In particular, pushing the "path_name()" call _into_ the show() function
would seem to allow
- more clarity into who "owns" the name (ie now when we free the name in
the show_object callback, it's because we generated it ourselves by
calling path_name())
- not calling path_name() at all, either because we don't care about the
name in the first place, or because we are actually happy walking the
linked list of "struct name_path *" and the last component.
Now, I didn't do that latter optimization, because it would require some
more coding, but especially looking at "builtin-pack-objects.c", we really
don't even want the whole pathname, we really would be better off with the
list of path components.
Why? We use that name for two things:
- add_preferred_base_object(), which actually _wants_ to traverse the
path, and now does it by looking for '/' characters!
- for 'name_hash()', which only cares about the last 16 characters of a
name, so again, generating the full name seems to be just unnecessary
work.
Anyway, so I didn't look any closer at those things, but it did convince
me that the "show_object()" calling convention was crazy, and we're
actually better off doing _less_ in list-objects.c, and giving people
access to the internal data structures so that they can decide whether
they want to generate a path-name or not.
This patch does that, and then for people who did use the name (even if
they might do something more clever in the future), it just does the
straightforward "name = path_name(path, component); .. free(name);" thing.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Here's a less trivial thing, and slightly more dubious one.
I was looking at that "struct object_array objects", and wondering why we
do that. I have honestly totally forgotten. Why not just call the "show()"
function as we encounter the objects? Rather than add the objects to the
object_array, and then at the very end going through the array and doing a
'show' on all, just do things more incrementally.
Now, there are possible downsides to this:
- the "buffer using object_array" _can_ in theory result in at least
better I-cache usage (two tight loops rather than one more spread out
one). I don't think this is a real issue, but in theory..
- this _does_ change the order of the objects printed. Instead of doing a
"process_tree(revs, commit->tree, &objects, NULL, "");" in the loop
over the commits (which puts all the root trees _first_ in the object
list, this patch just adds them to the list of pending objects, and
then we'll traverse them in that order (and thus show each root tree
object together with the objects we discover under it)
I _think_ the new ordering actually makes more sense, but the object
ordering is actually a subtle thing when it comes to packing
efficiency, so any change in order is going to have implications for
packing. Good or bad, I dunno.
- There may be some reason why we did it that odd way with the object
array, that I have simply forgotten.
Anyway, now that we don't buffer up the objects before showing them
that may actually result in lower memory usage during that whole
traverse_commit_list() phase.
This is seriously not very deeply tested. It makes sense to me, it seems
to pass all the tests, it looks ok, but...
Does anybody remember why we did that "object_array" thing? It used to be
an "object_list" a long long time ago, but got changed into the array due
to better memory usage patterns (those linked lists of obejcts are
horrible from a memory allocation standpoint). But I wonder why we didn't
do this back then. Maybe there's a reason for it.
Or maybe there _used_ to be a reason, and no longer is.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
These links inside "stalenotes" section need to be updated on the master
branch every time a new stable or maintenance release is made.
Signed-off-by: Nanako Shiraishi <nanako3@lavabit.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
* jk/show-upstream:
branch: show upstream branch when double verbose
make get_short_ref a public function
for-each-ref: add "upstream" format field
for-each-ref: refactor refname handling
for-each-ref: refactor get_short_ref function
* fg/remote-prune:
add tests for remote groups
git remote update: Fallback to remote if group does not exist
remote: New function remote_is_configured()
git remote update: Report error for non-existing groups
git remote update: New option --prune
builtin-remote.c: Split out prune_remote as a separate function.
* cc/bisect-filter: (21 commits)
rev-list: add "int bisect_show_flags" in "struct rev_list_info"
rev-list: remove last static vars used in "show_commit"
list-objects: add "void *data" parameter to show functions
bisect--helper: string output variables together with "&&"
rev-list: pass "int flags" as last argument of "show_bisect_vars"
t6030: test bisecting with paths
bisect: use "bisect--helper" and remove "filter_skipped" function
bisect: implement "read_bisect_paths" to read paths in "$GIT_DIR/BISECT_NAMES"
bisect--helper: implement "git bisect--helper"
bisect: use the new generic "sha1_pos" function to lookup sha1
rev-list: call new "filter_skip" function
patch-ids: use the new generic "sha1_pos" function to lookup sha1
sha1-lookup: add new "sha1_pos" function to efficiently lookup sha1
rev-list: pass "revs" to "show_bisect_vars"
rev-list: make "show_bisect_vars" non static
rev-list: move code to show bisect vars into its own function
rev-list: move bisect related code into its own file
rev-list: make "bisect_list" variable local to "cmd_rev_list"
refs: add "for_each_ref_in" function to refactor "for_each_*_ref" functions
quote: add "sq_dequote_to_argv" to put unwrapped args in an argv array
...
* maint:
GIT 1.6.2.3
State the effect of filter-branch on graft explicitly
process_{tree,blob}: Remove useless xstrdup calls
Conflicts:
GIT-VERSION-GEN
The mergetool--lib scriplet was tricky to use because it relied upon
the existance of several global shell variables. This removes more
global variables so that things are simpler for callers.
A side effect is that some variables are recomputed each time
run_merge_tool() is called, but the overhead for recomputing
them is justified by the simpler implementation.
Signed-off-by: David Aguilar <davvid@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
On Wed, 8 Apr 2009, Björn Steinbrink wrote:
>
> The name of the processed object was duplicated for passing it to
> add_object(), but that already calls path_name, which allocates a new
> string anyway. So the memory allocated by the xstrdup calls just went
> nowhere, leaking memory.
Ack, ack.
There's another easy 5% or so for the built-in object walker: once we've
created the hash from the name, the name isn't interesting any more, and
so something trivial like this can help a bit.
Does it matter? Probably not on its own. But a few more memory saving
tricks and it might all make a difference.
Linus
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This test was added recently (5a688fe, "core.sharedrepository = 0mode"
should set, not loosen; 2009-03-28). It checked the result of a sed
invocation for emptyness, but in some cases it forgot to print anything
at all, so that those checks would never be false.
Due to this mistake, it went unnoticed that the files in objects/info are
not necessarily 0440, but can also be 0660. Because the 0mode setting
tries to guarantee that the files are accessible only to the people they
are meant to be used by, we should only make sure that they are readable
by the user and the group when the configuration is set to 0660. It is a
separate matter from the core.shredrepository settings that w-bit from
immutable object files under objects/[0-9a-f][0-9a-f] directories should
be dropped.
COMMIT_EDITMSG is still world-readable, but it (and any transient files
that are meant for repositories with a work tree) does not matter. If you
are working on a shared machine and on a sekrit stuff, the root of the
work tree would be with mode 0700 (or 0750 to allow peeking by other
people in the group), and that would mean that .git/COMMIT_EDITMSG in such
a repository would not be readable by the strangers anyway.
Also, in the real-world use case, .git/COMMIT_EDITMSG will be given to an
arbitrary editor the user happens to use, and we have no guarantee what it
does (e.g. it may create a new file with umask and replace, it may rewrite
in place, it may leave an editor backup file but use umask to create it,
etc.), and the protection of the file lies majorly on the protection of
the root of the work tree.
This test cannot be run on Windows; it requires POSIXPERM when merged to
'master'.
Signed-off-by: Johannes Sixt <j6t@kdbg.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
With "git add -e [<files>]", Git will fire up an editor with the current
diff relative to the index (i.e. what you would get with "git diff
[<files>]").
Now you can edit the patch as much as you like, including adding/removing
lines, editing the text, whatever. Make sure, though, that the first
character of the hunk lines is still a space, a plus or a minus.
After you closed the editor, Git will adjust the line counts of the hunks
if necessary, thanks to the --recount option of apply, and commit the
patch. Except if you deleted everything, in which case nothing happens
(for obvious reasons).
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Ensure that the Makefile that generates and installs the Documentation is
aware of any SHELL_PATH setting. Use this value if found or the current
setting for SHELL if not. This is an accommodation for systems where sh
is not POSIX enough.
Signed-off-by: Ben Walton <bwalton@artsci.utoronto.ca>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
I told people on the kernel mailing list to please use "-M" when sending
me rename patches, so that I can see what they do while reading email
rather than having to apply the patch and then look at the end result.
I also told them that if they want to make it the default, they can just
add
[diff]
renames
to their ~/.gitconfig file. And while I was thinking about that, I wanted
to also check whether you can then mark individual projects to _not_ have
that default in the per-repository .git/config file.
And you can't. Currently you cannot have a global "enable renames by
default" and then a local ".. but not for _this_ project". Why? Because if
somebody writes
[diff]
renames = no
we simply ignore it, rather than resetting "diff_detect_rename_default"
back to zero.
Fixed thusly.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
If the repo url or the user email contain XML special characters, the
remote DAV server is likely to reject the LOCK requests because the XML
is then malformed.
Signed-off-by: Mike Hommey <mh@glandium.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
It has been reported time and time again in relation to msysGit that
git-svn does not work well when core.autocrlf has any value other than
'false'. So let's make it so by default.
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Acked-by: Eric Wong <normalperson@yhbt.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
In the case of a small repository, pack-objects is smart enough to not
start more threads than necessary. However, the output to the user always
reports the value of the pack.threads configuration and not the real
number of threads to be used.
Signed-off-by: Dan McGee <dpmcgee@gmail.com>
Acked-by: Nicolas Pitre <nico@cam.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The --ignored-paths argument is now stored as
"svn-remote.$REMOTE_NAME.ignore-paths" in the config file.
[ew: edited subject and message]
Signed-off-by: Ben Jackson <ben@ben.com>
Acked-by: Eric Wong <normalperson@yhbt.net>
The --ignore-paths option to fetch is very useful for working on a subset
of a SVN repository. For proper operation, every command that causes a
fetch (explicit or implied) must include a matching --ignore-paths option.
This patch adds a persistent svn-remote.$repo_id.ignore-paths config by
promoting Fetcher::is_path_ignored to a member function and initializing
$self->{ignore_regex} in Fetcher::new. Command line --ignore-paths is
still recognized and acts in addition to the config value.
Signed-off-by: Ben Jackson <ben@ben.com>
Acked-by: Eric Wong <normalperson@yhbt.net>
'git svn blame' now uses the 'git cat-file --batch' command to
speed up resolving SVN revision number out of commit SHA by
removing fork+exec overhead.
[ew: enforced 80-column line wrap]
Signed-off-by: Boris Byk <boris.byk@gmail.com>
Acked-by: Eric Wong <normalperson@yhbt.net>
In several places mergetool.keepBackup was misspelled as merge.keepBackup.
Signed-off-by: Ferry Huberts <ferry.huberts@pelagic.nl>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This allows for example gitosis to allow use of 'git archive --remote' in a
controlled environment.
Signed-off-by: Erik Broes <erikbroes@ripe.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
On Thu, 9 Apr 2009, Linus Torvalds wrote:
>
> [diff]
> renames = no
Btw, while doing this, I also though that "renames = on/off" made more
sense, but while we allow yes/no and true/false for booleans, we don't
allow on/off.
Should we? Maybe. Here's a stupid patch.
Linus
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This adds a row context menu command to compare this commit and its
descendants with the marked commit and its descendants. The results
are shown in the bottom-left pane. Commits are compared by checking
whether their headlines are the same and their patches have the same
patch ID as generated by git patch-id.
Merges are ignored and skipped over (as long as they have one
descendant). If two commits have the same patch ID then the process
will continue and compare their descendants, as long as they both have
exactly one descendant. If either commit has 0 or 2 or more descendants,
the comparison stops there. There is currently a limit of 100
comparisons.
This can be useful for checking whether one string of commits is just
a rebased version of another string of commits. Mark the end of one
string (i.e. the oldest commit in the string) and invoke "Compare with
marked commit" on the end of the other string.
As this is implemented, the UI will be unresponsive while the results
are being generated. This should be fixed.
Signed-off-by: Paul Mackerras <paulus@samba.org>
This tries to systematically cover existing behavior, and
also mark some expect_failure cases for desired behavior.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
* jc/shared-literally:
t1301: loosen test for forced modes
set_shared_perm(): sometimes we know what the final mode bits should look like
move_temp_to_file(): do not forget to chmod() in "Coda hack" codepath
Move chmod(foo, 0444) into move_temp_to_file()
"core.sharedrepository = 0mode" should set, not loosen
* jc/maint-1.6.0-keep-pack:
pack-objects: don't loosen objects available in alternate or kept packs
t7700: demonstrate repack flaw which may loosen objects unnecessarily
Remove --kept-pack-only option and associated infrastructure
pack-objects: only repack or loosen objects residing in "local" packs
git-repack.sh: don't use --kept-pack-only option to pack-objects
t7700-repack: add two new tests demonstrating repacking flaws
is_kept_pack(): final clean-up
Simplify is_kept_pack()
Consolidate ignore_packed logic more
has_sha1_kept_pack(): take "struct rev_info"
has_sha1_pack(): refactor "pretend these packs do not exist" interface
git-repack: resist stray environment variable
Conflicts:
t/t7700-repack.sh
The name of the processed object was duplicated for passing it to
add_object(), but that already calls path_name, which allocates a new
string anyway. So the memory allocated by the xstrdup calls just went
nowhere, leaking memory.
This reduces the RSS usage for a "rev-list --all --objects" by about 10% on
the gentoo repo (fully packed) as well as linux-2.6.git:
gentoo:
| old | new
----------------|-------------------------------
RSS | 1537284 | 1388408
VSZ | 1816852 | 1667952
time elapsed | 1:49.62 | 1:48.99
min. page faults| 417178 | 379919
linux-2.6.git:
| old | new
----------------|-------------------------------
RSS | 324452 | 292996
VSZ | 491792 | 460376
time elapsed | 0:14.53 | 0:14.28
min. page faults| 89360 | 81613
Signed-off-by: Björn Steinbrink <B.Steinbrink@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
These scripts all test git programs that are written in
perl, and thus obviously won't work if NO_PERL is defined.
We pass NO_PERL to the scripts from the building Makefile
via the GIT-BUILD-OPTIONS file.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
For systems with a missing or broken perl, it is nicer to
explicitly say "we don't want perl" because:
1. The Makefile knows not to bother with Perl-ish things
like Git.pm.
2. We can print a more user-friendly error message
than "foo is not a git command" or whatever the broken
perl might barf
3. Test scripts that require perl can mark themselves and
such and be skipped
This patch implements parts (1) and (2). The perl/
subdirectory is skipped entirely, gitweb is not built, and
any git commands which rely on perl will print a
human-readable message and exit with an error code.
This patch is based on one from Robin H. Johnson.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This adds a context-menu command to put a mark on this commit. There
is at most one marked commit at any time, and it is indicated by a box
drawn around the headline. Once a commit is marked, two other
context-menu commands become available: one to select the marked commit,
and another to find the closest common descendant of this commit and
the marked commit.
The "find common descendant" command uses the displayed parent/child
relationships (i.e. the rewritten parent pointers produced by git log),
not the real parent/child relationships. Currently the UI will be
unresponsive while gitk is working out the nearest common descendant;
this should be improved in future.
Signed-off-by: Paul Mackerras <paulus@samba.org>
