ if (!data)
+ continue;
+ if (!strcmp(type, blob_type)) obj_type = OBJ_BLOB;
+ else if (!strcmp(type, tree_type)) obj_type = OBJ_TREE;
+ else if (!strcmp(type, commit_type)) obj_type = OBJ_COMMIT;
+ else if (!strcmp(type, tag_type)) obj_type = OBJ_TAG;
+ else die("base object %s is of type '%s'",
+ sha1_to_hex(d->base.sha1), type);
+
+ find_delta_children(&d->base, &first, &last);
+ for (j = first; j <= last; j++) {
+ struct object_entry *child = objects + deltas[j].obj_no;
+ if (child->real_type == OBJ_REF_DELTA)
+ resolve_delta(child, data, size, obj_type);
+ }
+
+ append_obj_to_pack(data, size, obj_type);
+ free(data);
+ if (verbose)
+ percent = display_progress(nr_resolved_deltas,
+ nr_deltas, percent);
+ }
+ free(sorted_by_pos);
+ if (verbose)
+ fputc('\n', stderr);
+}
+
+static void readjust_pack_header_and_sha1(unsigned char *sha1)
+{
+ struct pack_header hdr;
+ SHA_CTX ctx;
+ int size;
+
+ /* Rewrite pack header with updated object number */
+ if (lseek(output_fd, 0, SEEK_SET) != 0)
+ die("cannot seek back: %s", strerror(errno));
+ if (xread(output_fd, &hdr, sizeof(hdr)) != sizeof(hdr))
+ die("cannot read pack header back: %s", strerror(errno));
+ hdr.hdr_entries = htonl(nr_objects);
+ if (lseek(output_fd, 0, SEEK_SET) != 0)
+ die("cannot seek back: %s", strerror(errno));
+ write_or_die(output_fd, &hdr, sizeof(hdr));
+ if (lseek(output_fd, 0, SEEK_SET) != 0)
+ die("cannot seek back: %s", strerror(errno));
+
+ /* Recompute and store the new pack's SHA1 */
+ SHA1_Init(&ctx);
+ do {
+ unsigned char *buf[4096];
+ size = xread(output_fd, buf, sizeof(buf));
+ if (size < 0)
+ die("cannot read pack data back: %s", strerror(errno));
+ SHA1_Update(&ctx, buf, size);
+ } while (size > 0);
+ SHA1_Final(sha1, &ctx);
+ write_or_die(output_fd, sha1, 20);
}
static int sha1_compare(const void *_a, const void *_b)
@@ -450,12 +675,12 @@ static int sha1_compare(const void *_a, const void *_b)
* On entry *sha1 contains the pack content SHA1 hash, on exit it is
* the SHA1 hash of sorted object names.
*/
-static void write_index_file(const char *index_name, unsigned char *sha1)
+static const char *write_index_file(const char *index_name, unsigned char *sha1)
{
struct sha1file *f;
struct object_entry **sorted_by_sha, **list, **last;
unsigned int array[256];
- int i;
+ int i, fd;
SHA_CTX ctx;
if (nr_objects) {
@@ -472,8 +697,19 @@ static void write_index_file(const char *index_name, unsigned char *sha1)
else
sorted_by_sha = list = last = NULL;
- unlink(index_name);
- f = sha1create("%s", index_name);
+ if (!index_name) {
+ static char tmpfile[PATH_MAX];
+ snprintf(tmpfile, sizeof(tmpfile),
+ "%s/index_XXXXXX", get_object_directory());
+ fd = mkstemp(tmpfile);
+ index_name = xstrdup(tmpfile);
+ } else {
+ unlink(index_name);
+ fd = open(index_name, O_CREAT|O_EXCL|O_WRONLY, 0600);
+ }
+ if (fd < 0)
+ die("unable to create %s: %s", index_name, strerror(errno));
+ f = sha1fd(fd, index_name);
/*
* Write the first-level table (the list is sorted,
@@ -513,12 +749,64 @@ static void write_index_file(const char *index_name, unsigned char *sha1)
sha1close(f, NULL, 1);
free(sorted_by_sha);
SHA1_Final(sha1, &ctx);
+ return index_name;
+}
+
+static void final(const char *final_pack_name, const char *curr_pack_name,
+ const char *final_index_name, const char *curr_index_name,
+ unsigned char *sha1)
+{
+ char name[PATH_MAX];
+ int err;
+
+ if (!from_stdin) {
+ close(input_fd);
+ } else {
+ err = close(output_fd);
+ if (err)
+ die("error while closing pack file: %s", strerror(errno));
+ chmod(curr_pack_name, 0444);
+
+ /*
+ * Let's just mimic git-unpack-objects here and write
+ * the last part of the buffer to stdout.
+ */
+ while (input_len) {
+ err = xwrite(1, input_buffer + input_offset, input_len);
+ if (err <= 0)
+ break;
+ input_len -= err;
+ input_offset += err;
+ }
+ }
+
+ if (final_pack_name != curr_pack_name) {
+ if (!final_pack_name) {
+ snprintf(name, sizeof(name), "%s/pack/pack-%s.pack",
+ get_object_directory(), sha1_to_hex(sha1));
+ final_pack_name = name;
+ }
+ if (move_temp_to_file(curr_pack_name, final_pack_name))
+ die("cannot store pack file");
+ }
+
+ chmod(curr_index_name, 0444);
+ if (final_index_name != curr_index_name) {
+ if (!final_index_name) {
+ snprintf(name, sizeof(name), "%s/pack/pack-%s.idx",
+ get_object_directory(), sha1_to_hex(sha1));
+ final_index_name = name;
+ }
+ if (move_temp_to_file(curr_index_name, final_index_name))
+ die("cannot store index file");
+ }
}
int main(int argc, char **argv)
{
- int i;
- char *index_name = NULL;
+ int i, fix_thin_pack = 0;
+ const char *curr_pack, *pack_name = NULL;
+ const char *curr_index, *index_name = NULL;
char *index_name_buf = NULL;
unsigned char sha1[20];
@@ -526,7 +814,13 @@ int main(int argc, char **argv)
const char *arg = argv[i];
if (*arg == '-') {
- if (!strcmp(arg, "-o")) {
+ if (!strcmp(arg, "--stdin")) {
+ from_stdin = 1;
+ } else if (!strcmp(arg, "--fix-thin")) {
+ fix_thin_pack = 1;
+ } else if (!strcmp(arg, "-v")) {
+ verbose = 1;
+ } else if (!strcmp(arg, "-o")) {
if (index_name || (i+1) >= argc)
usage(index_pack_usage);
index_name = argv[++i];
@@ -540,9 +834,11 @@ int main(int argc, char **argv)
pack_name = arg;
}
- if (!pack_name)
+ if (!pack_name && !from_stdin)
usage(index_pack_usage);
- if (!index_name) {
+ if (fix_thin_pack && !from_stdin)
+ die("--fix-thin cannot be used without --stdin");
+ if (!index_name && pack_name) {
int len = strlen(pack_name);
if (!has_extension(pack_name, ".pack"))
die("packfile name '%s' does not end with '.pack'",
@@ -553,17 +849,43 @@ int main(int argc, char **argv)
index_name = index_name_buf;
}
- open_pack_file();
+ curr_pack = open_pack_file(pack_name);
parse_pack_header();
- objects = xcalloc(nr_objects + 1, sizeof(struct object_entry));
- deltas = xcalloc(nr_objects, sizeof(struct delta_entry));
+ objects = xmalloc((nr_objects + 1) * sizeof(struct object_entry));
+ deltas = xmalloc(nr_objects * sizeof(struct delta_entry));
+ if (verbose)
+ setup_progress_signal();
parse_pack_objects(sha1);
+ if (nr_deltas != nr_resolved_deltas) {
+ if (fix_thin_pack) {
+ int nr_unresolved = nr_deltas - nr_resolved_deltas;
+ int nr_objects_initial = nr_objects;
+ if (nr_unresolved <= 0)
+ die("confusion beyond insanity");
+ objects = xrealloc(objects,
+ (nr_objects + nr_unresolved + 1)
+ * sizeof(*objects));
+ fix_unresolved_deltas(nr_unresolved);
+ if (verbose)
+ fprintf(stderr, "%d objects were added to complete this thin pack.\n",
+ nr_objects - nr_objects_initial);
+ readjust_pack_header_and_sha1(sha1);
+ }
+ if (nr_deltas != nr_resolved_deltas)
+ die("pack has %d unresolved deltas",
+ nr_deltas - nr_resolved_deltas);
+ } else {
+ /* Flush remaining pack final 20-byte SHA1. */
+ flush();
+ }
free(deltas);
- write_index_file(index_name, sha1);
+ curr_index = write_index_file(index_name, sha1);
+ final(pack_name, curr_pack, index_name, curr_index, sha1);
free(objects);
free(index_name_buf);
- printf("%s\n", sha1_to_hex(sha1));
+ if (!from_stdin)
+ printf("%s\n", sha1_to_hex(sha1));
return 0;
}