Code

git.git
17 years agoShow total transferred as part of throughput progress
Nicolas Pitre [Thu, 1 Nov 2007 20:59:57 +0000 (16:59 -0400)]
Show total transferred as part of throughput progress

Right now it is infeasible to offer to the user a reasonable concept
of when a clone will be complete as we aren't able to come up with
the final pack size until after we have actually transferred the
entire thing to the client.  However in many cases users can work
with a rough rule-of-thumb; for example it is somewhat well known
that git.git is about 16 MiB today and that linux-2.6.git is over
120 MiB.

We now show the total amount of data we have transferred over
the network as part of the throughput meter, organizing it in
"human friendly" terms like `ls -h` would do.  Users can glance at
this, see that the total transferred size is about 3 MiB, see the
throughput of X KiB/sec, and determine a reasonable figure of about
when the clone will be complete, assuming they know the rough size
of the source repository or are able to obtain it.

This is also a helpful indicator that there is progress being made
even if we stall on a very large object.  The thoughput meter may
remain relatively constant and the percentage complete and object
count won't be changing, but the total transferred will be increasing
as additional data is received for this object.

[from an initial proposal from Shawn O. Pearce]

Signed-off-by: Nicolas Pitre <nico@cam.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
17 years agomake sure throughput display gets updated even if progress doesn't move
Nicolas Pitre [Thu, 1 Nov 2007 20:59:56 +0000 (16:59 -0400)]
make sure throughput display gets updated even if progress doesn't move

Currently the progress/throughput display update happens only through
display_progress().  If the progress based on object count remains
unchanged because a large object is being received, the latest throughput
won't be displayed.  The display update should occur through
display_throughput() as well.

Signed-off-by: Nicolas Pitre <nico@cam.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
17 years agoreturn the prune-packed progress display to the inner loop
Nicolas Pitre [Thu, 1 Nov 2007 20:59:55 +0000 (16:59 -0400)]
return the prune-packed progress display to the inner loop

This reverts commit 0e549137966feb016927a827fb6e359aec8264a3 so to return
to the same state as commit b5d72f0a4cd3cce945ca0d37e4fa0ebbfcdcdb52.

On Wed, 31 Oct 2007, Shawn O. Pearce wrote:
> During my testing with a 40,000 loose object case (yea, I fully
> unpacked a git.git clone I had laying around) my system stalled
> hard in the first object directory.  A *lot* longer than 1 second.
> So I got no progress meter for a long time, and then a progress
> meter appeared on the second directory.

Signed-off-by: Nicolas Pitre <nico@cam.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
17 years agoadd throughput display to git-push
Nicolas Pitre [Tue, 30 Oct 2007 21:06:21 +0000 (17:06 -0400)]
add throughput display to git-push

This one triggers only when git-pack-objects is called with
--all-progress and --stdout which is the combination used by
git-push.

Signed-off-by: Nicolas Pitre <nico@cam.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
17 years agoadd some copyright notice to the progress display code
Nicolas Pitre [Tue, 30 Oct 2007 19:41:13 +0000 (15:41 -0400)]
add some copyright notice to the progress display code

Some self patting on the back to keep my ego alive.

Signed-off-by: Nicolas Pitre <nico@cam.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
17 years agoadd throughput display to index-pack
Nicolas Pitre [Tue, 30 Oct 2007 18:57:35 +0000 (14:57 -0400)]
add throughput display to index-pack

... and call it "Receiving objects" when over stdin to look clearer
to end users.

Signed-off-by: Nicolas Pitre <nico@cam.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
17 years agoadd throughput to progress display
Nicolas Pitre [Tue, 30 Oct 2007 18:57:34 +0000 (14:57 -0400)]
add throughput to progress display

This adds the ability for the progress code to also display transfer
throughput when that makes sense.

The math was inspired by commit c548cf4ee0737a321ffe94f6a97c65baf87281be
from Linus.

Signed-off-by: Nicolas Pitre <nico@cam.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
17 years agorelax usage of the progress API
Nicolas Pitre [Tue, 30 Oct 2007 18:57:33 +0000 (14:57 -0400)]
relax usage of the progress API

Since it is now OK to pass a null pointer to display_progress() and
stop_progress() resulting in a no-op, then we can simplify the code
and remove a bunch of lines by not making those calls conditional all
the time.

Signed-off-by: Nicolas Pitre <nico@cam.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
17 years agomake struct progress an opaque type
Nicolas Pitre [Tue, 30 Oct 2007 18:57:32 +0000 (14:57 -0400)]
make struct progress an opaque type

This allows for better management of progress "object" existence,
as well as making the progress display implementation more independent
from its callers.

Signed-off-by: Nicolas Pitre <nico@cam.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
17 years agoprune-packed: don't call display_progress() for every file
Nicolas Pitre [Tue, 30 Oct 2007 18:57:31 +0000 (14:57 -0400)]
prune-packed: don't call display_progress() for every file

The progress count is per fanout directory, so it is useless to call
it for every file as the count doesn't change that often.

Signed-off-by: Nicolas Pitre <nico@cam.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
17 years agoStop displaying "Pack pack-$ID created." during git-gc
Shawn O. Pearce [Fri, 19 Oct 2007 05:01:40 +0000 (01:01 -0400)]
Stop displaying "Pack pack-$ID created." during git-gc

Discussion on the list tonight came to the conclusion that showing
the name of the packfile we just created during git-repack is not
a very useful message for any end-user.  For the really technical
folk who need to have the name of the newest packfile they can use
something such as `ls -t .git/objects/pack | head -2` to find the
most recently created packfile.

Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agoTeach prune-packed to use the standard progress meter
Shawn O. Pearce [Fri, 19 Oct 2007 04:08:37 +0000 (00:08 -0400)]
Teach prune-packed to use the standard progress meter

Rather than reimplementing the progress meter logic and always
showing 100 lines of output while pruning already packed objects
we now use a delayed progress meter and only show it if there are
enough objects to make us take a little while.

Most users won't see the message anymore as it usually doesn't take
very long to delete the already packed loose objects.  This neatens
the output of a git-gc or git-repack execution, which is especially
important for a `git gc --auto` triggered from within another
command.

We perform the display_progress() call from within the very innermost
loop in case we spend more than 1 second within any single object
directory.  This ensures that a progress_update event from the
timer will still trigger in a timely fashion and allow the user to
see the progress meter.

While I'm in here I changed the message to be more descriptive of
its actual task.  "Removing unused objects" is a little scary for
new users as they wonder where these unused objects came from and
how they should avoid them.  Truth is these objects aren't unused
in the sense of what git-prune would call a dangling object, these
are used but are just duplicates of things we have already stored
in a packfile.

Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agoChange 'Deltifying objects' to 'Compressing objects'
Shawn O. Pearce [Fri, 19 Oct 2007 00:42:20 +0000 (20:42 -0400)]
Change 'Deltifying objects' to 'Compressing objects'

Recently I was referred to the Grammar Police as the git-pack-objects
progress message 'Deltifying %u objects' is considered to be not
proper English to at least some small but vocal segment of the
English speaking population.  Techncially we are applying delta
compression to these objects at this stage, so the new term is
slightly more acceptable to the Grammar Police but is also just
as correct.

Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agofix for more minor memory leaks
Nicolas Pitre [Wed, 17 Oct 2007 01:55:50 +0000 (21:55 -0400)]
fix for more minor memory leaks

Now that some pointers have lost their const attribute, we can free their
associated memory when done with them.  This is more a correctness issue
about the rule for freeing those pointers which isn't completely trivial
more than the leak itself which didn't matter as the program is
exiting anyway.

Signed-off-by: Nicolas Pitre <nico@cam.org>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agofix const issues with some functions
Nicolas Pitre [Wed, 17 Oct 2007 01:55:49 +0000 (21:55 -0400)]
fix const issues with some functions

Two functions, namely write_idx_file() and open_pack_file(), currently
return a const pointer.  However that pointer is either a copy of the
first argument, or set to a malloc'd buffer when that first argument
is null.  In the later case it is wrong to qualify that pointer as const
since ownership of the buffer is transferred to the caller to dispose of,
and obviously the free() function is not meant to be passed const
pointers.

Making the return pointer not const causes a warning when the first
argument is returned since that argument is also marked const.

The correct thing to do is therefore to remove the const qualifiers,
avoiding the need for ugly casts only to silence some warnings.

Signed-off-by: Nicolas Pitre <nico@cam.org>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agopack-objects.c: fix some global variable abuse and memory leaks
Nicolas Pitre [Wed, 17 Oct 2007 01:55:48 +0000 (21:55 -0400)]
pack-objects.c: fix some global variable abuse and memory leaks

To keep things well layered, sha1close() now returns the file descriptor
when it doesn't close the file.

An ugly cast was added to the return of write_idx_file() to avoid a
warning.  A proper fix will come separately.

Signed-off-by: Nicolas Pitre <nico@cam.org>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agopack-objects: no delta possible with only one object in the list
Nicolas Pitre [Wed, 17 Oct 2007 01:55:47 +0000 (21:55 -0400)]
pack-objects: no delta possible with only one object in the list

... so don't even try in that case, and save another useless line of
progress display.

Signed-off-by: Nicolas Pitre <nico@cam.org>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agocope with multiple line breaks within sideband progress messages
Nicolas Pitre [Wed, 17 Oct 2007 01:55:46 +0000 (21:55 -0400)]
cope with multiple line breaks within sideband progress messages

A single sideband packet may sometimes contain multiple lines of progress
messages, but we prepend "remote: " only to the whole buffer which creates
a messed up display in that case.  Make sure that the "remote: " prefix
is applied to every remote lines.

Signed-off-by: Nicolas Pitre <nico@cam.org>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agomore compact progress display
Nicolas Pitre [Wed, 17 Oct 2007 01:55:45 +0000 (21:55 -0400)]
more compact progress display

Each progress can be on a single line instead of two.

[sp: Changed "Checking files out" to "Checking out files" at
     Johannes Sixt's suggestion as it better explains the
 action that is taking place]

Signed-off-by: Nicolas Pitre <nico@cam.org>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agogit-svn: simplify the handling of fatal errors
Benoit Sigoure [Tue, 16 Oct 2007 14:36:52 +0000 (16:36 +0200)]
git-svn: simplify the handling of fatal errors

* git-svn.perl (&fatal): Append the newline at the end of the error
message.
Adjust all callers.

Signed-off-by: Benoit Sigoure <tsuna@lrde.epita.fr>
Acked-by: Eric Wong <normalperson@yhbt.net>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agogit-svn: add git svn proplist
Benoit Sigoure [Tue, 16 Oct 2007 14:36:51 +0000 (16:36 +0200)]
git-svn: add git svn proplist

This allows one to easily retrieve a list of svn properties from within
git-svn without requiring svn or knowing the URL of a repository.

* git-svn.perl (%cmd): Add the command `proplist'.
(&cmd_proplist): New.
* t/t9101-git-svn-props.sh: Test git svn proplist.

Signed-off-by: Benoit Sigoure <tsuna@lrde.epita.fr>
Acked-by: Eric Wong <normalperson@yhbt.net>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agogit-svn: add git svn propget
Benoit Sigoure [Tue, 16 Oct 2007 14:36:50 +0000 (16:36 +0200)]
git-svn: add git svn propget

This allows one to easily retrieve a single SVN property from within
git-svn without requiring svn or remembering the URL of a repository

* git-svn.perl (%cmd): Add the new command `propget'.
($cmd_dir_prefix): New global.
(&get_svnprops): New helper.
(&cmd_propget): New.  Use &get_svnprops.
* t/t9101-git-svn-props.sh: Add a test case for propget.

[ew: make sure the rev-parse --show-prefix call doesn't break
     the `git-svn clone' command]

Signed-off-by: Benoit Sigoure <tsuna@lrde.epita.fr>
Acked-by: Eric Wong <normalperson@yhbt.net>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agogit-svn: implement git svn create-ignore
Benoit Sigoure [Tue, 16 Oct 2007 14:36:49 +0000 (16:36 +0200)]
git-svn: implement git svn create-ignore

git svn create-ignore (to create one .gitignore per directory
from the svn:ignore properties.  This has the disadvantage of
committing the .gitignore during the next dcommit, but when you
import a repo with tons of ignores (>1000), using git svn show-ignore
to build .git/info/exclude is *not* a good idea, because things like
git-status will end up doing >1000 fnmatch *per file* in the repo,
which leads to git-status taking more than 4s on my Core2Duo 2Ghz 2G
RAM)

* git-svn.perl (%cmd): Add the new command `create-ignore'.
(&cmd_create_ignore): New.
* t/t9101-git-svn-props.sh: Adjust the test-case for show-ignore and
add a test case for create-ignore.

[ew: added commit message from
  <05CAB148-56ED-4FF1-8AAB-4BA2A0B70C2C@lrde.epita.fr> ]

Signed-off-by: Benoit Sigoure <tsuna@lrde.epita.fr>
Acked-by: Eric Wong <normalperson@yhbt.net>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agogit-svn: add a generic tree traversal to fetch SVN properties
Benoit Sigoure [Tue, 16 Oct 2007 14:36:48 +0000 (16:36 +0200)]
git-svn: add a generic tree traversal to fetch SVN properties

* git-svn.perl (&traverse_ignore): Remove.
(&prop_walk): New.
(&cmd_show_ignore): Use prop_walk.

[ew: This will ease the implementation of the `create-ignore',
     `propget', and `proplist' commands]

Signed-off-by: Benoit Sigoure <tsuna@lrde.epita.fr>
Acked-by: Eric Wong <normalperson@yhbt.net>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agogitweb: speed up project listing on large work trees by limiting find depth
Luke Lu [Wed, 17 Oct 2007 03:45:25 +0000 (20:45 -0700)]
gitweb: speed up project listing on large work trees by limiting find depth

Signed-off-by: Luke Lu <git@vicaya.com>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agoMerge branch 'maint'
Shawn O. Pearce [Wed, 17 Oct 2007 03:32:03 +0000 (23:32 -0400)]
Merge branch 'maint'

* maint:
  Document additional 1.5.3.5 fixes in release notes
  Avoid 'expr index' on Mac OS X as it isn't supported
  filter-branch: update current branch when rewritten
  fix filter-branch documentation
  helpful error message when send-pack finds no refs in common.
  Fix setup_git_directory_gently() with relative GIT_DIR & GIT_WORK_TREE
  Correct typos in release notes for 1.5.3.5

17 years agoDocument additional 1.5.3.5 fixes in release notes
Shawn O. Pearce [Wed, 17 Oct 2007 03:31:58 +0000 (23:31 -0400)]
Document additional 1.5.3.5 fixes in release notes

Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agogit-cvsexportcommit.perl: git-apply no longer needs --binary
Michael Witten [Tue, 16 Oct 2007 08:08:14 +0000 (04:08 -0400)]
git-cvsexportcommit.perl: git-apply no longer needs --binary

Signed-off-by: Michael Witten <mfwitten@mit.edu>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agoAvoid 'expr index' on Mac OS X as it isn't supported
Shawn O. Pearce [Wed, 17 Oct 2007 01:33:11 +0000 (21:33 -0400)]
Avoid 'expr index' on Mac OS X as it isn't supported

This fixes git-instaweb so it can start an httpd without warning
about an invalid test command.  Yes its ugly, but its also quite
portable.

Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agofilter-branch: update current branch when rewritten
Johannes Schindelin [Wed, 17 Oct 2007 02:23:10 +0000 (03:23 +0100)]
filter-branch: update current branch when rewritten

Earlier, "git filter-branch --<options> HEAD" would not update the
working tree after rewriting the branch.  This commit fixes it.

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agofix filter-branch documentation
Johannes Schindelin [Wed, 17 Oct 2007 02:22:25 +0000 (03:22 +0100)]
fix filter-branch documentation

The man page for filter-branch still talked about writing the result
to the branch "newbranch".  This is hopefully the last place where the
old behaviour was described.

Noticed by Bill Lear.

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agohelpful error message when send-pack finds no refs in common.
Andrew Clausen [Tue, 16 Oct 2007 21:16:05 +0000 (17:16 -0400)]
helpful error message when send-pack finds no refs in common.

Signed-off-by: Andrew Clausen <clausen@econ.upenn.edu>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agogit-svn: use "no warnings 'once'" to disable false-positives
Eygene Ryabinkin [Mon, 15 Oct 2007 07:19:12 +0000 (11:19 +0400)]
git-svn: use "no warnings 'once'" to disable false-positives

Some variables coming from the Subversion's Perl bindings are used
in our code only once, so the interpreter warns us about it.  These
warnings are false-positives, because the variables themselves are
initialized in the binding's guts, that are made by SWIG.

Credits to Sam Vilain for his note about "no warnings 'once'".

[ew: minor formatting change]

Signed-off-by: Eygene Ryabinkin <rea-git@codelabs.ru>
Acked-by: Eric Wong <normalperson@yhbt.net>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agoFix setup_git_directory_gently() with relative GIT_DIR & GIT_WORK_TREE
Johannes Schindelin [Tue, 16 Oct 2007 23:37:36 +0000 (00:37 +0100)]
Fix setup_git_directory_gently() with relative GIT_DIR & GIT_WORK_TREE

There are a few programs, such as config and diff, which allow running
without a git repository.  Therefore, they have to call
setup_git_directory_gently().

However, when GIT_DIR and GIT_WORK_TREE were set, and the current
directory was a subdirectory of the work tree,
setup_git_directory_gently() would return a bogus NULL prefix.

This patch fixes that.

Noticed by REPLeffect on IRC.

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agoCorrect typos in release notes for 1.5.3.5
Shawn O. Pearce [Wed, 17 Oct 2007 00:09:21 +0000 (20:09 -0400)]
Correct typos in release notes for 1.5.3.5

Noticed by Michele Ballabio.

Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agocvsexportcommit: get rid of temporary directory
Johannes Schindelin [Mon, 15 Oct 2007 22:32:45 +0000 (23:32 +0100)]
cvsexportcommit: get rid of temporary directory

Since commit e86ad71fe5f53ae4434566bd09ea4256090e5a3a we do not use
a temporary directory in cvsexportcommit.  So there is no need to set
one up.

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agogit-rebase: document suppression of duplicate commits
Jeff King [Mon, 15 Oct 2007 04:47:30 +0000 (00:47 -0400)]
git-rebase: document suppression of duplicate commits

git-rebase uses format-patch's --ignore-if-in-upstream
option, but we never document the user-visible behavior. The
example is placed near the top of the example list rather
than at the bottom because it is:
  a. a simple example
  b. a reasonably common scenario for many projects (mail
     some patches which get accepted upstream, then rebase)

[sp: Corrected direction of 'HEAD..<upstream>' set comparsion]

Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agoDocumentation/git-tag.txt: Document how to backdate tags
Michael Olson [Mon, 15 Oct 2007 17:51:30 +0000 (13:51 -0400)]
Documentation/git-tag.txt: Document how to backdate tags

Added a new section beneath "On Automatic following" called "On
Backdating Tags".  This includes an explanation of when to use this
method, a brief explanation of the kind of date that can be used in
GIT_AUTHOR_DATE, and an example invocation of git-tag using a custom
setting of GIT_AUTHOR_DATE.

[sp: Corrected s/you/your/, noticed by Jeff King]

Signed-off-by: Michael W. Olson <mwolson@gnu.org>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agogit-rebase--interactive.sh: Quote arguments to test
Michael W. Olson [Mon, 15 Oct 2007 17:48:27 +0000 (13:48 -0400)]
git-rebase--interactive.sh: Quote arguments to test

If /bin/sh is /bin/dash, then the script will display an error if
$parent_sha1 is undefined.  This patch works fixes the issue by
quoting both arguments to `test'.  Arguments composed solely of
variable expansions should always be quoted, unless we know for
certain that the contents are defined.

Signed-off-by: Michael W. Olson <mwolson@gnu.org>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agoMerge branch 'maint'
Shawn O. Pearce [Tue, 16 Oct 2007 02:31:47 +0000 (22:31 -0400)]
Merge branch 'maint'

* maint:
  Whip post 1.5.3.4 maintenance series into shape.
  rebase -i: use diff plumbing instead of porcelain
  Do not remove distributed configure script
  git-archive: document --exec
  git-reflog: document --verbose
  git-config: handle --file option with relative pathname properly
  clear_commit_marks(): avoid deep recursion
  git add -i: Remove unused variables
  git add -i: Fix parsing of abbreviated hunk headers
  git-config: don't silently ignore options after --list
  Clean up "git log" format with DIFF_FORMAT_NO_OUTPUT
  Fix embarrassing "git log --follow" bug

Conflicts:

RelNotes
git-rebase--interactive.sh

17 years agoWhip post 1.5.3.4 maintenance series into shape.
Shawn O. Pearce [Tue, 16 Oct 2007 02:29:50 +0000 (22:29 -0400)]
Whip post 1.5.3.4 maintenance series into shape.

Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agoMinor usage update in setgitperms.perl
Josh England [Tue, 9 Oct 2007 16:04:42 +0000 (10:04 -0600)]
Minor usage update in setgitperms.perl

Signed-off-by: Josh England <jjengla@sandia.gov>
Signed-off-by: Lars Hjemli <hjemli@gmail.com>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agomanual: use 'URL' instead of 'url'.
Ralf Wildenhues [Tue, 9 Oct 2007 21:03:55 +0000 (23:03 +0200)]
manual: use 'URL' instead of 'url'.

Just for consistency, use the spelling URL everywhere.

Signed-off-by: Lars Hjemli <hjemli@gmail.com>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agomanual: add some markup.
Ralf Wildenhues [Tue, 9 Oct 2007 21:03:11 +0000 (23:03 +0200)]
manual: add some markup.

Signed-off-by: Lars Hjemli <hjemli@gmail.com>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agomanual: Fix example finding commits referencing given content.
Ralf Wildenhues [Tue, 9 Oct 2007 21:02:19 +0000 (23:02 +0200)]
manual: Fix example finding commits referencing given content.

If I'm handed a file, then it typically lives outside the
working directory.  git-log only operates on in-tree files,
so the first 'filename' should be an in-tree one, or it should
look at all files.  This patch does the latter, so it would
also find renamed files.  However, it is also slower.

Signed-off-by: Lars Hjemli <hjemli@gmail.com>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agoFix wording in push definition.
Ralf Wildenhues [Tue, 9 Oct 2007 21:01:05 +0000 (23:01 +0200)]
Fix wording in push definition.

Make the definition of push in the glossary readable.

Signed-off-by: Lars Hjemli <hjemli@gmail.com>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agoFix some typos, punctuation, missing words, minor markup.
Ralf Wildenhues [Tue, 9 Oct 2007 21:00:03 +0000 (23:00 +0200)]
Fix some typos, punctuation, missing words, minor markup.

Signed-off-by: Lars Hjemli <hjemli@gmail.com>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agomanual: Fix or remove em dashes.
Ralf Wildenhues [Tue, 9 Oct 2007 21:05:30 +0000 (23:05 +0200)]
manual: Fix or remove em dashes.

em dashes were used inconsistently in the manual.
This changes them to the way they are used in US English.

Signed-off-by: Lars Hjemli <hjemli@gmail.com>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agoAdd a --dry-run option to git-push.
Brian Ewins [Thu, 11 Oct 2007 19:32:27 +0000 (20:32 +0100)]
Add a --dry-run option to git-push.

The default behaviour of git-push is potentially confusing
for new users, since it will push changes that are not on
the current branch. Publishing patches that were still
cooking on a development branch is hard to undo.

It would also be nice to be able to verify the expansion
of refspecs if you've edited them, so that you know
what branches matched on the server.

Adding a --dry-run flag allows the user to experiment
safely and learn how to use git-push properly. Originally
suggested by Steffen Prohaska.

Signed-off-by: Brian Ewins <brian.ewins@gmail.com>
Signed-off-by: Lars Hjemli <hjemli@gmail.com>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agoAdd a --dry-run option to git-send-pack.
Brian Ewins [Thu, 11 Oct 2007 19:32:26 +0000 (20:32 +0100)]
Add a --dry-run option to git-send-pack.

Implement support for --dry-run, so that it can be used
in calls from git-push. With this flag set, git-send-pack
will not send any updates to the server.

Signed-off-by: Brian Ewins <brian.ewins@gmail.com>
Signed-off-by: Lars Hjemli <hjemli@gmail.com>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agoFix in-place editing functions in convert.c
Pierre Habouzit [Fri, 5 Oct 2007 08:11:59 +0000 (10:11 +0200)]
Fix in-place editing functions in convert.c

* crlf_to_git and ident_to_git:

  Don't grow the buffer if there is enough space in the first place.
  As a side effect, when the editing is done "in place", we don't grow, so
  the buffer pointer doesn't changes, and `src' isn't invalidated anymore.

  Thanks to Bernt Hansen for the bug report.

* apply_filter:

  Fix memory leak due to fake in-place editing that didn't collected the
  old buffer when the filter succeeds. Also a cosmetic fix.

Signed-off-by: Pierre Habouzit <madcoder@debian.org>
Signed-off-by: Lars Hjemli <hjemli@gmail.com>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agoinstaweb: support for Ruby's WEBrick server
Mike Dalessio [Sat, 6 Oct 2007 17:29:49 +0000 (13:29 -0400)]
instaweb: support for Ruby's WEBrick server

running the webrick server with git requires Ruby and Ruby's YAML and
Webrick libraries (both of which come standard with Ruby). nice for
single-user standalone invocations.

the --httpd=webrick option generates a ruby script on the fly to read
httpd.conf options and invoke the web server via library call. this
script is placed in the .git/gitweb directory. it also generates a
shell script in a feeble attempt to invoke ruby in a portable manner,
which assumes that 'ruby' is in the user's $PATH.

Signed-off-by: Mike Dalessio <mike@csa.net>
Signed-off-by: Lars Hjemli <hjemli@gmail.com>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agoinstaweb: allow for use of auto-generated scripts
Mike Dalessio [Sat, 6 Oct 2007 17:29:48 +0000 (13:29 -0400)]
instaweb: allow for use of auto-generated scripts

this patch allows scripts that reside in $fqgitdir/gitweb to be used
for firing up an instaweb server. this lays the groundwork for
extending instaweb support to non-standard web servers, which may
require a script for proper invocation.

Signed-off-by: Mike Dalessio <mike@csa.net>
Signed-off-by: Lars Hjemli <hjemli@gmail.com>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agoAdd 'git-p4 commit' as an alias for 'git-p4 submit'
Marius Storm-Olsen [Tue, 9 Oct 2007 14:16:09 +0000 (16:16 +0200)]
Add 'git-p4 commit' as an alias for 'git-p4 submit'

Given that git uses 'commit', git-p4's 'sumbit' was a bit confusing at times;
often making me do 'git submit' and 'git-p4 commit' instead.

Signed-off-by: Marius Storm-Olsen <marius@trolltech.com>
Acked-By: Simon Hausmann <simon@lst.de>
Signed-off-by: Lars Hjemli <hjemli@gmail.com>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agohg-to-git speedup through selectable repack intervals
Michael Gebetsroither [Sat, 6 Oct 2007 21:16:51 +0000 (23:16 +0200)]
hg-to-git speedup through selectable repack intervals

Signed-off-by: Lars Hjemli <hjemli@gmail.com>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agogit-svn: respect Subversion's [auth] section configuration values
Eygene Ryabinkin [Sat, 6 Oct 2007 18:57:19 +0000 (22:57 +0400)]
git-svn: respect Subversion's [auth] section configuration values

Parameters 'store-passwords' and 'store-auth-creds' from Subversion's
configuration (~/.subversion/config) were not respected.  This was
fixed: the default values for these parameters are set to 'yes' to
follow Subversion behaviour.

Signed-off-by: Eygene Ryabinkin <rea-git@codelabs.ru>
Signed-off-by: Lars Hjemli <hjemli@gmail.com>
Acked-by: Eric Wong <normalperson@yhbt.net>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agogtksourceview2 support for gitview
Frederick Akalin [Fri, 5 Oct 2007 07:20:49 +0000 (00:20 -0700)]
gtksourceview2 support for gitview

Added support for gtksourceview2 module (pygtksourceview 1.90.x) in
gitview.  Also refactored code that creates the source buffer and view.

Signed-off-by: Frederick Akalin <akalin@akalin.cx>
Signed-off-by: Lars Hjemli <hjemli@gmail.com>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agofix contrib/hooks/post-receive-email hooks.recipients error message
Jeff Muizelaar [Thu, 11 Oct 2007 21:49:21 +0000 (17:49 -0400)]
fix contrib/hooks/post-receive-email hooks.recipients error message

Have the error message for missing recipients actually report the
missing config variable and not a fictional one.

Signed-off-by: Lars Hjemli <hjemli@gmail.com>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agoSupport cvs via git-shell
Johannes Schindelin [Tue, 9 Oct 2007 14:33:25 +0000 (15:33 +0100)]
Support cvs via git-shell

This adds cvs support to the git-shell; You can now give new users
a restricted git-shell and they still can commit via git's cvs
emulator.

Note that either the gecos information must be accurate, or you must
provide a $HOME/.gitconfig with the appropriate user credentials.
Since the git-shell is too restricted to allow the user to do it
(on purpose!), it is up to the administrator to take care of that.

Based on an idea by Jan Wielemaker.

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Lars Hjemli <hjemli@gmail.com>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agorebase -i: use diff plumbing instead of porcelain
Johannes Schindelin [Tue, 9 Oct 2007 12:59:43 +0000 (13:59 +0100)]
rebase -i: use diff plumbing instead of porcelain

When diff drivers are installed, calling "git diff <tree1>..<tree2>"
calls those drivers.  This borks the patch generation of rebase -i.
So use "git diff-tree -p" instead, which does not call diff drivers.

Noticed by Johannes Sixt.

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Lars Hjemli <hjemli@gmail.com>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agoDo not remove distributed configure script
Mathias Megyei [Thu, 4 Oct 2007 21:49:19 +0000 (23:49 +0200)]
Do not remove distributed configure script

Before this patch the clean target has removed the
configure script that comes with Git tar file.
That made compiling Git for different architectures
inconvenient.
This patch excludes configure from the files to be
deleted by 'make clean' and adds new target 'distclean'
to preserve old functionality.

Signed-off-by: Mathias Megyei <mathias@mnet-mail.de>
Signed-off-by: Lars Hjemli <hjemli@gmail.com>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agogit-archive: document --exec
Michele Ballabio [Thu, 4 Oct 2007 22:41:11 +0000 (00:41 +0200)]
git-archive: document --exec

Signed-off-by: Michele Ballabio <barra_cuda@katamail.com>
Signed-off-by: Lars Hjemli <hjemli@gmail.com>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agogit-reflog: document --verbose
Michele Ballabio [Thu, 4 Oct 2007 12:26:54 +0000 (14:26 +0200)]
git-reflog: document --verbose

Signed-off-by: Michele Ballabio <barra_cuda@katamail.com>
Signed-off-by: Lars Hjemli <hjemli@gmail.com>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agomakefile: Add a cscope target
Kristof Provost [Sat, 6 Oct 2007 14:24:42 +0000 (16:24 +0200)]
makefile: Add a cscope target

The current makefile supports ctags but not cscope. Some people prefer
cscope (I do), so this patch adds a cscope target.

I've also added cscope* to the .gitignore file. For some reason tags
and TAGS weren't in there either so I've added them too.

Signed-off-by: Kristof Provost <Kristof@provost-engineering.be>
Signed-off-by: Lars Hjemli <hjemli@gmail.com>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agogit-config: handle --file option with relative pathname properly
Gerrit Pape [Fri, 12 Oct 2007 11:32:51 +0000 (11:32 +0000)]
git-config: handle --file option with relative pathname properly

When calling git-config not from the top level directory of a repository,
it changes directory before trying to open the config file specified
through the --file option, which then fails if the config file was
specified by a relative pathname.  This patch adjusts the pathname to
the config file if applicable.

The problem was noticed by Joey Hess, reported through
 http://bugs.debian.org/445208

Signed-off-by: Gerrit Pape <pape@smarden.org>
Signed-off-by: Lars Hjemli <hjemli@gmail.com>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agoclear_commit_marks(): avoid deep recursion
Johannes Schindelin [Wed, 10 Oct 2007 22:14:35 +0000 (23:14 +0100)]
clear_commit_marks(): avoid deep recursion

Before this patch, clear_commit_marks() recursed for each parent.  This
could be potentially very expensive in terms of stack space.  Probably
the only reason that this did not lead to problems is the fact that we
typically call clear_commit_marks() after marking a relatively small set
of commits.

Use (sort of) a tail recursion instead: first recurse on the parents
other than the first one, and then continue the loop with the first
parent.

Noticed by Shawn Pearce.

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Lars Hjemli <hjemli@gmail.com>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agogit add -i: Remove unused variables
Jean-Luc Herren [Tue, 9 Oct 2007 19:34:17 +0000 (21:34 +0200)]
git add -i: Remove unused variables

Signed-off-by: Jean-Luc Herren <jlh@gmx.ch>
Signed-off-by: Lars Hjemli <hjemli@gmail.com>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agogit add -i: Fix parsing of abbreviated hunk headers
Jean-Luc Herren [Tue, 9 Oct 2007 19:29:26 +0000 (21:29 +0200)]
git add -i: Fix parsing of abbreviated hunk headers

The unified diff format allows one-line ranges to be abbreviated
by omiting the size.  The hunk header "@@ -10,1 +10,1 @@" can be
expressed as "@@ -10 +10 @@", but this wasn't properly parsed in
all cases.

Such abbreviated hunk headers are generated when a one-line change
(add, remove or modify) appears without context; for example
because the file is a one-liner itself or because GIT_DIFF_OPTS
was set to '-u0'.  If the user then runs 'git add -i' and enters
the 'patch' command for that file, perl complains about undefined
variables.

Signed-off-by: Jean-Luc Herren <jlh@gmx.ch>
Signed-off-by: Lars Hjemli <hjemli@gmail.com>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agogit-config: don't silently ignore options after --list
Frank Lichtenheld [Fri, 5 Oct 2007 20:16:44 +0000 (22:16 +0200)]
git-config: don't silently ignore options after --list

Error out if someone gives options after --list since that is
not a valid syntax.

Signed-off-by: Frank Lichtenheld <frank@lichtenheld.de>
Signed-off-by: Lars Hjemli <hjemli@gmail.com>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agogit-gc --auto: simplify "repack" command line building
Brandon Casey [Thu, 11 Oct 2007 01:00:27 +0000 (20:00 -0500)]
git-gc --auto: simplify "repack" command line building

Since "-a" is removed from the base repack command line,
we can simplify how we add additional options to this
command line when using --auto.

Signed-off-by: Brandon Casey <casey@nrlssc.navy.mil>
Signed-off-by: Lars Hjemli <hjemli@gmail.com>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agogit-gc: by default use safer "-A" option to repack when not --prune'ing
Brandon Casey [Thu, 11 Oct 2007 00:59:19 +0000 (19:59 -0500)]
git-gc: by default use safer "-A" option to repack when not --prune'ing

This makes use of repack's new "-A" option which does not drop packed
unreachable objects. This makes git-gc safe to call at any time,
particularly when a repository is referenced as an alternate by
another repository.

git-gc --prune will use the "-a" option to repack instead of "-A", so
that packed unreachable objects will be removed.

Signed-off-by: Brandon Casey <casey@nrlssc.navy.mil>
Signed-off-by: Lars Hjemli <hjemli@gmail.com>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agofix t5403-post-checkout-hook.sh: built-in test in dash does not have "=="
Alex Riesen [Thu, 4 Oct 2007 13:32:53 +0000 (15:32 +0200)]
fix t5403-post-checkout-hook.sh: built-in test in dash does not have "=="

Signed-off-by: Alex Riesen <raa.lkml@gmail.com>
Signed-off-by: Lars Hjemli <hjemli@gmail.com>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agoClean up "git log" format with DIFF_FORMAT_NO_OUTPUT
Linus Torvalds [Tue, 9 Oct 2007 16:35:22 +0000 (09:35 -0700)]
Clean up "git log" format with DIFF_FORMAT_NO_OUTPUT

This fixes an unnecessary empty line that we add to the log message when
we generate diffs, but don't actually end up printing any due to having
DIFF_FORMAT_NO_OUTPUT set.

This can happen with pickaxe or with rename following. The reason is that
we normally add an empty line between the commit and the diff, but we do
that even for the case where we've then suppressed the actual printing of
the diff.

This also updates a couple of tests that assumed the extraneous empty
line would exist at the end of output.

Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Lars Hjemli <hjemli@gmail.com>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agoFix embarrassing "git log --follow" bug
Linus Torvalds [Mon, 8 Oct 2007 20:42:41 +0000 (13:42 -0700)]
Fix embarrassing "git log --follow" bug

It turns out that I completely broke "git log --follow" with my recent
patch to revision.c ("Fix revision log diff setup, avoid unnecessary diff
generation", commit b7bb760d5ed4881422673d32f869d140221d3564).

Why? Because --follow obviously requires the diff machinery to function,
exactly the same way pickaxe does.

So everybody is away right now, but considering that nobody even noticed
this bug, I don't think it matters. But for the record, here's the trivial
one-liner fix (well, two, since I also fixed the comment).

Because of the nature of the bug, if you ask for patches when following
(which is one of the things I normally do), the bug is hidden, because
then the request for diff output will automatically also enable the diffs
themselves.

So while "git log --follow <filename>" didn't work, adding a "-p"
magically made it work again even without this fix.

Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Lars Hjemli <hjemli@gmail.com>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agoUpdate state documentation link for 1.5.3.4
Junio C Hamano [Thu, 4 Oct 2007 08:35:01 +0000 (01:35 -0700)]
Update state documentation link for 1.5.3.4

Signed-off-by: Junio C Hamano <gitster@pobox.com>
17 years agoDon't checkout the full tree if avoidable
Steven Walter [Fri, 28 Sep 2007 17:24:19 +0000 (13:24 -0400)]
Don't checkout the full tree if avoidable

In most cases of branching, the tree is copied unmodified from the trunk
to the branch.  When that is done, we can simply start with the parent's
index and apply the changes on the branch as usual.

[ew: rewritten from Steven's original to use SVN::Client instead
     of the command-line svn client.

     Since SVN::Client connects separately, we'll share our
     authentication providers array between our usages of
     SVN::Client and SVN::Ra, too.  Bypassing the high-level
     SVN::Client library can avoid this, but the code will be
     much more complex.  Regardless, any implementation of this
     seems to require restarting a connection to the remote
     server.

     Also of note is that SVN 1.4 and later allows a more
     efficient diff_summary to be done instead of a full diff,
     but since this code is only to support SVN < 1.4.4, we'll
     ignore it for now.]

Signed-off-by: Steven Walter <stevenrwalter@gmail.com>
Signed-off-by: Eric Wong <normalperson@yhbt.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
17 years agobuiltin-apply: fix conversion error in strbuf series
Junio C Hamano [Thu, 4 Oct 2007 00:42:52 +0000 (17:42 -0700)]
builtin-apply: fix conversion error in strbuf series

Signed-off-by: Junio C Hamano <gitster@pobox.com>
17 years agoMerge branch 'mv/unknown'
Junio C Hamano [Wed, 3 Oct 2007 11:28:24 +0000 (04:28 -0700)]
Merge branch 'mv/unknown'

* mv/unknown:
  Don't use "<unknown>" for placeholders and suppress printing of empty user formats.

17 years agoMerge branch 'ph/strbuf'
Junio C Hamano [Wed, 3 Oct 2007 10:06:02 +0000 (03:06 -0700)]
Merge branch 'ph/strbuf'

* ph/strbuf: (44 commits)
  Make read_patch_file work on a strbuf.
  strbuf_read_file enhancement, and use it.
  strbuf change: be sure ->buf is never ever NULL.
  double free in builtin-update-index.c
  Clean up stripspace a bit, use strbuf even more.
  Add strbuf_read_file().
  rerere: Fix use of an empty strbuf.buf
  Small cache_tree_write refactor.
  Make builtin-rerere use of strbuf nicer and more efficient.
  Add strbuf_cmp.
  strbuf_setlen(): do not barf on setting length of an empty buffer to 0
  sq_quote_argv and add_to_string rework with strbuf's.
  Full rework of quote_c_style and write_name_quoted.
  Rework unquote_c_style to work on a strbuf.
  strbuf API additions and enhancements.
  nfv?asprintf are broken without va_copy, workaround them.
  Fix the expansion pattern of the pseudo-static path buffer.
  builtin-for-each-ref.c::copy_name() - do not overstep the buffer.
  builtin-apply.c: fix a tiny leak introduced during xmemdupz() conversion.
  Use xmemdupz() in many places.
  ...

17 years agoMerge branch 'lh/merge'
Junio C Hamano [Wed, 3 Oct 2007 10:05:58 +0000 (03:05 -0700)]
Merge branch 'lh/merge'

* lh/merge:
  git-merge: add --ff and --no-ff options
  git-merge: add support for --commit and --no-squash
  git-merge: add support for branch.<name>.mergeoptions
  git-merge: refactor option parsing
  git-merge: fix faulty SQUASH_MSG
  Add test-script for git-merge porcelain

17 years agoMerge branch 'js/rebase-i'
Junio C Hamano [Wed, 3 Oct 2007 10:05:39 +0000 (03:05 -0700)]
Merge branch 'js/rebase-i'

* js/rebase-i:
  rebase -i: work on a detached HEAD

17 years agoMerge branch 'jc/autogc'
Junio C Hamano [Wed, 3 Oct 2007 10:05:32 +0000 (03:05 -0700)]
Merge branch 'jc/autogc'

* jc/autogc:
  git-gc --auto: run "repack -A -d -l" as necessary.
  git-gc --auto: restructure the way "repack" command line is built.
  git-gc --auto: protect ourselves from accumulated cruft
  git-gc --auto: add documentation.
  git-gc --auto: move threshold check to need_to_gc() function.
  repack -A -d: use --keep-unreachable when repacking
  pack-objects --keep-unreachable
  Export matches_pack_name() and fix its return value
  Invoke "git gc --auto" from commit, merge, am and rebase.
  Implement git gc --auto

17 years agoMerge branch 'je/hooks'
Junio C Hamano [Wed, 3 Oct 2007 10:03:40 +0000 (03:03 -0700)]
Merge branch 'je/hooks'

* je/hooks:
  post-checkout hook, tests, and docs

17 years agoMerge branch 'ap/dateformat'
Junio C Hamano [Wed, 3 Oct 2007 10:03:22 +0000 (03:03 -0700)]
Merge branch 'ap/dateformat'

* ap/dateformat:
  Add a test script for for-each-ref, including test of date formatting
  dateformat: parse %(xxdate) %(yydate:format) correctly
  Make for-each-ref's grab_date() support per-atom formatting
  Make for-each-ref allow atom names like "<name>:<something>"
  parse_date_format(): convert a format name to an enum date_mode

17 years agoSync with GIT 1.5.3.4
Junio C Hamano [Wed, 3 Oct 2007 10:03:15 +0000 (03:03 -0700)]
Sync with GIT 1.5.3.4

Signed-off-by: Junio C Hamano <gitster@pobox.com>
17 years agoGIT 1.5.3.4 v1.5.3.4
Junio C Hamano [Wed, 3 Oct 2007 09:33:11 +0000 (02:33 -0700)]
GIT 1.5.3.4

Signed-off-by: Junio C Hamano <gitster@pobox.com>
17 years agoAdd a test script for for-each-ref, including test of date formatting
Andy Parkins [Tue, 2 Oct 2007 11:02:57 +0000 (12:02 +0100)]
Add a test script for for-each-ref, including test of date formatting

Signed-off-by: Andy Parkins <andyparkins@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
17 years agoAdd test case for ls-files --with-tree
Carl Worth [Wed, 3 Oct 2007 07:03:53 +0000 (00:03 -0700)]
Add test case for ls-files --with-tree

This tests basic functionality and also exercises a bug noticed
by Keith Packard, (prune_cache followed by add_index_entry can
trigger an attempt to realloc a pointer into the middle of an
allocated buffer).

Signed-off-by: Junio C Hamano <gitster@pobox.com>
17 years agoMust not modify the_index.cache as it may be passed to realloc at some point.
Keith Packard [Wed, 3 Oct 2007 05:44:15 +0000 (22:44 -0700)]
Must not modify the_index.cache as it may be passed to realloc at some point.

The index cache is not static, growing as new entries are added. If
entries are added after prune_cache is called, cache will no longer
point at the base of the allocation, and realloc will not be happy.

I verified that this was the only place in the current source which
modified any index_state.cache elements aside from the alloc/realloc
calls in read-cache by changing the type of the element to 'struct
cache_entry ** const cache' and recompiling.

A more efficient patch would create a separate 'cache_base' value to
track the allocation and then fix things up when reallocation was
necessary, instead of the brute-force memmove used here.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
17 years agothe ar tool is called gar on some systems
Robert Schiele [Wed, 3 Oct 2007 01:49:34 +0000 (03:49 +0200)]
the ar tool is called gar on some systems

Some systems that have only installed the GNU toolchain (prefixed with "g")
do not provide "ar" but only "gar".  Make configure find this tool as well.

Signed-off-by: Robert Schiele <rschiele@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
17 years agorename diff_free_filespec_data_large() to diff_free_filespec_blob()
Junio C Hamano [Wed, 3 Oct 2007 04:01:03 +0000 (21:01 -0700)]
rename diff_free_filespec_data_large() to diff_free_filespec_blob()

Signed-off-by: Junio C Hamano <gitster@pobox.com>
17 years agodiffcore-rename: cache file deltas
Jeff King [Tue, 25 Sep 2007 19:29:42 +0000 (15:29 -0400)]
diffcore-rename: cache file deltas

We find rename candidates by computing a fingerprint hash of
each file, and then comparing those fingerprints. There are
inherently O(n^2) comparisons, so it pays in CPU time to
hoist the (rather expensive) computation of the fingerprint
out of that loop (or to cache it once we have computed it once).

Previously, we didn't keep the filespec information around
because then we had the potential to consume a great deal of
memory. However, instead of keeping all of the filespec
data, we can instead just keep the fingerprint.

This patch implements and uses diff_free_filespec_data_large
to accomplish that goal. We also have to change
estimate_similarity not to needlessly repopulate the
filespec data when we already have the hash.

Practical tests showed 4.5x speedup for a 10% memory usage
increase.

Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
17 years agoMention 'cpio' dependency in INSTALL
Johan Herland [Tue, 2 Oct 2007 23:42:29 +0000 (01:42 +0200)]
Mention 'cpio' dependency in INSTALL

Signed-off-by: Johan Herland <johan@herland.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
17 years agoMake git-pull complain and give advice when there is nothing to merge with
Federico Mena Quintero [Tue, 2 Oct 2007 23:36:30 +0000 (18:36 -0500)]
Make git-pull complain and give advice when there is nothing to merge with

Signed-off-by: Federico Mena Quintero <federico@gnu.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
17 years agoNote that git-branch will not automatically checkout the new branch
Federico Mena Quintero [Tue, 2 Oct 2007 23:34:32 +0000 (18:34 -0500)]
Note that git-branch will not automatically checkout the new branch

Signed-off-by: Federico Mena Quintero <federico@gnu.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
17 years agoAdd documentation for --track and --no-track to the git-branch docs.
Federico Mena Quintero [Tue, 2 Oct 2007 23:33:30 +0000 (18:33 -0500)]
Add documentation for --track and --no-track to the git-branch docs.

Signed-off-by: Federico Mena Quintero <federico@gnu.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
17 years agoSay when --track is useful in the git-checkout docs.
Federico Mena Quintero [Tue, 2 Oct 2007 23:32:32 +0000 (18:32 -0500)]
Say when --track is useful in the git-checkout docs.

The documentation used to say what the option does, but it
didn't mention a use case.

Signed-off-by: Federico Mena Quintero <federico@gnu.org>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
17 years agoFix typo in config.txt
Johannes Schindelin [Tue, 2 Oct 2007 20:14:30 +0000 (21:14 +0100)]
Fix typo in config.txt

There was an 'l' (ell) instead of a '1' (one) in one of the gitlinks.

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
17 years agofor-each-ref: fix %(numparent) and %(parent)
Junio C Hamano [Tue, 2 Oct 2007 22:09:41 +0000 (15:09 -0700)]
for-each-ref: fix %(numparent) and %(parent)

The string value of %(numparent) was not returned correctly.
Also %(parent) misbehaved for the root commits (returned garbage)
and merge commits (returned first parent, followed by a space).

Signed-off-by: Junio C Hamano <gitster@pobox.com>
17 years agodateformat: parse %(xxdate) %(yydate:format) correctly
Junio C Hamano [Tue, 2 Oct 2007 21:31:37 +0000 (14:31 -0700)]
dateformat: parse %(xxdate) %(yydate:format) correctly

Andy Parkins noticed that parsing of the above would not
correctly notice that xxdate does not have any format
specifier.

Signed-off-by: Junio C Hamano <gitster@pobox.com>