2005-04-24 04:04:40 +02:00
|
|
|
#include "cache.h"
|
2005-10-05 23:49:54 +02:00
|
|
|
#include "refs.h"
|
2005-06-29 20:30:24 +02:00
|
|
|
#include "tag.h"
|
2005-04-24 04:04:40 +02:00
|
|
|
#include "commit.h"
|
2005-06-25 07:56:58 +02:00
|
|
|
#include "tree.h"
|
|
|
|
|
#include "blob.h"
|
2006-03-30 08:55:43 +02:00
|
|
|
#include "tree-walk.h"
|
2006-04-11 03:14:54 +02:00
|
|
|
#include "diff.h"
|
2006-02-26 01:19:46 +01:00
|
|
|
#include "revision.h"
|
2006-05-18 23:19:20 +02:00
|
|
|
#include "builtin.h"
|
2006-02-26 01:19:46 +01:00
|
|
|
|
2006-04-17 03:12:49 +02:00
|
|
|
/* bits #0-15 in revision.h */
|
2005-04-24 04:04:40 +02:00
|
|
|
|
2006-04-17 03:12:49 +02:00
|
|
|
#define COUNTED (1u<<16)
|
2005-05-31 03:46:32 +02:00
|
|
|
|
2005-05-26 03:29:09 +02:00
|
|
|
static const char rev_list_usage[] =
|
2005-10-30 10:03:45 +01:00
|
|
|
"git-rev-list [OPTION] <commit-id>... [ -- paths... ]\n"
|
|
|
|
|
" limiting output:\n"
|
|
|
|
|
" --max-count=nr\n"
|
|
|
|
|
" --max-age=epoch\n"
|
|
|
|
|
" --min-age=epoch\n"
|
|
|
|
|
" --sparse\n"
|
|
|
|
|
" --no-merges\n"
|
2006-01-27 10:39:24 +01:00
|
|
|
" --remove-empty\n"
|
2005-10-30 10:03:45 +01:00
|
|
|
" --all\n"
|
|
|
|
|
" ordering output:\n"
|
|
|
|
|
" --topo-order\n"
|
2006-02-16 07:05:33 +01:00
|
|
|
" --date-order\n"
|
2005-10-30 10:03:45 +01:00
|
|
|
" formatting output:\n"
|
|
|
|
|
" --parents\n"
|
2006-02-19 12:32:31 +01:00
|
|
|
" --objects | --objects-edge\n"
|
2005-10-30 10:03:45 +01:00
|
|
|
" --unpacked\n"
|
|
|
|
|
" --header | --pretty\n"
|
2006-02-10 20:56:42 +01:00
|
|
|
" --abbrev=nr | --no-abbrev\n"
|
2006-04-07 06:32:36 +02:00
|
|
|
" --abbrev-commit\n"
|
2005-10-30 10:03:45 +01:00
|
|
|
" special purpose:\n"
|
|
|
|
|
" --bisect"
|
|
|
|
|
;
|
2005-05-26 03:29:09 +02:00
|
|
|
|
2006-05-18 23:19:20 +02:00
|
|
|
static struct rev_info revs;
|
2006-02-26 01:19:46 +01:00
|
|
|
|
2005-06-18 07:54:50 +02:00
|
|
|
static int bisect_list = 0;
|
2006-03-22 09:22:00 +01:00
|
|
|
static int show_timestamp = 0;
|
2005-06-02 18:19:53 +02:00
|
|
|
static int hdr_termination = 0;
|
Log message printout cleanups
On Sun, 16 Apr 2006, Junio C Hamano wrote:
>
> In the mid-term, I am hoping we can drop the generate_header()
> callchain _and_ the custom code that formats commit log in-core,
> found in cmd_log_wc().
Ok, this was nastier than expected, just because the dependencies between
the different log-printing stuff were absolutely _everywhere_, but here's
a patch that does exactly that.
The patch is not very easy to read, and the "--patch-with-stat" thing is
still broken (it does not call the "show_log()" thing properly for
merges). That's not a new bug. In the new world order it _should_ do
something like
if (rev->logopt)
show_log(rev, rev->logopt, "---\n");
but it doesn't. I haven't looked at the --with-stat logic, so I left it
alone.
That said, this patch removes more lines than it adds, and in particular,
the "cmd_log_wc()" loop is now a very clean:
while ((commit = get_revision(rev)) != NULL) {
log_tree_commit(rev, commit);
free(commit->buffer);
commit->buffer = NULL;
}
so it doesn't get much prettier than this. All the complexity is entirely
hidden in log-tree.c, and any code that needs to flush the log literally
just needs to do the "if (rev->logopt) show_log(...)" incantation.
I had to make the combined_diff() logic take a "struct rev_info" instead
of just a "struct diff_options", but that part is pretty clean.
This does change "git whatchanged" from using "diff-tree" as the commit
descriptor to "commit", and I changed one of the tests to reflect that new
reality. Otherwise everything still passes, and my other tests look fine
too.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-04-17 20:59:32 +02:00
|
|
|
static const char *header_prefix;
|
2006-02-23 07:10:24 +01:00
|
|
|
|
2005-06-02 18:19:53 +02:00
|
|
|
static void show_commit(struct commit *commit)
|
|
|
|
|
{
|
2006-03-22 09:22:00 +01:00
|
|
|
if (show_timestamp)
|
|
|
|
|
printf("%lu ", commit->date);
|
Log message printout cleanups
On Sun, 16 Apr 2006, Junio C Hamano wrote:
>
> In the mid-term, I am hoping we can drop the generate_header()
> callchain _and_ the custom code that formats commit log in-core,
> found in cmd_log_wc().
Ok, this was nastier than expected, just because the dependencies between
the different log-printing stuff were absolutely _everywhere_, but here's
a patch that does exactly that.
The patch is not very easy to read, and the "--patch-with-stat" thing is
still broken (it does not call the "show_log()" thing properly for
merges). That's not a new bug. In the new world order it _should_ do
something like
if (rev->logopt)
show_log(rev, rev->logopt, "---\n");
but it doesn't. I haven't looked at the --with-stat logic, so I left it
alone.
That said, this patch removes more lines than it adds, and in particular,
the "cmd_log_wc()" loop is now a very clean:
while ((commit = get_revision(rev)) != NULL) {
log_tree_commit(rev, commit);
free(commit->buffer);
commit->buffer = NULL;
}
so it doesn't get much prettier than this. All the complexity is entirely
hidden in log-tree.c, and any code that needs to flush the log literally
just needs to do the "if (rev->logopt) show_log(...)" incantation.
I had to make the combined_diff() logic take a "struct rev_info" instead
of just a "struct diff_options", but that part is pretty clean.
This does change "git whatchanged" from using "diff-tree" as the commit
descriptor to "commit", and I changed one of the tests to reflect that new
reality. Otherwise everything still passes, and my other tests look fine
too.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-04-17 20:59:32 +02:00
|
|
|
if (header_prefix)
|
|
|
|
|
fputs(header_prefix, stdout);
|
2006-03-28 09:58:34 +02:00
|
|
|
if (commit->object.flags & BOUNDARY)
|
|
|
|
|
putchar('-');
|
2006-04-16 08:48:27 +02:00
|
|
|
if (revs.abbrev_commit && revs.abbrev)
|
|
|
|
|
fputs(find_unique_abbrev(commit->object.sha1, revs.abbrev),
|
|
|
|
|
stdout);
|
2006-04-07 06:32:36 +02:00
|
|
|
else
|
|
|
|
|
fputs(sha1_to_hex(commit->object.sha1), stdout);
|
2006-03-31 02:52:42 +02:00
|
|
|
if (revs.parents) {
|
2005-06-02 18:19:53 +02:00
|
|
|
struct commit_list *parents = commit->parents;
|
|
|
|
|
while (parents) {
|
2006-01-30 00:24:42 +01:00
|
|
|
struct object *o = &(parents->item->object);
|
2005-06-02 18:19:53 +02:00
|
|
|
parents = parents->next;
|
2006-01-30 00:24:42 +01:00
|
|
|
if (o->flags & TMP_MARK)
|
|
|
|
|
continue;
|
|
|
|
|
printf(" %s", sha1_to_hex(o->sha1));
|
|
|
|
|
o->flags |= TMP_MARK;
|
2005-06-02 18:19:53 +02:00
|
|
|
}
|
2006-01-30 00:24:42 +01:00
|
|
|
/* TMP_MARK is a general purpose flag that can
|
|
|
|
|
* be used locally, but the user should clean
|
|
|
|
|
* things up after it is done with them.
|
|
|
|
|
*/
|
|
|
|
|
for (parents = commit->parents;
|
|
|
|
|
parents;
|
|
|
|
|
parents = parents->next)
|
|
|
|
|
parents->item->object.flags &= ~TMP_MARK;
|
2005-06-02 18:19:53 +02:00
|
|
|
}
|
2006-04-16 08:48:27 +02:00
|
|
|
if (revs.commit_format == CMIT_FMT_ONELINE)
|
2005-08-09 07:15:40 +02:00
|
|
|
putchar(' ');
|
|
|
|
|
else
|
|
|
|
|
putchar('\n');
|
|
|
|
|
|
2006-04-16 08:48:27 +02:00
|
|
|
if (revs.verbose_header) {
|
2005-06-05 18:02:03 +02:00
|
|
|
static char pretty_header[16384];
|
2006-04-16 08:48:27 +02:00
|
|
|
pretty_print_commit(revs.commit_format, commit, ~0,
|
|
|
|
|
pretty_header, sizeof(pretty_header),
|
2006-05-20 15:40:29 +02:00
|
|
|
revs.abbrev, NULL, NULL);
|
2005-06-05 18:02:03 +02:00
|
|
|
printf("%s%c", pretty_header, hdr_termination);
|
2005-07-05 01:36:48 +02:00
|
|
|
}
|
|
|
|
|
fflush(stdout);
|
2006-06-18 03:47:58 +02:00
|
|
|
if (commit->parents) {
|
|
|
|
|
free_commit_list(commit->parents);
|
|
|
|
|
commit->parents = NULL;
|
|
|
|
|
}
|
|
|
|
|
if (commit->buffer) {
|
|
|
|
|
free(commit->buffer);
|
|
|
|
|
commit->buffer = NULL;
|
|
|
|
|
}
|
2005-06-06 17:39:40 +02:00
|
|
|
}
|
|
|
|
|
|
Add "named object array" concept
We've had this notion of a "object_list" for a long time, which eventually
grew a "name" member because some users (notably git-rev-list) wanted to
name each object as it is generated.
That object_list is great for some things, but it isn't all that wonderful
for others, and the "name" member is generally not used by everybody.
This patch splits the users of the object_list array up into two: the
traditional list users, who want the list-like format, and who don't
actually use or want the name. And another class of users that really used
the list as an extensible array, and generally wanted to name the objects.
The patch is fairly straightforward, but it's also biggish. Most of it
really just cleans things up: switching the revision parsing and listing
over to the array makes things like the builtin-diff usage much simpler
(we now see exactly how many members the array has, and we don't get the
objects reversed from the order they were on the command line).
One of the main reasons for doing this at all is that the malloc overhead
of the simple object list was actually pretty high, and the array is just
a lot denser. So this patch brings down memory usage by git-rev-list by
just under 3% (on top of all the other memory use optimizations) on the
mozilla archive.
It does add more lines than it removes, and more importantly, it adds a
whole new infrastructure for maintaining lists of objects, but on the
other hand, the new dynamic array code is pretty obvious. The change to
builtin-diff-tree.c shows a fairly good example of why an array interface
is sometimes more natural, and just much simpler for everybody.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-06-20 02:42:35 +02:00
|
|
|
static void process_blob(struct blob *blob,
|
|
|
|
|
struct object_array *p,
|
|
|
|
|
struct name_path *path,
|
|
|
|
|
const char *name)
|
2005-06-25 07:56:58 +02:00
|
|
|
{
|
|
|
|
|
struct object *obj = &blob->object;
|
|
|
|
|
|
2006-02-26 01:19:46 +01:00
|
|
|
if (!revs.blob_objects)
|
Add "named object array" concept
We've had this notion of a "object_list" for a long time, which eventually
grew a "name" member because some users (notably git-rev-list) wanted to
name each object as it is generated.
That object_list is great for some things, but it isn't all that wonderful
for others, and the "name" member is generally not used by everybody.
This patch splits the users of the object_list array up into two: the
traditional list users, who want the list-like format, and who don't
actually use or want the name. And another class of users that really used
the list as an extensible array, and generally wanted to name the objects.
The patch is fairly straightforward, but it's also biggish. Most of it
really just cleans things up: switching the revision parsing and listing
over to the array makes things like the builtin-diff usage much simpler
(we now see exactly how many members the array has, and we don't get the
objects reversed from the order they were on the command line).
One of the main reasons for doing this at all is that the malloc overhead
of the simple object list was actually pretty high, and the array is just
a lot denser. So this patch brings down memory usage by git-rev-list by
just under 3% (on top of all the other memory use optimizations) on the
mozilla archive.
It does add more lines than it removes, and more importantly, it adds a
whole new infrastructure for maintaining lists of objects, but on the
other hand, the new dynamic array code is pretty obvious. The change to
builtin-diff-tree.c shows a fairly good example of why an array interface
is sometimes more natural, and just much simpler for everybody.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-06-20 02:42:35 +02:00
|
|
|
return;
|
2005-06-25 07:56:58 +02:00
|
|
|
if (obj->flags & (UNINTERESTING | SEEN))
|
Add "named object array" concept
We've had this notion of a "object_list" for a long time, which eventually
grew a "name" member because some users (notably git-rev-list) wanted to
name each object as it is generated.
That object_list is great for some things, but it isn't all that wonderful
for others, and the "name" member is generally not used by everybody.
This patch splits the users of the object_list array up into two: the
traditional list users, who want the list-like format, and who don't
actually use or want the name. And another class of users that really used
the list as an extensible array, and generally wanted to name the objects.
The patch is fairly straightforward, but it's also biggish. Most of it
really just cleans things up: switching the revision parsing and listing
over to the array makes things like the builtin-diff usage much simpler
(we now see exactly how many members the array has, and we don't get the
objects reversed from the order they were on the command line).
One of the main reasons for doing this at all is that the malloc overhead
of the simple object list was actually pretty high, and the array is just
a lot denser. So this patch brings down memory usage by git-rev-list by
just under 3% (on top of all the other memory use optimizations) on the
mozilla archive.
It does add more lines than it removes, and more importantly, it adds a
whole new infrastructure for maintaining lists of objects, but on the
other hand, the new dynamic array code is pretty obvious. The change to
builtin-diff-tree.c shows a fairly good example of why an array interface
is sometimes more natural, and just much simpler for everybody.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-06-20 02:42:35 +02:00
|
|
|
return;
|
2005-06-25 07:56:58 +02:00
|
|
|
obj->flags |= SEEN;
|
2006-05-28 20:37:23 +02:00
|
|
|
name = strdup(name);
|
Add "named object array" concept
We've had this notion of a "object_list" for a long time, which eventually
grew a "name" member because some users (notably git-rev-list) wanted to
name each object as it is generated.
That object_list is great for some things, but it isn't all that wonderful
for others, and the "name" member is generally not used by everybody.
This patch splits the users of the object_list array up into two: the
traditional list users, who want the list-like format, and who don't
actually use or want the name. And another class of users that really used
the list as an extensible array, and generally wanted to name the objects.
The patch is fairly straightforward, but it's also biggish. Most of it
really just cleans things up: switching the revision parsing and listing
over to the array makes things like the builtin-diff usage much simpler
(we now see exactly how many members the array has, and we don't get the
objects reversed from the order they were on the command line).
One of the main reasons for doing this at all is that the malloc overhead
of the simple object list was actually pretty high, and the array is just
a lot denser. So this patch brings down memory usage by git-rev-list by
just under 3% (on top of all the other memory use optimizations) on the
mozilla archive.
It does add more lines than it removes, and more importantly, it adds a
whole new infrastructure for maintaining lists of objects, but on the
other hand, the new dynamic array code is pretty obvious. The change to
builtin-diff-tree.c shows a fairly good example of why an array interface
is sometimes more natural, and just much simpler for everybody.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-06-20 02:42:35 +02:00
|
|
|
add_object(obj, p, path, name);
|
2005-06-25 07:56:58 +02:00
|
|
|
}
|
|
|
|
|
|
Add "named object array" concept
We've had this notion of a "object_list" for a long time, which eventually
grew a "name" member because some users (notably git-rev-list) wanted to
name each object as it is generated.
That object_list is great for some things, but it isn't all that wonderful
for others, and the "name" member is generally not used by everybody.
This patch splits the users of the object_list array up into two: the
traditional list users, who want the list-like format, and who don't
actually use or want the name. And another class of users that really used
the list as an extensible array, and generally wanted to name the objects.
The patch is fairly straightforward, but it's also biggish. Most of it
really just cleans things up: switching the revision parsing and listing
over to the array makes things like the builtin-diff usage much simpler
(we now see exactly how many members the array has, and we don't get the
objects reversed from the order they were on the command line).
One of the main reasons for doing this at all is that the malloc overhead
of the simple object list was actually pretty high, and the array is just
a lot denser. So this patch brings down memory usage by git-rev-list by
just under 3% (on top of all the other memory use optimizations) on the
mozilla archive.
It does add more lines than it removes, and more importantly, it adds a
whole new infrastructure for maintaining lists of objects, but on the
other hand, the new dynamic array code is pretty obvious. The change to
builtin-diff-tree.c shows a fairly good example of why an array interface
is sometimes more natural, and just much simpler for everybody.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-06-20 02:42:35 +02:00
|
|
|
static void process_tree(struct tree *tree,
|
|
|
|
|
struct object_array *p,
|
|
|
|
|
struct name_path *path,
|
|
|
|
|
const char *name)
|
2005-06-25 07:56:58 +02:00
|
|
|
{
|
|
|
|
|
struct object *obj = &tree->object;
|
2006-05-29 21:18:33 +02:00
|
|
|
struct tree_desc desc;
|
tree_entry(): new tree-walking helper function
This adds a "tree_entry()" function that combines the common operation of
doing a "tree_entry_extract()" + "update_tree_entry()".
It also has a simplified calling convention, designed for simple loops
that traverse over a whole tree: the arguments are pointers to the tree
descriptor and a name_entry structure to fill in, and it returns a boolean
"true" if there was an entry left to be gotten in the tree.
This allows tree traversal with
struct tree_desc desc;
struct name_entry entry;
desc.buf = tree->buffer;
desc.size = tree->size;
while (tree_entry(&desc, &entry) {
... use "entry.{path, sha1, mode, pathlen}" ...
}
which is not only shorter than writing it out in full, it's hopefully less
error prone too.
[ It's actually a tad faster too - we don't need to recalculate the entry
pathlength in both extract and update, but need to do it only once.
Also, some callers can avoid doing a "strlen()" on the result, since
it's returned as part of the name_entry structure.
However, by now we're talking just 1% speedup on "git-rev-list --objects
--all", and we're definitely at the point where tree walking is no
longer the issue any more. ]
NOTE! Not everybody wants to use this new helper function, since some of
the tree walkers very much on purpose do the descriptor update separately
from the entry extraction. So the "extract + update" sequence still
remains as the core sequence, this is just a simplified interface.
We should probably add a silly two-line inline helper function for
initializing the descriptor from the "struct tree" too, just to cut down
on the noise from that common "desc" initializer.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-05-30 18:45:45 +02:00
|
|
|
struct name_entry entry;
|
2006-02-23 07:10:24 +01:00
|
|
|
struct name_path me;
|
2005-06-25 07:56:58 +02:00
|
|
|
|
2006-02-26 01:19:46 +01:00
|
|
|
if (!revs.tree_objects)
|
Add "named object array" concept
We've had this notion of a "object_list" for a long time, which eventually
grew a "name" member because some users (notably git-rev-list) wanted to
name each object as it is generated.
That object_list is great for some things, but it isn't all that wonderful
for others, and the "name" member is generally not used by everybody.
This patch splits the users of the object_list array up into two: the
traditional list users, who want the list-like format, and who don't
actually use or want the name. And another class of users that really used
the list as an extensible array, and generally wanted to name the objects.
The patch is fairly straightforward, but it's also biggish. Most of it
really just cleans things up: switching the revision parsing and listing
over to the array makes things like the builtin-diff usage much simpler
(we now see exactly how many members the array has, and we don't get the
objects reversed from the order they were on the command line).
One of the main reasons for doing this at all is that the malloc overhead
of the simple object list was actually pretty high, and the array is just
a lot denser. So this patch brings down memory usage by git-rev-list by
just under 3% (on top of all the other memory use optimizations) on the
mozilla archive.
It does add more lines than it removes, and more importantly, it adds a
whole new infrastructure for maintaining lists of objects, but on the
other hand, the new dynamic array code is pretty obvious. The change to
builtin-diff-tree.c shows a fairly good example of why an array interface
is sometimes more natural, and just much simpler for everybody.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-06-20 02:42:35 +02:00
|
|
|
return;
|
2005-06-25 07:56:58 +02:00
|
|
|
if (obj->flags & (UNINTERESTING | SEEN))
|
Add "named object array" concept
We've had this notion of a "object_list" for a long time, which eventually
grew a "name" member because some users (notably git-rev-list) wanted to
name each object as it is generated.
That object_list is great for some things, but it isn't all that wonderful
for others, and the "name" member is generally not used by everybody.
This patch splits the users of the object_list array up into two: the
traditional list users, who want the list-like format, and who don't
actually use or want the name. And another class of users that really used
the list as an extensible array, and generally wanted to name the objects.
The patch is fairly straightforward, but it's also biggish. Most of it
really just cleans things up: switching the revision parsing and listing
over to the array makes things like the builtin-diff usage much simpler
(we now see exactly how many members the array has, and we don't get the
objects reversed from the order they were on the command line).
One of the main reasons for doing this at all is that the malloc overhead
of the simple object list was actually pretty high, and the array is just
a lot denser. So this patch brings down memory usage by git-rev-list by
just under 3% (on top of all the other memory use optimizations) on the
mozilla archive.
It does add more lines than it removes, and more importantly, it adds a
whole new infrastructure for maintaining lists of objects, but on the
other hand, the new dynamic array code is pretty obvious. The change to
builtin-diff-tree.c shows a fairly good example of why an array interface
is sometimes more natural, and just much simpler for everybody.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-06-20 02:42:35 +02:00
|
|
|
return;
|
2005-06-25 07:56:58 +02:00
|
|
|
if (parse_tree(tree) < 0)
|
|
|
|
|
die("bad tree object %s", sha1_to_hex(obj->sha1));
|
|
|
|
|
obj->flags |= SEEN;
|
2006-05-28 20:37:23 +02:00
|
|
|
name = strdup(name);
|
Add "named object array" concept
We've had this notion of a "object_list" for a long time, which eventually
grew a "name" member because some users (notably git-rev-list) wanted to
name each object as it is generated.
That object_list is great for some things, but it isn't all that wonderful
for others, and the "name" member is generally not used by everybody.
This patch splits the users of the object_list array up into two: the
traditional list users, who want the list-like format, and who don't
actually use or want the name. And another class of users that really used
the list as an extensible array, and generally wanted to name the objects.
The patch is fairly straightforward, but it's also biggish. Most of it
really just cleans things up: switching the revision parsing and listing
over to the array makes things like the builtin-diff usage much simpler
(we now see exactly how many members the array has, and we don't get the
objects reversed from the order they were on the command line).
One of the main reasons for doing this at all is that the malloc overhead
of the simple object list was actually pretty high, and the array is just
a lot denser. So this patch brings down memory usage by git-rev-list by
just under 3% (on top of all the other memory use optimizations) on the
mozilla archive.
It does add more lines than it removes, and more importantly, it adds a
whole new infrastructure for maintaining lists of objects, but on the
other hand, the new dynamic array code is pretty obvious. The change to
builtin-diff-tree.c shows a fairly good example of why an array interface
is sometimes more natural, and just much simpler for everybody.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-06-20 02:42:35 +02:00
|
|
|
add_object(obj, p, path, name);
|
2006-02-23 07:10:24 +01:00
|
|
|
me.up = path;
|
|
|
|
|
me.elem = name;
|
|
|
|
|
me.elem_len = strlen(name);
|
2006-05-29 21:18:33 +02:00
|
|
|
|
|
|
|
|
desc.buf = tree->buffer;
|
|
|
|
|
desc.size = tree->size;
|
|
|
|
|
|
tree_entry(): new tree-walking helper function
This adds a "tree_entry()" function that combines the common operation of
doing a "tree_entry_extract()" + "update_tree_entry()".
It also has a simplified calling convention, designed for simple loops
that traverse over a whole tree: the arguments are pointers to the tree
descriptor and a name_entry structure to fill in, and it returns a boolean
"true" if there was an entry left to be gotten in the tree.
This allows tree traversal with
struct tree_desc desc;
struct name_entry entry;
desc.buf = tree->buffer;
desc.size = tree->size;
while (tree_entry(&desc, &entry) {
... use "entry.{path, sha1, mode, pathlen}" ...
}
which is not only shorter than writing it out in full, it's hopefully less
error prone too.
[ It's actually a tad faster too - we don't need to recalculate the entry
pathlength in both extract and update, but need to do it only once.
Also, some callers can avoid doing a "strlen()" on the result, since
it's returned as part of the name_entry structure.
However, by now we're talking just 1% speedup on "git-rev-list --objects
--all", and we're definitely at the point where tree walking is no
longer the issue any more. ]
NOTE! Not everybody wants to use this new helper function, since some of
the tree walkers very much on purpose do the descriptor update separately
from the entry extraction. So the "extract + update" sequence still
remains as the core sequence, this is just a simplified interface.
We should probably add a silly two-line inline helper function for
initializing the descriptor from the "struct tree" too, just to cut down
on the noise from that common "desc" initializer.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-05-30 18:45:45 +02:00
|
|
|
while (tree_entry(&desc, &entry)) {
|
|
|
|
|
if (S_ISDIR(entry.mode))
|
Add "named object array" concept
We've had this notion of a "object_list" for a long time, which eventually
grew a "name" member because some users (notably git-rev-list) wanted to
name each object as it is generated.
That object_list is great for some things, but it isn't all that wonderful
for others, and the "name" member is generally not used by everybody.
This patch splits the users of the object_list array up into two: the
traditional list users, who want the list-like format, and who don't
actually use or want the name. And another class of users that really used
the list as an extensible array, and generally wanted to name the objects.
The patch is fairly straightforward, but it's also biggish. Most of it
really just cleans things up: switching the revision parsing and listing
over to the array makes things like the builtin-diff usage much simpler
(we now see exactly how many members the array has, and we don't get the
objects reversed from the order they were on the command line).
One of the main reasons for doing this at all is that the malloc overhead
of the simple object list was actually pretty high, and the array is just
a lot denser. So this patch brings down memory usage by git-rev-list by
just under 3% (on top of all the other memory use optimizations) on the
mozilla archive.
It does add more lines than it removes, and more importantly, it adds a
whole new infrastructure for maintaining lists of objects, but on the
other hand, the new dynamic array code is pretty obvious. The change to
builtin-diff-tree.c shows a fairly good example of why an array interface
is sometimes more natural, and just much simpler for everybody.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-06-20 02:42:35 +02:00
|
|
|
process_tree(lookup_tree(entry.sha1), p, &me, entry.path);
|
2005-06-25 07:56:58 +02:00
|
|
|
else
|
Add "named object array" concept
We've had this notion of a "object_list" for a long time, which eventually
grew a "name" member because some users (notably git-rev-list) wanted to
name each object as it is generated.
That object_list is great for some things, but it isn't all that wonderful
for others, and the "name" member is generally not used by everybody.
This patch splits the users of the object_list array up into two: the
traditional list users, who want the list-like format, and who don't
actually use or want the name. And another class of users that really used
the list as an extensible array, and generally wanted to name the objects.
The patch is fairly straightforward, but it's also biggish. Most of it
really just cleans things up: switching the revision parsing and listing
over to the array makes things like the builtin-diff usage much simpler
(we now see exactly how many members the array has, and we don't get the
objects reversed from the order they were on the command line).
One of the main reasons for doing this at all is that the malloc overhead
of the simple object list was actually pretty high, and the array is just
a lot denser. So this patch brings down memory usage by git-rev-list by
just under 3% (on top of all the other memory use optimizations) on the
mozilla archive.
It does add more lines than it removes, and more importantly, it adds a
whole new infrastructure for maintaining lists of objects, but on the
other hand, the new dynamic array code is pretty obvious. The change to
builtin-diff-tree.c shows a fairly good example of why an array interface
is sometimes more natural, and just much simpler for everybody.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-06-20 02:42:35 +02:00
|
|
|
process_blob(lookup_blob(entry.sha1), p, &me, entry.path);
|
2005-06-25 07:56:58 +02:00
|
|
|
}
|
2006-05-29 21:16:12 +02:00
|
|
|
free(tree->buffer);
|
|
|
|
|
tree->buffer = NULL;
|
2005-06-25 07:56:58 +02:00
|
|
|
}
|
|
|
|
|
|
2006-02-28 20:24:00 +01:00
|
|
|
static void show_commit_list(struct rev_info *revs)
|
2005-06-02 18:19:53 +02:00
|
|
|
{
|
Add "named object array" concept
We've had this notion of a "object_list" for a long time, which eventually
grew a "name" member because some users (notably git-rev-list) wanted to
name each object as it is generated.
That object_list is great for some things, but it isn't all that wonderful
for others, and the "name" member is generally not used by everybody.
This patch splits the users of the object_list array up into two: the
traditional list users, who want the list-like format, and who don't
actually use or want the name. And another class of users that really used
the list as an extensible array, and generally wanted to name the objects.
The patch is fairly straightforward, but it's also biggish. Most of it
really just cleans things up: switching the revision parsing and listing
over to the array makes things like the builtin-diff usage much simpler
(we now see exactly how many members the array has, and we don't get the
objects reversed from the order they were on the command line).
One of the main reasons for doing this at all is that the malloc overhead
of the simple object list was actually pretty high, and the array is just
a lot denser. So this patch brings down memory usage by git-rev-list by
just under 3% (on top of all the other memory use optimizations) on the
mozilla archive.
It does add more lines than it removes, and more importantly, it adds a
whole new infrastructure for maintaining lists of objects, but on the
other hand, the new dynamic array code is pretty obvious. The change to
builtin-diff-tree.c shows a fairly good example of why an array interface
is sometimes more natural, and just much simpler for everybody.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-06-20 02:42:35 +02:00
|
|
|
int i;
|
2006-02-28 20:24:00 +01:00
|
|
|
struct commit *commit;
|
Add "named object array" concept
We've had this notion of a "object_list" for a long time, which eventually
grew a "name" member because some users (notably git-rev-list) wanted to
name each object as it is generated.
That object_list is great for some things, but it isn't all that wonderful
for others, and the "name" member is generally not used by everybody.
This patch splits the users of the object_list array up into two: the
traditional list users, who want the list-like format, and who don't
actually use or want the name. And another class of users that really used
the list as an extensible array, and generally wanted to name the objects.
The patch is fairly straightforward, but it's also biggish. Most of it
really just cleans things up: switching the revision parsing and listing
over to the array makes things like the builtin-diff usage much simpler
(we now see exactly how many members the array has, and we don't get the
objects reversed from the order they were on the command line).
One of the main reasons for doing this at all is that the malloc overhead
of the simple object list was actually pretty high, and the array is just
a lot denser. So this patch brings down memory usage by git-rev-list by
just under 3% (on top of all the other memory use optimizations) on the
mozilla archive.
It does add more lines than it removes, and more importantly, it adds a
whole new infrastructure for maintaining lists of objects, but on the
other hand, the new dynamic array code is pretty obvious. The change to
builtin-diff-tree.c shows a fairly good example of why an array interface
is sometimes more natural, and just much simpler for everybody.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-06-20 02:42:35 +02:00
|
|
|
struct object_array objects = { 0, 0, NULL };
|
2005-06-02 18:19:53 +02:00
|
|
|
|
2006-02-28 20:24:00 +01:00
|
|
|
while ((commit = get_revision(revs)) != NULL) {
|
Add "named object array" concept
We've had this notion of a "object_list" for a long time, which eventually
grew a "name" member because some users (notably git-rev-list) wanted to
name each object as it is generated.
That object_list is great for some things, but it isn't all that wonderful
for others, and the "name" member is generally not used by everybody.
This patch splits the users of the object_list array up into two: the
traditional list users, who want the list-like format, and who don't
actually use or want the name. And another class of users that really used
the list as an extensible array, and generally wanted to name the objects.
The patch is fairly straightforward, but it's also biggish. Most of it
really just cleans things up: switching the revision parsing and listing
over to the array makes things like the builtin-diff usage much simpler
(we now see exactly how many members the array has, and we don't get the
objects reversed from the order they were on the command line).
One of the main reasons for doing this at all is that the malloc overhead
of the simple object list was actually pretty high, and the array is just
a lot denser. So this patch brings down memory usage by git-rev-list by
just under 3% (on top of all the other memory use optimizations) on the
mozilla archive.
It does add more lines than it removes, and more importantly, it adds a
whole new infrastructure for maintaining lists of objects, but on the
other hand, the new dynamic array code is pretty obvious. The change to
builtin-diff-tree.c shows a fairly good example of why an array interface
is sometimes more natural, and just much simpler for everybody.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-06-20 02:42:35 +02:00
|
|
|
process_tree(commit->tree, &objects, NULL, "");
|
2006-03-01 00:07:20 +01:00
|
|
|
show_commit(commit);
|
2005-06-02 18:19:53 +02:00
|
|
|
}
|
Add "named object array" concept
We've had this notion of a "object_list" for a long time, which eventually
grew a "name" member because some users (notably git-rev-list) wanted to
name each object as it is generated.
That object_list is great for some things, but it isn't all that wonderful
for others, and the "name" member is generally not used by everybody.
This patch splits the users of the object_list array up into two: the
traditional list users, who want the list-like format, and who don't
actually use or want the name. And another class of users that really used
the list as an extensible array, and generally wanted to name the objects.
The patch is fairly straightforward, but it's also biggish. Most of it
really just cleans things up: switching the revision parsing and listing
over to the array makes things like the builtin-diff usage much simpler
(we now see exactly how many members the array has, and we don't get the
objects reversed from the order they were on the command line).
One of the main reasons for doing this at all is that the malloc overhead
of the simple object list was actually pretty high, and the array is just
a lot denser. So this patch brings down memory usage by git-rev-list by
just under 3% (on top of all the other memory use optimizations) on the
mozilla archive.
It does add more lines than it removes, and more importantly, it adds a
whole new infrastructure for maintaining lists of objects, but on the
other hand, the new dynamic array code is pretty obvious. The change to
builtin-diff-tree.c shows a fairly good example of why an array interface
is sometimes more natural, and just much simpler for everybody.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-06-20 02:42:35 +02:00
|
|
|
for (i = 0; i < revs->pending.nr; i++) {
|
|
|
|
|
struct object_array_entry *pending = revs->pending.objects + i;
|
2005-06-29 20:30:24 +02:00
|
|
|
struct object *obj = pending->item;
|
|
|
|
|
const char *name = pending->name;
|
|
|
|
|
if (obj->flags & (UNINTERESTING | SEEN))
|
|
|
|
|
continue;
|
Shrink "struct object" a bit
This shrinks "struct object" by a small amount, by getting rid of the
"struct type *" pointer and replacing it with a 3-bit bitfield instead.
In addition, we merge the bitfields and the "flags" field, which
incidentally should also remove a useless 4-byte padding from the object
when in 64-bit mode.
Now, our "struct object" is still too damn large, but it's now less
obviously bloated, and of the remaining fields, only the "util" (which is
not used by most things) is clearly something that should be eventually
discarded.
This shrinks the "git-rev-list --all" memory use by about 2.5% on the
kernel archive (and, perhaps more importantly, on the larger mozilla
archive). That may not sound like much, but I suspect it's more on a
64-bit platform.
There are other remaining inefficiencies (the parent lists, for example,
probably have horrible malloc overhead), but this was pretty obvious.
Most of the patch is just changing the comparison of the "type" pointer
from one of the constant string pointers to the appropriate new TYPE_xxx
small integer constant.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-06-15 01:45:13 +02:00
|
|
|
if (obj->type == TYPE_TAG) {
|
2005-06-29 20:30:24 +02:00
|
|
|
obj->flags |= SEEN;
|
Add "named object array" concept
We've had this notion of a "object_list" for a long time, which eventually
grew a "name" member because some users (notably git-rev-list) wanted to
name each object as it is generated.
That object_list is great for some things, but it isn't all that wonderful
for others, and the "name" member is generally not used by everybody.
This patch splits the users of the object_list array up into two: the
traditional list users, who want the list-like format, and who don't
actually use or want the name. And another class of users that really used
the list as an extensible array, and generally wanted to name the objects.
The patch is fairly straightforward, but it's also biggish. Most of it
really just cleans things up: switching the revision parsing and listing
over to the array makes things like the builtin-diff usage much simpler
(we now see exactly how many members the array has, and we don't get the
objects reversed from the order they were on the command line).
One of the main reasons for doing this at all is that the malloc overhead
of the simple object list was actually pretty high, and the array is just
a lot denser. So this patch brings down memory usage by git-rev-list by
just under 3% (on top of all the other memory use optimizations) on the
mozilla archive.
It does add more lines than it removes, and more importantly, it adds a
whole new infrastructure for maintaining lists of objects, but on the
other hand, the new dynamic array code is pretty obvious. The change to
builtin-diff-tree.c shows a fairly good example of why an array interface
is sometimes more natural, and just much simpler for everybody.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-06-20 02:42:35 +02:00
|
|
|
add_object_array(obj, name, &objects);
|
2005-06-29 20:30:24 +02:00
|
|
|
continue;
|
|
|
|
|
}
|
Shrink "struct object" a bit
This shrinks "struct object" by a small amount, by getting rid of the
"struct type *" pointer and replacing it with a 3-bit bitfield instead.
In addition, we merge the bitfields and the "flags" field, which
incidentally should also remove a useless 4-byte padding from the object
when in 64-bit mode.
Now, our "struct object" is still too damn large, but it's now less
obviously bloated, and of the remaining fields, only the "util" (which is
not used by most things) is clearly something that should be eventually
discarded.
This shrinks the "git-rev-list --all" memory use by about 2.5% on the
kernel archive (and, perhaps more importantly, on the larger mozilla
archive). That may not sound like much, but I suspect it's more on a
64-bit platform.
There are other remaining inefficiencies (the parent lists, for example,
probably have horrible malloc overhead), but this was pretty obvious.
Most of the patch is just changing the comparison of the "type" pointer
from one of the constant string pointers to the appropriate new TYPE_xxx
small integer constant.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-06-15 01:45:13 +02:00
|
|
|
if (obj->type == TYPE_TREE) {
|
Add "named object array" concept
We've had this notion of a "object_list" for a long time, which eventually
grew a "name" member because some users (notably git-rev-list) wanted to
name each object as it is generated.
That object_list is great for some things, but it isn't all that wonderful
for others, and the "name" member is generally not used by everybody.
This patch splits the users of the object_list array up into two: the
traditional list users, who want the list-like format, and who don't
actually use or want the name. And another class of users that really used
the list as an extensible array, and generally wanted to name the objects.
The patch is fairly straightforward, but it's also biggish. Most of it
really just cleans things up: switching the revision parsing and listing
over to the array makes things like the builtin-diff usage much simpler
(we now see exactly how many members the array has, and we don't get the
objects reversed from the order they were on the command line).
One of the main reasons for doing this at all is that the malloc overhead
of the simple object list was actually pretty high, and the array is just
a lot denser. So this patch brings down memory usage by git-rev-list by
just under 3% (on top of all the other memory use optimizations) on the
mozilla archive.
It does add more lines than it removes, and more importantly, it adds a
whole new infrastructure for maintaining lists of objects, but on the
other hand, the new dynamic array code is pretty obvious. The change to
builtin-diff-tree.c shows a fairly good example of why an array interface
is sometimes more natural, and just much simpler for everybody.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-06-20 02:42:35 +02:00
|
|
|
process_tree((struct tree *)obj, &objects, NULL, name);
|
2005-06-29 20:30:24 +02:00
|
|
|
continue;
|
|
|
|
|
}
|
Shrink "struct object" a bit
This shrinks "struct object" by a small amount, by getting rid of the
"struct type *" pointer and replacing it with a 3-bit bitfield instead.
In addition, we merge the bitfields and the "flags" field, which
incidentally should also remove a useless 4-byte padding from the object
when in 64-bit mode.
Now, our "struct object" is still too damn large, but it's now less
obviously bloated, and of the remaining fields, only the "util" (which is
not used by most things) is clearly something that should be eventually
discarded.
This shrinks the "git-rev-list --all" memory use by about 2.5% on the
kernel archive (and, perhaps more importantly, on the larger mozilla
archive). That may not sound like much, but I suspect it's more on a
64-bit platform.
There are other remaining inefficiencies (the parent lists, for example,
probably have horrible malloc overhead), but this was pretty obvious.
Most of the patch is just changing the comparison of the "type" pointer
from one of the constant string pointers to the appropriate new TYPE_xxx
small integer constant.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-06-15 01:45:13 +02:00
|
|
|
if (obj->type == TYPE_BLOB) {
|
Add "named object array" concept
We've had this notion of a "object_list" for a long time, which eventually
grew a "name" member because some users (notably git-rev-list) wanted to
name each object as it is generated.
That object_list is great for some things, but it isn't all that wonderful
for others, and the "name" member is generally not used by everybody.
This patch splits the users of the object_list array up into two: the
traditional list users, who want the list-like format, and who don't
actually use or want the name. And another class of users that really used
the list as an extensible array, and generally wanted to name the objects.
The patch is fairly straightforward, but it's also biggish. Most of it
really just cleans things up: switching the revision parsing and listing
over to the array makes things like the builtin-diff usage much simpler
(we now see exactly how many members the array has, and we don't get the
objects reversed from the order they were on the command line).
One of the main reasons for doing this at all is that the malloc overhead
of the simple object list was actually pretty high, and the array is just
a lot denser. So this patch brings down memory usage by git-rev-list by
just under 3% (on top of all the other memory use optimizations) on the
mozilla archive.
It does add more lines than it removes, and more importantly, it adds a
whole new infrastructure for maintaining lists of objects, but on the
other hand, the new dynamic array code is pretty obvious. The change to
builtin-diff-tree.c shows a fairly good example of why an array interface
is sometimes more natural, and just much simpler for everybody.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-06-20 02:42:35 +02:00
|
|
|
process_blob((struct blob *)obj, &objects, NULL, name);
|
2005-06-29 20:30:24 +02:00
|
|
|
continue;
|
|
|
|
|
}
|
|
|
|
|
die("unknown pending object %s (%s)", sha1_to_hex(obj->sha1), name);
|
|
|
|
|
}
|
Add "named object array" concept
We've had this notion of a "object_list" for a long time, which eventually
grew a "name" member because some users (notably git-rev-list) wanted to
name each object as it is generated.
That object_list is great for some things, but it isn't all that wonderful
for others, and the "name" member is generally not used by everybody.
This patch splits the users of the object_list array up into two: the
traditional list users, who want the list-like format, and who don't
actually use or want the name. And another class of users that really used
the list as an extensible array, and generally wanted to name the objects.
The patch is fairly straightforward, but it's also biggish. Most of it
really just cleans things up: switching the revision parsing and listing
over to the array makes things like the builtin-diff usage much simpler
(we now see exactly how many members the array has, and we don't get the
objects reversed from the order they were on the command line).
One of the main reasons for doing this at all is that the malloc overhead
of the simple object list was actually pretty high, and the array is just
a lot denser. So this patch brings down memory usage by git-rev-list by
just under 3% (on top of all the other memory use optimizations) on the
mozilla archive.
It does add more lines than it removes, and more importantly, it adds a
whole new infrastructure for maintaining lists of objects, but on the
other hand, the new dynamic array code is pretty obvious. The change to
builtin-diff-tree.c shows a fairly good example of why an array interface
is sometimes more natural, and just much simpler for everybody.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-06-20 02:42:35 +02:00
|
|
|
for (i = 0; i < objects.nr; i++) {
|
|
|
|
|
struct object_array_entry *p = objects.objects + i;
|
|
|
|
|
|
2006-02-22 10:27:02 +01:00
|
|
|
/* An object with name "foo\n0000000..." can be used to
|
|
|
|
|
* confuse downstream git-pack-objects very badly.
|
2005-10-03 02:29:21 +02:00
|
|
|
*/
|
Add "named object array" concept
We've had this notion of a "object_list" for a long time, which eventually
grew a "name" member because some users (notably git-rev-list) wanted to
name each object as it is generated.
That object_list is great for some things, but it isn't all that wonderful
for others, and the "name" member is generally not used by everybody.
This patch splits the users of the object_list array up into two: the
traditional list users, who want the list-like format, and who don't
actually use or want the name. And another class of users that really used
the list as an extensible array, and generally wanted to name the objects.
The patch is fairly straightforward, but it's also biggish. Most of it
really just cleans things up: switching the revision parsing and listing
over to the array makes things like the builtin-diff usage much simpler
(we now see exactly how many members the array has, and we don't get the
objects reversed from the order they were on the command line).
One of the main reasons for doing this at all is that the malloc overhead
of the simple object list was actually pretty high, and the array is just
a lot denser. So this patch brings down memory usage by git-rev-list by
just under 3% (on top of all the other memory use optimizations) on the
mozilla archive.
It does add more lines than it removes, and more importantly, it adds a
whole new infrastructure for maintaining lists of objects, but on the
other hand, the new dynamic array code is pretty obvious. The change to
builtin-diff-tree.c shows a fairly good example of why an array interface
is sometimes more natural, and just much simpler for everybody.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-06-20 02:42:35 +02:00
|
|
|
const char *ep = strchr(p->name, '\n');
|
2005-10-03 02:29:21 +02:00
|
|
|
if (ep) {
|
Add "named object array" concept
We've had this notion of a "object_list" for a long time, which eventually
grew a "name" member because some users (notably git-rev-list) wanted to
name each object as it is generated.
That object_list is great for some things, but it isn't all that wonderful
for others, and the "name" member is generally not used by everybody.
This patch splits the users of the object_list array up into two: the
traditional list users, who want the list-like format, and who don't
actually use or want the name. And another class of users that really used
the list as an extensible array, and generally wanted to name the objects.
The patch is fairly straightforward, but it's also biggish. Most of it
really just cleans things up: switching the revision parsing and listing
over to the array makes things like the builtin-diff usage much simpler
(we now see exactly how many members the array has, and we don't get the
objects reversed from the order they were on the command line).
One of the main reasons for doing this at all is that the malloc overhead
of the simple object list was actually pretty high, and the array is just
a lot denser. So this patch brings down memory usage by git-rev-list by
just under 3% (on top of all the other memory use optimizations) on the
mozilla archive.
It does add more lines than it removes, and more importantly, it adds a
whole new infrastructure for maintaining lists of objects, but on the
other hand, the new dynamic array code is pretty obvious. The change to
builtin-diff-tree.c shows a fairly good example of why an array interface
is sometimes more natural, and just much simpler for everybody.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-06-20 02:42:35 +02:00
|
|
|
printf("%s %.*s\n", sha1_to_hex(p->item->sha1),
|
|
|
|
|
(int) (ep - p->name),
|
|
|
|
|
p->name);
|
2005-10-03 02:29:21 +02:00
|
|
|
}
|
|
|
|
|
else
|
Add "named object array" concept
We've had this notion of a "object_list" for a long time, which eventually
grew a "name" member because some users (notably git-rev-list) wanted to
name each object as it is generated.
That object_list is great for some things, but it isn't all that wonderful
for others, and the "name" member is generally not used by everybody.
This patch splits the users of the object_list array up into two: the
traditional list users, who want the list-like format, and who don't
actually use or want the name. And another class of users that really used
the list as an extensible array, and generally wanted to name the objects.
The patch is fairly straightforward, but it's also biggish. Most of it
really just cleans things up: switching the revision parsing and listing
over to the array makes things like the builtin-diff usage much simpler
(we now see exactly how many members the array has, and we don't get the
objects reversed from the order they were on the command line).
One of the main reasons for doing this at all is that the malloc overhead
of the simple object list was actually pretty high, and the array is just
a lot denser. So this patch brings down memory usage by git-rev-list by
just under 3% (on top of all the other memory use optimizations) on the
mozilla archive.
It does add more lines than it removes, and more importantly, it adds a
whole new infrastructure for maintaining lists of objects, but on the
other hand, the new dynamic array code is pretty obvious. The change to
builtin-diff-tree.c shows a fairly good example of why an array interface
is sometimes more natural, and just much simpler for everybody.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-06-20 02:42:35 +02:00
|
|
|
printf("%s %s\n", sha1_to_hex(p->item->sha1), p->name);
|
2005-06-25 07:56:58 +02:00
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2005-06-18 07:54:50 +02:00
|
|
|
/*
|
|
|
|
|
* This is a truly stupid algorithm, but it's only
|
|
|
|
|
* used for bisection, and we just don't care enough.
|
|
|
|
|
*
|
|
|
|
|
* We care just barely enough to avoid recursing for
|
|
|
|
|
* non-merge entries.
|
|
|
|
|
*/
|
|
|
|
|
static int count_distance(struct commit_list *entry)
|
|
|
|
|
{
|
|
|
|
|
int nr = 0;
|
|
|
|
|
|
|
|
|
|
while (entry) {
|
|
|
|
|
struct commit *commit = entry->item;
|
|
|
|
|
struct commit_list *p;
|
|
|
|
|
|
|
|
|
|
if (commit->object.flags & (UNINTERESTING | COUNTED))
|
|
|
|
|
break;
|
2006-03-10 10:21:39 +01:00
|
|
|
if (!revs.prune_fn || (commit->object.flags & TREECHANGE))
|
2005-11-27 20:32:03 +01:00
|
|
|
nr++;
|
2005-06-18 07:54:50 +02:00
|
|
|
commit->object.flags |= COUNTED;
|
|
|
|
|
p = commit->parents;
|
|
|
|
|
entry = p;
|
|
|
|
|
if (p) {
|
|
|
|
|
p = p->next;
|
|
|
|
|
while (p) {
|
|
|
|
|
nr += count_distance(p);
|
|
|
|
|
p = p->next;
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
}
|
2005-11-27 20:32:03 +01:00
|
|
|
|
2005-06-18 07:54:50 +02:00
|
|
|
return nr;
|
|
|
|
|
}
|
|
|
|
|
|
2005-06-19 05:02:49 +02:00
|
|
|
static void clear_distance(struct commit_list *list)
|
2005-06-18 07:54:50 +02:00
|
|
|
{
|
|
|
|
|
while (list) {
|
|
|
|
|
struct commit *commit = list->item;
|
|
|
|
|
commit->object.flags &= ~COUNTED;
|
|
|
|
|
list = list->next;
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static struct commit_list *find_bisection(struct commit_list *list)
|
|
|
|
|
{
|
|
|
|
|
int nr, closest;
|
|
|
|
|
struct commit_list *p, *best;
|
|
|
|
|
|
|
|
|
|
nr = 0;
|
|
|
|
|
p = list;
|
|
|
|
|
while (p) {
|
2006-03-10 10:21:39 +01:00
|
|
|
if (!revs.prune_fn || (p->item->object.flags & TREECHANGE))
|
2005-11-27 20:32:03 +01:00
|
|
|
nr++;
|
2005-06-18 07:54:50 +02:00
|
|
|
p = p->next;
|
|
|
|
|
}
|
|
|
|
|
closest = 0;
|
|
|
|
|
best = list;
|
|
|
|
|
|
2005-11-27 20:32:03 +01:00
|
|
|
for (p = list; p; p = p->next) {
|
|
|
|
|
int distance;
|
|
|
|
|
|
2006-03-10 10:21:39 +01:00
|
|
|
if (revs.prune_fn && !(p->item->object.flags & TREECHANGE))
|
2005-11-27 20:32:03 +01:00
|
|
|
continue;
|
|
|
|
|
|
|
|
|
|
distance = count_distance(p);
|
2005-06-18 07:54:50 +02:00
|
|
|
clear_distance(list);
|
|
|
|
|
if (nr - distance < distance)
|
|
|
|
|
distance = nr - distance;
|
|
|
|
|
if (distance > closest) {
|
|
|
|
|
best = p;
|
|
|
|
|
closest = distance;
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
if (best)
|
|
|
|
|
best->next = NULL;
|
|
|
|
|
return best;
|
|
|
|
|
}
|
|
|
|
|
|
2006-02-19 12:32:31 +01:00
|
|
|
static void mark_edge_parents_uninteresting(struct commit *commit)
|
|
|
|
|
{
|
|
|
|
|
struct commit_list *parents;
|
|
|
|
|
|
|
|
|
|
for (parents = commit->parents; parents; parents = parents->next) {
|
|
|
|
|
struct commit *parent = parents->item;
|
|
|
|
|
if (!(parent->object.flags & UNINTERESTING))
|
|
|
|
|
continue;
|
|
|
|
|
mark_tree_uninteresting(parent->tree);
|
2006-02-26 01:19:46 +01:00
|
|
|
if (revs.edge_hint && !(parent->object.flags & SHOWN)) {
|
2006-02-24 08:44:15 +01:00
|
|
|
parent->object.flags |= SHOWN;
|
2006-02-19 12:32:31 +01:00
|
|
|
printf("-%s\n", sha1_to_hex(parent->object.sha1));
|
2006-02-24 08:44:15 +01:00
|
|
|
}
|
2006-02-19 12:32:31 +01:00
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2005-09-16 00:14:29 +02:00
|
|
|
static void mark_edges_uninteresting(struct commit_list *list)
|
|
|
|
|
{
|
|
|
|
|
for ( ; list; list = list->next) {
|
2006-02-19 12:32:31 +01:00
|
|
|
struct commit *commit = list->item;
|
2005-09-16 00:14:29 +02:00
|
|
|
|
2006-02-19 12:32:31 +01:00
|
|
|
if (commit->object.flags & UNINTERESTING) {
|
|
|
|
|
mark_tree_uninteresting(commit->tree);
|
|
|
|
|
continue;
|
2005-09-16 00:14:29 +02:00
|
|
|
}
|
2006-02-19 12:32:31 +01:00
|
|
|
mark_edge_parents_uninteresting(commit);
|
2005-09-16 00:14:29 +02:00
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2006-05-18 23:19:20 +02:00
|
|
|
int cmd_rev_list(int argc, const char **argv, char **envp)
|
2005-04-24 04:04:40 +02:00
|
|
|
{
|
2006-02-26 01:19:46 +01:00
|
|
|
struct commit_list *list;
|
2006-02-27 17:54:36 +01:00
|
|
|
int i;
|
2005-04-24 04:04:40 +02:00
|
|
|
|
2006-04-16 08:48:27 +02:00
|
|
|
init_revisions(&revs);
|
|
|
|
|
revs.abbrev = 0;
|
|
|
|
|
revs.commit_format = CMIT_FMT_UNSPECIFIED;
|
2006-02-28 20:24:00 +01:00
|
|
|
argc = setup_revisions(argc, argv, &revs, NULL);
|
2006-02-26 01:19:46 +01:00
|
|
|
|
2005-05-06 10:00:11 +02:00
|
|
|
for (i = 1 ; i < argc; i++) {
|
2005-10-21 06:25:09 +02:00
|
|
|
const char *arg = argv[i];
|
2005-05-06 10:00:11 +02:00
|
|
|
|
2005-05-26 03:29:09 +02:00
|
|
|
if (!strcmp(arg, "--header")) {
|
2006-04-16 08:48:27 +02:00
|
|
|
revs.verbose_header = 1;
|
2005-06-01 17:42:22 +02:00
|
|
|
continue;
|
|
|
|
|
}
|
2006-03-22 09:22:00 +01:00
|
|
|
if (!strcmp(arg, "--timestamp")) {
|
|
|
|
|
show_timestamp = 1;
|
|
|
|
|
continue;
|
|
|
|
|
}
|
2005-06-18 07:54:50 +02:00
|
|
|
if (!strcmp(arg, "--bisect")) {
|
|
|
|
|
bisect_list = 1;
|
|
|
|
|
continue;
|
|
|
|
|
}
|
2006-02-26 01:19:46 +01:00
|
|
|
usage(rev_list_usage);
|
2005-05-26 03:29:09 +02:00
|
|
|
|
2005-05-06 10:00:11 +02:00
|
|
|
}
|
2006-04-16 08:48:27 +02:00
|
|
|
if (revs.commit_format != CMIT_FMT_UNSPECIFIED) {
|
|
|
|
|
/* The command line has a --pretty */
|
|
|
|
|
hdr_termination = '\n';
|
|
|
|
|
if (revs.commit_format == CMIT_FMT_ONELINE)
|
Log message printout cleanups
On Sun, 16 Apr 2006, Junio C Hamano wrote:
>
> In the mid-term, I am hoping we can drop the generate_header()
> callchain _and_ the custom code that formats commit log in-core,
> found in cmd_log_wc().
Ok, this was nastier than expected, just because the dependencies between
the different log-printing stuff were absolutely _everywhere_, but here's
a patch that does exactly that.
The patch is not very easy to read, and the "--patch-with-stat" thing is
still broken (it does not call the "show_log()" thing properly for
merges). That's not a new bug. In the new world order it _should_ do
something like
if (rev->logopt)
show_log(rev, rev->logopt, "---\n");
but it doesn't. I haven't looked at the --with-stat logic, so I left it
alone.
That said, this patch removes more lines than it adds, and in particular,
the "cmd_log_wc()" loop is now a very clean:
while ((commit = get_revision(rev)) != NULL) {
log_tree_commit(rev, commit);
free(commit->buffer);
commit->buffer = NULL;
}
so it doesn't get much prettier than this. All the complexity is entirely
hidden in log-tree.c, and any code that needs to flush the log literally
just needs to do the "if (rev->logopt) show_log(...)" incantation.
I had to make the combined_diff() logic take a "struct rev_info" instead
of just a "struct diff_options", but that part is pretty clean.
This does change "git whatchanged" from using "diff-tree" as the commit
descriptor to "commit", and I changed one of the tests to reflect that new
reality. Otherwise everything still passes, and my other tests look fine
too.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-04-17 20:59:32 +02:00
|
|
|
header_prefix = "";
|
2006-04-16 08:48:27 +02:00
|
|
|
else
|
Log message printout cleanups
On Sun, 16 Apr 2006, Junio C Hamano wrote:
>
> In the mid-term, I am hoping we can drop the generate_header()
> callchain _and_ the custom code that formats commit log in-core,
> found in cmd_log_wc().
Ok, this was nastier than expected, just because the dependencies between
the different log-printing stuff were absolutely _everywhere_, but here's
a patch that does exactly that.
The patch is not very easy to read, and the "--patch-with-stat" thing is
still broken (it does not call the "show_log()" thing properly for
merges). That's not a new bug. In the new world order it _should_ do
something like
if (rev->logopt)
show_log(rev, rev->logopt, "---\n");
but it doesn't. I haven't looked at the --with-stat logic, so I left it
alone.
That said, this patch removes more lines than it adds, and in particular,
the "cmd_log_wc()" loop is now a very clean:
while ((commit = get_revision(rev)) != NULL) {
log_tree_commit(rev, commit);
free(commit->buffer);
commit->buffer = NULL;
}
so it doesn't get much prettier than this. All the complexity is entirely
hidden in log-tree.c, and any code that needs to flush the log literally
just needs to do the "if (rev->logopt) show_log(...)" incantation.
I had to make the combined_diff() logic take a "struct rev_info" instead
of just a "struct diff_options", but that part is pretty clean.
This does change "git whatchanged" from using "diff-tree" as the commit
descriptor to "commit", and I changed one of the tests to reflect that new
reality. Otherwise everything still passes, and my other tests look fine
too.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-04-17 20:59:32 +02:00
|
|
|
header_prefix = "commit ";
|
2006-04-16 08:48:27 +02:00
|
|
|
}
|
2006-04-17 21:42:36 +02:00
|
|
|
else if (revs.verbose_header)
|
|
|
|
|
/* Only --header was specified */
|
|
|
|
|
revs.commit_format = CMIT_FMT_RAW;
|
2005-05-06 10:00:11 +02:00
|
|
|
|
2006-02-26 01:19:46 +01:00
|
|
|
list = revs.commits;
|
|
|
|
|
|
2006-04-15 07:43:34 +02:00
|
|
|
if ((!list &&
|
|
|
|
|
(!(revs.tag_objects||revs.tree_objects||revs.blob_objects) &&
|
Add "named object array" concept
We've had this notion of a "object_list" for a long time, which eventually
grew a "name" member because some users (notably git-rev-list) wanted to
name each object as it is generated.
That object_list is great for some things, but it isn't all that wonderful
for others, and the "name" member is generally not used by everybody.
This patch splits the users of the object_list array up into two: the
traditional list users, who want the list-like format, and who don't
actually use or want the name. And another class of users that really used
the list as an extensible array, and generally wanted to name the objects.
The patch is fairly straightforward, but it's also biggish. Most of it
really just cleans things up: switching the revision parsing and listing
over to the array makes things like the builtin-diff usage much simpler
(we now see exactly how many members the array has, and we don't get the
objects reversed from the order they were on the command line).
One of the main reasons for doing this at all is that the malloc overhead
of the simple object list was actually pretty high, and the array is just
a lot denser. So this patch brings down memory usage by git-rev-list by
just under 3% (on top of all the other memory use optimizations) on the
mozilla archive.
It does add more lines than it removes, and more importantly, it adds a
whole new infrastructure for maintaining lists of objects, but on the
other hand, the new dynamic array code is pretty obvious. The change to
builtin-diff-tree.c shows a fairly good example of why an array interface
is sometimes more natural, and just much simpler for everybody.
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-06-20 02:42:35 +02:00
|
|
|
!revs.pending.nr)) ||
|
2006-04-15 07:43:34 +02:00
|
|
|
revs.diff)
|
2005-10-26 00:24:55 +02:00
|
|
|
usage(rev_list_usage);
|
|
|
|
|
|
2006-04-16 08:48:27 +02:00
|
|
|
save_commit_buffer = revs.verbose_header;
|
2006-03-29 03:28:04 +02:00
|
|
|
track_object_refs = 0;
|
rev-list --bisect: limit list before bisecting.
I noticed bisect does not work well without both good and bad.
Running this script in git.git repository would give you quite
different results:
#!/bin/sh
initial=e83c5163316f89bfbde7d9ab23ca2e25604af290
mid0=`git rev-list --bisect ^$initial --all`
git rev-list $mid0 | wc -l
git rev-list ^$mid0 --all | wc -l
mid1=`git rev-list --bisect --all`
git rev-list $mid1 | wc -l
git rev-list ^$mid1 --all | wc -l
The $initial commit is the very first commit you made. The
first midpoint bisects things evenly as designed, but the latter
does not.
The reason I got interested in this was because I was wondering
if something like the following would help people converting a
huge repository from foreign SCM, or preparing a repository to
be fetched over plain dumb HTTP only:
#!/bin/sh
N=4
P=.git/objects/pack
bottom=
while test 0 \< $N
do
N=$((N-1))
if test -z "$bottom"
then
newbottom=`git rev-list --bisect --all`
else
newbottom=`git rev-list --bisect ^$bottom --all`
fi
if test -z "$bottom"
then
rev_list="$newbottom"
elif test 0 = $N
then
rev_list="^$bottom --all"
else
rev_list="^$bottom $newbottom"
fi
p=$(git rev-list --unpacked --objects $rev_list |
git pack-objects $P/pack)
git show-index <$P/pack-$p.idx | wc -l
bottom=$newbottom
done
The idea is to pack older half of the history to one pack, then
older half of the remaining history to another, to continue a
few times, using finer granularity as we get closer to the tip.
This may not matter, since for a truly huge history, running
bisect number of times could be quite time consuming, and we
might be better off running "git rev-list --all" once into a
temporary file, and manually pick cut-off points from the
resulting list of commits. After all we are talking about
"approximately half" for such an usage, and older history does
not matter much.
Signed-off-by: Junio C Hamano <junkio@cox.net>
2006-04-15 00:57:32 +02:00
|
|
|
if (bisect_list)
|
|
|
|
|
revs.limited = 1;
|
2006-03-29 03:28:04 +02:00
|
|
|
|
2006-02-28 20:24:00 +01:00
|
|
|
prepare_revision_walk(&revs);
|
|
|
|
|
if (revs.tree_objects)
|
|
|
|
|
mark_edges_uninteresting(revs.commits);
|
|
|
|
|
|
|
|
|
|
if (bisect_list)
|
|
|
|
|
revs.commits = find_bisection(revs.commits);
|
2005-10-26 00:24:55 +02:00
|
|
|
|
2006-03-01 00:07:20 +01:00
|
|
|
show_commit_list(&revs);
|
2005-05-31 03:46:32 +02:00
|
|
|
|
2005-04-24 04:04:40 +02:00
|
|
|
return 0;
|
|
|
|
|
}
|