Code

git.git
17 years agoMake sure test-genrandom and test-chmtime are builtas part of the main build.
Junio C Hamano [Sat, 28 Apr 2007 22:32:49 +0000 (15:32 -0700)]
Make sure test-genrandom and test-chmtime are builtas part of the main build.

Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoMake macros to prevent double-inclusion in headers consistent.
Junio C Hamano [Sun, 29 Apr 2007 06:38:52 +0000 (23:38 -0700)]
Make macros to prevent double-inclusion in headers consistent.

Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoApply mailmap in git-blame output.
Junio C Hamano [Fri, 27 Apr 2007 07:42:15 +0000 (00:42 -0700)]
Apply mailmap in git-blame output.

This makes git-blame to use the same mailmap used by
git-shortlog.

Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoSplit out mailmap handling out of shortlog
Junio C Hamano [Fri, 27 Apr 2007 07:41:15 +0000 (00:41 -0700)]
Split out mailmap handling out of shortlog

This splits out a few functions to deal with mailmap from
shortlog and makes it a bit more usable from other programs.
Most notably, it does not clobber input e-mail address anymore.

Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoblame -s: suppress author name and time.
Junio C Hamano [Thu, 12 Apr 2007 22:50:45 +0000 (15:50 -0700)]
blame -s: suppress author name and time.

With this "git blame -b -s HEAD~n..HEAD" becomes a nicer way to
review the result of recent changes in context.

Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoFall back to $EMAIL for missing GIT_AUTHOR_EMAIL and GIT_COMMITTER_EMAIL
Josh Triplett [Sun, 29 Apr 2007 01:40:28 +0000 (18:40 -0700)]
Fall back to $EMAIL for missing GIT_AUTHOR_EMAIL and GIT_COMMITTER_EMAIL

Some other programs get the user's email address from $EMAIL, so fall back to
that if we don't have a Git-specific email address.

Signed-off-by: Josh Triplett <josh@freedesktop.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoMerge commit 'gfi/master'
Junio C Hamano [Sun, 29 Apr 2007 08:54:28 +0000 (01:54 -0700)]
Merge commit 'gfi/master'

* commit 'gfi/master':

17 years agoMerge branch 'maint'
Junio C Hamano [Sun, 29 Apr 2007 08:52:43 +0000 (01:52 -0700)]
Merge branch 'maint'

* maint:
  http.c: Fix problem with repeated calls of http_init
  Add missing reference to GIT_COMMITTER_DATE in git-commit-tree documentation
  Fix import-tars fix.
  Update .mailmap with "Michael"
  Do not barf on too long action description
  Catch empty pathnames in trees during fsck
  Don't allow empty pathnames in fast-import
  import-tars: be nice to wrong directory modes
  git-svn: Added 'find-rev' command
  git shortlog documentation: add long options and fix a typo

17 years agohttp.c: Fix problem with repeated calls of http_init
Julian Phillips [Sun, 29 Apr 2007 02:46:42 +0000 (03:46 +0100)]
http.c: Fix problem with repeated calls of http_init

Calling http_init after calling http_cleanup causes a segfault.  This
is due to the pragma_header curl_slist being freed but not being set
to NULL.  The subsequent call to http_init tries to setup the slist
again, but it now points to an invalid memory location.

Signed-off-by: Julian Phillips <julian@quantumfyre.co.uk>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoAdd missing reference to GIT_COMMITTER_DATE in git-commit-tree documentation
Josh Triplett [Sun, 29 Apr 2007 01:40:12 +0000 (18:40 -0700)]
Add missing reference to GIT_COMMITTER_DATE in git-commit-tree documentation

Signed-off-by: Josh Triplett <josh@freedesktop.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoFix import-tars fix.
Junio C Hamano [Sun, 29 Apr 2007 07:31:14 +0000 (00:31 -0700)]
Fix import-tars fix.

This heeds advice from our resident Perl expert to make sure
the script is not confused with a string that ends with /\n

Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoUpdate .mailmap with "Michael"
Junio C Hamano [Sun, 29 Apr 2007 07:22:00 +0000 (00:22 -0700)]
Update .mailmap with "Michael"

Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoDo not barf on too long action description
Junio C Hamano [Sat, 28 Apr 2007 07:15:48 +0000 (00:15 -0700)]
Do not barf on too long action description

Reflog message is primarily about easier identification, and
leaving truncated entry is much better than dying.

Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoMerge branch 'maint' of git://repo.or.cz/git/fastimport into maint
Junio C Hamano [Sun, 29 Apr 2007 01:15:00 +0000 (18:15 -0700)]
Merge branch 'maint' of git://repo.or.cz/git/fastimport into maint

* 'maint' of git://repo.or.cz/git/fastimport:
  Don't allow empty pathnames in fast-import
  import-tars: be nice to wrong directory modes

17 years agoCatch empty pathnames in trees during fsck
Shawn O. Pearce [Sun, 29 Apr 2007 00:29:23 +0000 (20:29 -0400)]
Catch empty pathnames in trees during fsck

Released versions of fast-import have been able to create a tree that
contains files or subtrees that contain no name.  Unfortunately these
trees aren't valid, but people may have actually tried to create them
due to bugs in import-tars.perl or their own fast-import frontend.

We now look for this unusual condition and warn the user if at
least one of their tree objects contains the problem.

Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoMerge branch 'gfi-maint' into gfi-master
Shawn O. Pearce [Sun, 29 Apr 2007 00:05:58 +0000 (20:05 -0400)]
Merge branch 'gfi-maint' into gfi-master

* gfi-maint:
  Don't allow empty pathnames in fast-import
  import-tars: be nice to wrong directory modes

17 years agoMerge commit 'jc/maint' into gfi-maint
Shawn O. Pearce [Sun, 29 Apr 2007 00:05:20 +0000 (20:05 -0400)]
Merge commit 'jc/maint' into gfi-maint

* commit 'jc/maint': (35 commits)
  Update git-http-fetch documentation
  Update git-local-fetch documentation
  Update git-http-push documentation
  Update -L documentation for git-blame/git-annotate
  Update git-grep documentation
  Update git-fmt-merge documentation
  Document additional options for git-fetch
  Removing -n option from git-diff-files documentation
  Start preparing for 1.5.1.3
  Sanitize @to recipients.
  git-svn: Ignore usernames in URLs in find_by_url
  Document --dry-run and envelope-sender for git-send-email.
  Allow users to optionally specify their envelope sender.
  Ensure clean addresses are always used with Net::SMTP
  Validate @recipients before using it for sendmail and Net::SMTP.
  Perform correct quoting of recipient names.
  Change the scope of the $cc variable as it is not needed outside of send_message.
  Debugging cleanup improvements
  Prefix Dry- to the message status to denote dry-runs.
  Document --dry-run parameter to send-email.
  ...

17 years agoDon't allow empty pathnames in fast-import
Shawn O. Pearce [Sun, 29 Apr 2007 00:01:27 +0000 (20:01 -0400)]
Don't allow empty pathnames in fast-import

riddochc on #git noticed corruption caused by import-tars.  This
was fixed in the prior commit by Dscho, but fast-import was wrong
to have allowed a tree to be created with an empty string as the
filename.  No operating system allows this, and Git itself doesn't
accept this into the index.

Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agoimport-tars: be nice to wrong directory modes
Johannes Schindelin [Sat, 28 Apr 2007 23:59:47 +0000 (01:59 +0200)]
import-tars: be nice to wrong directory modes

Some tars seem to have modes 0755 for directories, not 01000755. Do
not generate an empty object for them, but ignore them.

Noticed by riddochc on IRC.

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agogit-svn: Added 'find-rev' command
Adam Roben [Fri, 27 Apr 2007 18:57:53 +0000 (11:57 -0700)]
git-svn: Added 'find-rev' command

This patch adds a new 'find-rev' command to git-svn that lets you easily
translate between SVN revision numbers and git tree-ish.

Signed-off-by: Adam Roben <aroben@apple.com>
Acked-by: Eric Wong <normalperson@yhbt.net>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agogit shortlog documentation: add long options and fix a typo
Michele Ballabio [Fri, 27 Apr 2007 19:56:47 +0000 (21:56 +0200)]
git shortlog documentation: add long options and fix a typo

Signed-off-by: Michele Ballabio <barra_cuda@katamail.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoMerge branch 'maint'
Junio C Hamano [Fri, 27 Apr 2007 06:29:09 +0000 (23:29 -0700)]
Merge branch 'maint'

* maint:
  Update git-http-fetch documentation
  Update git-local-fetch documentation
  Update git-http-push documentation
  Update -L documentation for git-blame/git-annotate
  Update git-grep documentation
  Update git-fmt-merge documentation
  Document additional options for git-fetch
  Removing -n option from git-diff-files documentation

17 years agoUpdate git-http-fetch documentation
Andrew Ruder [Fri, 27 Apr 2007 04:59:02 +0000 (23:59 -0500)]
Update git-http-fetch documentation

Documentation/git-http-fetch.txt: --recover to resume a failed fetch
operation.

Signed-off-by: Andrew Ruder <andy@aeruder.net>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoUpdate git-local-fetch documentation
Andrew Ruder [Fri, 27 Apr 2007 04:59:01 +0000 (23:59 -0500)]
Update git-local-fetch documentation

Documentation/git-local-fetch.txt: -s to use
symbolic links instead of file-to-file copy, -l
to use hardlinks, -n to never use file-to-file
copies, --recover to resume a failed fetch.

Signed-off-by: Andrew Ruder <andy@aeruder.net>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoUpdate git-http-push documentation
Andrew Ruder [Fri, 27 Apr 2007 04:59:00 +0000 (23:59 -0500)]
Update git-http-push documentation

Documentation/git-http-push.txt: Changing --complete to --all.  Added
documentation for -d and -D to remote remote refs.

Signed-off-by: Andrew Ruder <andy@aeruder.net>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoUpdate -L documentation for git-blame/git-annotate
Andrew Ruder [Fri, 27 Apr 2007 04:58:59 +0000 (23:58 -0500)]
Update -L documentation for git-blame/git-annotate

Documenting alternate ways to use -L:

-L /regex/,end
-L start,+offset

Signed-off-by: Andrew Ruder <andy@aeruder.net>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoUpdate git-grep documentation
Andrew Ruder [Fri, 27 Apr 2007 04:58:58 +0000 (23:58 -0500)]
Update git-grep documentation

Documentation/git-grep.txt: Document -F/--fixed-strings to
search for non-regexp patterns.  Document -I to not search
binary files.  Document -<num> as a shortcut for -C<num>.

Signed-off-by: Andrew Ruder <andy@aeruder.net>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoUpdate git-fmt-merge documentation
Andrew Ruder [Fri, 27 Apr 2007 04:58:57 +0000 (23:58 -0500)]
Update git-fmt-merge documentation

Documentation/git-fmt-merge-msg.txt:
--summary to list commit summaries on merge
--no-summary
--file to take merged objects from a file.
Configuration option merge.summary

Signed-off-by: Andrew Ruder <andy@aeruder.net>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoDocument additional options for git-fetch
Andrew Ruder [Fri, 27 Apr 2007 04:58:56 +0000 (23:58 -0500)]
Document additional options for git-fetch

Document --quiet/-q and --verbose/-v
Add -n as an alternate for --no-tags
Fix some whitespace issues

Signed-off-by: Andrew Ruder <andy@aeruder.net>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoRemoving -n option from git-diff-files documentation
Andrew Ruder [Fri, 27 Apr 2007 04:58:55 +0000 (23:58 -0500)]
Removing -n option from git-diff-files documentation

-n is not a short form of --no-index as the documentation
suggests.  Removing it from the documentation and command
usage string.

Signed-off-by: Andrew Ruder <andy@aeruder.net>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agopost-receive-email example hook: sed command for getting description was wrong
Andy Parkins [Thu, 26 Apr 2007 21:37:16 +0000 (22:37 +0100)]
post-receive-email example hook: sed command for getting description was wrong

The sed command that extracted the first line of the project description
didn't include the -n switch and hence the project name was being
printed twice.  This was ruining the email header generation because it
was assumed that the description was only one line and was included in
the subject.  This turned the subject into a two line item and
prematurely finished the header.

Signed-off-by: Andy Parkins <andyparkins@gmail.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agopost-receive-email example hook: detect rewind-only updates and output sensible message
Andy Parkins [Thu, 26 Apr 2007 21:36:24 +0000 (22:36 +0100)]
post-receive-email example hook: detect rewind-only updates and output sensible message

Sometimes a non-fast-forward update doesn't add new commits, it merely
removes old commits.  This patch adds support for detecting that and
outputting a more correct message.

Signed-off-by: Andy Parkins <andyparkins@gmail.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agopost-receive-email example hook: fastforward should have been fast_forward
Andy Parkins [Thu, 26 Apr 2007 21:35:39 +0000 (22:35 +0100)]
post-receive-email example hook: fastforward should have been fast_forward

Signed-off-by: Andy Parkins <andyparkins@gmail.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoIgnore merged status of the file-level merge
Alex Riesen [Thu, 26 Apr 2007 19:13:49 +0000 (21:13 +0200)]
Ignore merged status of the file-level merge

as it is not relevant for whether the result should be written.
Even if no real merge happened, there might be _no_ reason to
rewrite the working tree file. Maybe even more so.

Signed-off-by: Alex Riesen <raa.lkml@gmail.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoAdd a test for merging changed and rename-changed branches
Alex Riesen [Wed, 25 Apr 2007 20:07:45 +0000 (22:07 +0200)]
Add a test for merging changed and rename-changed branches

Also leave a warning for future merge-recursive explorers.

Signed-off-by: Alex Riesen <raa.lkml@gmail.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoAvoid excessive rewrites in merge-recursive
Alex Riesen [Wed, 25 Apr 2007 20:06:59 +0000 (22:06 +0200)]
Avoid excessive rewrites in merge-recursive

If a file is changed in one branch, and renamed and changed to the
same content in another branch than we can skip the rewrite of this
file in the working directory, as the content does not change.

Signed-off-by: Alex Riesen <raa.lkml@gmail.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoMerge branch 'maint'
Junio C Hamano [Thu, 26 Apr 2007 06:31:45 +0000 (23:31 -0700)]
Merge branch 'maint'

* maint:
  Start preparing for 1.5.1.3
  Sanitize @to recipients.
  git-svn: Ignore usernames in URLs in find_by_url
  Document --dry-run and envelope-sender for git-send-email.
  Allow users to optionally specify their envelope sender.
  Ensure clean addresses are always used with Net::SMTP
  Validate @recipients before using it for sendmail and Net::SMTP.
  Perform correct quoting of recipient names.
  Change the scope of the $cc variable as it is not needed outside of send_message.
  Debugging cleanup improvements
  Prefix Dry- to the message status to denote dry-runs.
  Document --dry-run parameter to send-email.
  git-svn: Don't rely on $_ after making a function call
  Fix handle leak in write_tree
  Actually handle some-low memory conditions

Conflicts:

RelNotes
git-send-email.perl

17 years agoStart preparing for 1.5.1.3
Junio C Hamano [Thu, 26 Apr 2007 06:27:07 +0000 (23:27 -0700)]
Start preparing for 1.5.1.3

Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoSanitize @to recipients.
Robin H. Johnson [Thu, 26 Apr 2007 04:53:22 +0000 (21:53 -0700)]
Sanitize @to recipients.

We need to sanitize @to as well to ensure that names are properly quoted.

Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agogit-svn: Ignore usernames in URLs in find_by_url
Adam Roben [Wed, 25 Apr 2007 19:42:58 +0000 (12:42 -0700)]
git-svn: Ignore usernames in URLs in find_by_url

Usernames don't matter for the purposes of find_by_url, so always remove them
before doing any comparisons.

Signed-off-by: Adam Roben <aroben@apple.com>
Acked-by: Eric Wong <normalperson@yhbt.net>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoAdd --date={local,relative,default}
Junio C Hamano [Wed, 25 Apr 2007 06:36:22 +0000 (23:36 -0700)]
Add --date={local,relative,default}

This adds --date={local,relative,default} option to log family of commands,
to allow displaying timestamps in user's local timezone, relative time, or
the default format.

Existing --relative-date option is a synonym of --date=relative; we could
probably deprecate it in the long run.

Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoDocument --dry-run and envelope-sender for git-send-email.
Robin H. Johnson [Thu, 26 Apr 2007 02:37:23 +0000 (19:37 -0700)]
Document --dry-run and envelope-sender for git-send-email.

Catch the documentation up with the rest of this patchset.

Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoAllow users to optionally specify their envelope sender.
Robin H. Johnson [Thu, 26 Apr 2007 02:37:22 +0000 (19:37 -0700)]
Allow users to optionally specify their envelope sender.

If your normal user is not the same user you are subscribed to a list with,
then the default envelope sender used will cause your messages to bounce or
silently vanish into the ether.

This patch provides an optional parameter to set the envelope sender.
To use it with the sendmail binary, you must have privileges to use the -f
parameter!

Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoEnsure clean addresses are always used with Net::SMTP
Robin H. Johnson [Thu, 26 Apr 2007 02:37:21 +0000 (19:37 -0700)]
Ensure clean addresses are always used with Net::SMTP

Always pass in clean addresses to Net::SMTP for the MAIL FROM, and use them on
the SMTP non-quiet output as well.

Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoValidate @recipients before using it for sendmail and Net::SMTP.
Robin H. Johnson [Thu, 26 Apr 2007 02:37:20 +0000 (19:37 -0700)]
Validate @recipients before using it for sendmail and Net::SMTP.

Ensure that @recipients is only raw addresses when it is handed to the sendmail
binary OR Net::SMTP, otherwise BCC cases might get an extra <, or wierd stuff
might be passed to the exec.

Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoPerform correct quoting of recipient names.
Robin H. Johnson [Thu, 26 Apr 2007 02:37:19 +0000 (19:37 -0700)]
Perform correct quoting of recipient names.

Always perform quoting of the recipient names if they contain periods,
previously only the author's address was treated this way. This stops sendmail
binaries from exploding the name into bad addresses.

Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoChange the scope of the $cc variable as it is not needed outside of send_message.
Robin H. Johnson [Thu, 26 Apr 2007 02:37:18 +0000 (19:37 -0700)]
Change the scope of the $cc variable as it is not needed outside of send_message.

$cc is only used inside the send_message scope, so lets clean it out of the global scope.

Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoDebugging cleanup improvements
Robin H. Johnson [Thu, 26 Apr 2007 02:37:17 +0000 (19:37 -0700)]
Debugging cleanup improvements

The debug output is much more helpful if it has the parameters that were used.
Pull the sendmail parameters into a seperate array for that, and also include
similar data during the Net::SMTP case.

Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoPrefix Dry- to the message status to denote dry-runs.
Robin H. Johnson [Thu, 26 Apr 2007 02:37:16 +0000 (19:37 -0700)]
Prefix Dry- to the message status to denote dry-runs.

While doing testing, it's useful to see that a dry run was actually done,
instead of a real one.

Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoDocument --dry-run parameter to send-email.
Robin H. Johnson [Thu, 26 Apr 2007 02:37:15 +0000 (19:37 -0700)]
Document --dry-run parameter to send-email.

Looks like --dry-run was added to the code, but never to the --help output.

Signed-off-by: Robin H. Johnson <robbat2@gentoo.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agogit-svn: Don't rely on $_ after making a function call
Adam Roben [Wed, 25 Apr 2007 18:50:32 +0000 (11:50 -0700)]
git-svn: Don't rely on $_ after making a function call

Many functions and operators in perl set $_, so its value cannot be relied upon
after calling arbitrary functions. The solution is simply to copy the value of
$_ into a local variable that will not get overwritten.

Signed-off-by: Adam Roben <aroben@apple.com>
Acked-by: Eric Wong <normalperson@yhbt.net>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoFix handle leak in write_tree
Alex Riesen [Wed, 25 Apr 2007 22:28:17 +0000 (00:28 +0200)]
Fix handle leak in write_tree

This is a quick and dirty fix for the broken "git cherry-pick -n" on
some broken OS, which does not remove the directory entry after unlink
succeeded(!) if the file is still open somewher.
The entry is left but "protected": no open, no unlink, no stat.
Very annoying.

Signed-off-by: Alex Riesen <raa.lkml@gmail.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoActually handle some-low memory conditions
Shawn O. Pearce [Wed, 25 Apr 2007 08:02:27 +0000 (04:02 -0400)]
Actually handle some-low memory conditions

Tim Ansell discovered his Debian server didn't permit git-daemon to
use as much memory as it needed to handle cloning a project with
a 128 MiB packfile.  Filtering the strace provided by Tim of the
rev-list child showed this gem of a sequence:

  open("./objects/pack/pack-*.pack", O_RDONLY|O_LARGEFILE <unfinished ...>
  <... open resumed> )              = 5

OK, so the packfile is fd 5...

  mmap2(NULL, 33554432, PROT_READ, MAP_PRIVATE, 5, 0 <unfinished ...>
   <... mmap2 resumed> )             = 0xb5e2d000

and we mapped one 32 MiB window from it at position 0...

   mmap2(NULL, 31020635, PROT_READ, MAP_PRIVATE, 5, 0x6000 <unfinished ...>
   <... mmap2 resumed> )             = -1 ENOMEM (Cannot allocate memory)

And we asked for another window further into the file.  But got
denied.  In Tim's case this was due to a resource limit on the
git-daemon process, and its children.

Now where are we in the code?  We're down inside use_pack(),
after we have called unuse_one_window() enough times to make sure
we stay within our allowed maximum window size.  However since we
didn't unmap the prior window at 0xb5e2d000 we aren't exceeding
the current limit (which probably was just the defaults).

But we're actually down inside xmmap()...

So we release the window we do have (by calling release_pack_memory),
assuming there is some memory pressure...

   munmap(0xb5e2d000, 33554432 <unfinished ...>
   <... munmap resumed> )            = 0
   close(5 <unfinished ...>
   <... close resumed> )             = 0

And that was the last window in this packfile.  So we closed it.
Way to go us.  Our xmmap did not expect release_pack_memory to
close the fd its about to map...

   mmap2(NULL, 31020635, PROT_READ, MAP_PRIVATE, 5, 0x6000 <unfinished ...>
   <... mmap2 resumed> )             = -1 EBADF (Bad file descriptor)

And so the Linux kernel happily tells us f' off.

   write(2, "fatal: ", 7 <unfinished ...>
   <... write resumed> )             = 7
   write(2, "Out of memory? mmap failed: Bad "..., 47 <unfinished ...>
   <... write resumed> )             = 47

And we report the bad file descriptor error, and not the ENOMEM,
and die, claiming we are out of memory.  But actually that mmap
should have succeeded, as we had enough memory for that window,
seeing as how we released the prior one.

Originally when I developed the sliding window mmap feature I had
this exact same bug in fast-import, and I dealt with it by handing
in the struct packed_git* we want to open the new window for, as the
caller wasn't prepared to reopen the packfile if unuse_one_window
closed it.  The same is true here from xmmap, but the caller doesn't
have the struct packed_git* handy.  So I'm using the file descriptor
instead to perform the same test.

Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoinit_buffer(): Kill buf pointer
Luiz Fernando N. Capitulino [Wed, 25 Apr 2007 14:18:41 +0000 (11:18 -0300)]
init_buffer(): Kill buf pointer

We don't need it, it's possible to assign the block of memory to bufp

Signed-off-by: Luiz Fernando N. Capitulino <lcapitulino@mandriva.com.br>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agocore-tutorial: minor fixes
Luiz Fernando N. Capitulino [Wed, 25 Apr 2007 14:18:28 +0000 (11:18 -0300)]
core-tutorial: minor fixes

- Do not break the line when it's not needed
- s/Your/You

Signed-off-by: Luiz Fernando N. Capitulino <lcapitulino@mandriva.com.br>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoread_cache_from(): small simplification
Luiz Fernando N. Capitulino [Wed, 25 Apr 2007 14:18:17 +0000 (11:18 -0300)]
read_cache_from(): small simplification

This change 'opens' the code block which maps the index file into
memory, making the code clearer and easier to read.

Signed-off-by: Luiz Fernando N. Capitulino <lcapitulino@mandriva.com.br>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoentry.c: Use const qualifier for 'struct checkout' parameters
Luiz Fernando N. Capitulino [Wed, 25 Apr 2007 14:18:08 +0000 (11:18 -0300)]
entry.c: Use const qualifier for 'struct checkout' parameters

Signed-off-by: Luiz Fernando N. Capitulino <lcapitulino@mandriva.com.br>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoremove_subtree(): Use strerror() when possible
Luiz Fernando N. Capitulino [Wed, 25 Apr 2007 14:17:56 +0000 (11:17 -0300)]
remove_subtree(): Use strerror() when possible

Signed-off-by: Luiz Fernando N. Capitulino <lcapitulino@mandriva.com.br>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoMerge branch 'jc/the-index'
Junio C Hamano [Wed, 25 Apr 2007 05:13:22 +0000 (22:13 -0700)]
Merge branch 'jc/the-index'

* jc/the-index:
  Make read-cache.c "the_index" free.
  Move index-related variables into a structure.

17 years agoMerge branch 'mk/diff'
Junio C Hamano [Wed, 25 Apr 2007 05:12:48 +0000 (22:12 -0700)]
Merge branch 'mk/diff'

* mk/diff:
  Diff between two blobs should take mode changes into account now.
  use mode of the tree in git-diff, if <tree>:<file> syntax is used
  store mode in rev_list, if <tree>:<filename> syntax is used
  add add_object_array_with_mode
  add get_sha1_with_mode
  Add S_IFINVALID mode

17 years agoMerge branch 'maint'
Junio C Hamano [Wed, 25 Apr 2007 05:07:34 +0000 (22:07 -0700)]
Merge branch 'maint'

* maint:
  Remove usernames from all commit messages, not just when using svmprops
  applymbox & quiltimport: typofix.
  Create a sysconfdir variable, and use it for ETC_GITCONFIG

17 years agoMerge branch 'master' of git://repo.or.cz/git/fastimport
Junio C Hamano [Wed, 25 Apr 2007 05:02:38 +0000 (22:02 -0700)]
Merge branch 'master' of git://repo.or.cz/git/fastimport

* 'master' of git://repo.or.cz/git/fastimport:
  fast-import: size_t vs ssize_t
  fix importing of subversion tars
  Don't repack existing objects in fast-import

17 years agoRemove usernames from all commit messages, not just when using svmprops
Adam Roben [Wed, 25 Apr 2007 01:02:07 +0000 (18:02 -0700)]
Remove usernames from all commit messages, not just when using svmprops

Signed-off-by: Sam Vilain <sam.vilain@catalyst.net.nz>
Signed-off-by: Adam Roben <aroben@apple.com>
Acked-by: Eric Wong <normalperson@yhbt.net>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoapplymbox & quiltimport: typofix.
Junio C Hamano [Tue, 24 Apr 2007 21:27:00 +0000 (14:27 -0700)]
applymbox & quiltimport: typofix.

6777c380 fixed only one of three typos introduced in an earlier
patch 87ab7992.  This fixes the other two.

Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agogitattributes documentation: clarify overriding
Junio C Hamano [Tue, 24 Apr 2007 20:46:02 +0000 (13:46 -0700)]
gitattributes documentation: clarify overriding

Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agofast-import: size_t vs ssize_t
Sami Farin [Tue, 24 Apr 2007 19:56:02 +0000 (22:56 +0300)]
fast-import: size_t vs ssize_t

size_t is unsigned, so (n < 0) is never true.

Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agot/test-lib.sh: Protect ourselves from common misconfiguration
Junio C Hamano [Tue, 24 Apr 2007 18:21:47 +0000 (11:21 -0700)]
t/test-lib.sh: Protect ourselves from common misconfiguration
that exports CDPATH to the environment

Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agofix importing of subversion tars
Uwe Kleine-König [Tue, 24 Apr 2007 11:51:04 +0000 (13:51 +0200)]
fix importing of subversion tars

add a / between the prefix and name fields of the tar archive if prefix
is non-empty.

Signed-off-by: Uwe Kleine-König <ukleinek@informatik.uni-freiburg.de>
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
17 years agoCreate a sysconfdir variable, and use it for ETC_GITCONFIG
Josh Triplett [Tue, 24 Apr 2007 07:51:35 +0000 (00:51 -0700)]
Create a sysconfdir variable, and use it for ETC_GITCONFIG

ETC_GITCONFIG defaults to $(prefix)/etc/gitconfig, so if you just set
prefix=/usr, you end up with a git that looks in /usr/etc/gitconfig, rather
than /etc/gitconfig as specified by the FHS.  Furthermore, setting
ETC_GITCONFIG does not fix the paths to any future system-wide configuration
files.

Factor out the path to the system-wide configuration directory into a variable
sysconfdir, normally set to $(prefix)/etc, but set to /etc when prefix=/usr .
This fixes the prefix=/usr problem for ETC_GITCONFIG, and allows centralized
configuration of any future system-wide configuration files without requiring
further action from package maintainers or other people building and
installing git.

Signed-off-by: Josh Triplett <josh@freedesktop.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoDiff between two blobs should take mode changes into account now.
Junio C Hamano [Mon, 23 Apr 2007 06:56:22 +0000 (23:56 -0700)]
Diff between two blobs should take mode changes into account now.

Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agouse mode of the tree in git-diff, if <tree>:<file> syntax is used
Martin Koegler [Sun, 22 Apr 2007 16:44:00 +0000 (18:44 +0200)]
use mode of the tree in git-diff, if <tree>:<file> syntax is used

Signed-off-by: Martin Koegler <mkoegler@auto.tuwien.ac.at>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agostore mode in rev_list, if <tree>:<filename> syntax is used
Martin Koegler [Sun, 22 Apr 2007 16:43:59 +0000 (18:43 +0200)]
store mode in rev_list, if <tree>:<filename> syntax is used

Signed-off-by: Martin Koegler <mkoegler@auto.tuwien.ac.at>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoadd add_object_array_with_mode
Martin Koegler [Sun, 22 Apr 2007 16:43:58 +0000 (18:43 +0200)]
add add_object_array_with_mode

Each object in struct object_array is extended with the mode.
If not specified, S_IFINVALID is used. An object with an mode value
can be added with add_object_array_with_mode.

Signed-off-by: Martin Koegler <mkoegler@auto.tuwien.ac.at>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoadd get_sha1_with_mode
Martin Koegler [Mon, 23 Apr 2007 20:55:05 +0000 (22:55 +0200)]
add get_sha1_with_mode

get_sha1_with_mode basically behaves as get_sha1. It has an additional
parameter for storing the mode of the object.

If the mode can not be determined, it stores S_IFINVALID.

Signed-off-by: Martin Koegler <mkoegler@auto.tuwien.ac.at>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoAdd S_IFINVALID mode
Martin Koegler [Sun, 22 Apr 2007 16:43:56 +0000 (18:43 +0200)]
Add S_IFINVALID mode

S_IFINVALID is used to signal, that no mode information is available.

Signed-off-by: Martin Koegler <mkoegler@auto.tuwien.ac.at>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoMerge branch 'maint'
Junio C Hamano [Tue, 24 Apr 2007 07:08:16 +0000 (00:08 -0700)]
Merge branch 'maint'

* maint:
  Documentation/git-reset.txt: suggest git commit --amend in example.
  Build RPM with ETC_GITCONFIG=/etc/gitconfig
  Ignore all man sections as they are generated files.
  Fix typo in git-am: s/Was is/Was it/
  Reverse the order of -b and --track in the man page.
  dir.c(common_prefix): Fix two bugs

Conflicts:

git.spec.in

17 years agogit-fetch: Fix "argument list too long"
OGAWA Hirofumi [Mon, 23 Apr 2007 19:26:26 +0000 (04:26 +0900)]
git-fetch: Fix "argument list too long"

If $ls_remote_result was too long,

    git-fetch--tool -s pick-rref "$rref" "$ls_remote_result"

in git-fetch will fail with "argument list too long".

This patch fixes git-fetch--tool and git-fetch by passing
$ls_remote_result via stdin.

Signed-off-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoDocumentation/git-reset.txt: suggest git commit --amend in example.
Gerrit Pape [Mon, 23 Apr 2007 12:06:29 +0000 (12:06 +0000)]
Documentation/git-reset.txt: suggest git commit --amend in example.

In example 'Undo a commit and redo', refer to 'git commit --amend', as
this is the easier alternative.

Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoBuild RPM with ETC_GITCONFIG=/etc/gitconfig
Junio C Hamano [Tue, 24 Apr 2007 06:17:41 +0000 (23:17 -0700)]
Build RPM with ETC_GITCONFIG=/etc/gitconfig

Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoIgnore all man sections as they are generated files.
Brian Gernhardt [Tue, 24 Apr 2007 00:02:34 +0000 (20:02 -0400)]
Ignore all man sections as they are generated files.

Signed-off-by: Brian Gernhardt <benji@silverinsanity.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoFix typo in git-am: s/Was is/Was it/
Josh Triplett [Tue, 24 Apr 2007 00:32:04 +0000 (17:32 -0700)]
Fix typo in git-am: s/Was is/Was it/

Signed-off-by: Josh Triplett <josh@freedesktop.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoAdd clean.requireForce option, and add -f option to git-clean to override it
Josh Triplett [Tue, 24 Apr 2007 00:18:16 +0000 (17:18 -0700)]
Add clean.requireForce option, and add -f option to git-clean to override it

Add a new configuration option clean.requireForce.  If set, git-clean will
refuse to run, unless forced with the new -f option, or not acting due to -n.

Signed-off-by: Josh Triplett <josh@freedesktop.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agot6030: grab commit object name as we go
Junio C Hamano [Tue, 24 Apr 2007 05:10:47 +0000 (22:10 -0700)]
t6030: grab commit object name as we go

Instead of running rev-list and picking earlier lines using head/tail pipeline,
grab commit object name as we build commits.  This also removes a non POSIX
use of tail with -linenum (more posixly-correct way to say it is "-n linenum")

Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agot5302: avoid using tail -c
Junio C Hamano [Tue, 24 Apr 2007 05:05:22 +0000 (22:05 -0700)]
t5302: avoid using tail -c

A Large Angry SCM (gitzilla) noticed that on an unnamed platform, tail -c
wants its byte count as part of the option, not as a separate argument.

Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agot4201: Do not display weird characters on the terminal
Johannes Schindelin [Mon, 23 Apr 2007 18:46:21 +0000 (20:46 +0200)]
t4201: Do not display weird characters on the terminal

Now that git-commit got chatty, we have to shut it up again.

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoadd file checkout progress
Nicolas Pitre [Mon, 23 Apr 2007 17:17:19 +0000 (13:17 -0400)]
add file checkout progress

It is nice to see what is happening when checking out large amount of
files, either with git-checkout or git-reset.  The new progress code
already decides what is a "significant amount" and displays progress
only in that case..

Signed-off-by: Nicolas Pitre <nico@cam.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoReverse the order of -b and --track in the man page.
Brian Gernhardt [Mon, 23 Apr 2007 23:56:45 +0000 (19:56 -0400)]
Reverse the order of -b and --track in the man page.

Using "-b --track newbranch oldbranch" gives the error:

  git checkout: updating paths is incompatible with switching
  branches/forcing

However, "--track -b ..." works just fine.

Signed-off-by: Brian Gernhardt <benji@silverinsanity.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agodir.c(common_prefix): Fix two bugs
Johannes Schindelin [Mon, 23 Apr 2007 08:21:25 +0000 (10:21 +0200)]
dir.c(common_prefix): Fix two bugs

The function common_prefix() is used to find the common subdirectory of
a couple of pathnames. When checking if the next pathname matches up with
the prefix, it incorrectly checked the whole path, not just the prefix
(including the slash). Thus, the expensive part of the loop was executed
always.

The other bug is more serious: if the first and the last pathname in the
list have a longer common prefix than the common prefix for _all_ pathnames
in the list, the longer one would be chosen. This bug was probably hidden
by the fact that bash's wildcard expansion sorts the results, and the code
just so happens to work with sorted input.

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoDocument "diff=driver" attribute
Junio C Hamano [Mon, 23 Apr 2007 07:21:02 +0000 (00:21 -0700)]
Document "diff=driver" attribute

Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoMake read-cache.c "the_index" free.
Junio C Hamano [Mon, 2 Apr 2007 06:26:07 +0000 (23:26 -0700)]
Make read-cache.c "the_index" free.

This makes all low-level functions defined in read-cache.c to
take an explicit index_state structure as their first parameter,
to specify which index to work on.  These functions
traditionally operated on "the_index" and were named foo_cache();
the counterparts this patch introduces are called foo_index().

The traditional foo_cache() functions are made into macros that
give "the_index" to their corresponding foo_index() functions.

Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoMove index-related variables into a structure.
Junio C Hamano [Mon, 2 Apr 2007 01:14:06 +0000 (18:14 -0700)]
Move index-related variables into a structure.

This defines a index_state structure and moves index-related
global variables into it.  Currently there is one instance of
it, the_index, and everybody accesses it, so there is no code
change.

Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agogitweb: Show "no difference" message for empty diff v1.5.2-rc0
Martin Koegler [Mon, 23 Apr 2007 05:49:25 +0000 (22:49 -0700)]
gitweb: Show "no difference" message for empty diff

Currently, gitweb shows only header and footer, if no differences are
found. This patch adds a "No differences found" message for the html
output.

Signed-off-by: Martin Koegler <mkoegler@auto.tuwien.ac.at>
Signed-off-by: Jakub Narebski <jnareb@gmail.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agodelay progress display when checking out files
Nicolas Pitre [Fri, 20 Apr 2007 19:27:44 +0000 (15:27 -0400)]
delay progress display when checking out files

Let's start displaying progress only if more than 50% of total number
of files remains to be checked out after 2 seconds.

Signed-off-by: Nicolas Pitre <nico@cam.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoprovide a facility for "delayed" progress reporting
Nicolas Pitre [Fri, 20 Apr 2007 19:05:27 +0000 (15:05 -0400)]
provide a facility for "delayed" progress reporting

This allows for progress to be displayed only if the progress has not
reached a specified percentage treshold within a given delay in seconds.

Signed-off-by: Nicolas Pitre <nico@cam.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agomake progress "title" part of the common progress interface
Nicolas Pitre [Fri, 20 Apr 2007 18:10:07 +0000 (14:10 -0400)]
make progress "title" part of the common progress interface

If the progress bar ends up in a box, better provide a title for it too.

Signed-off-by: Nicolas Pitre <nico@cam.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agocommon progress display support
Nicolas Pitre [Wed, 18 Apr 2007 18:27:45 +0000 (14:27 -0400)]
common progress display support

Instead of having this code duplicated in multiple places, let's have
a common interface for progress display.  If someday someone wishes to
display a cheezy progress bar instead then only one file will have to
be changed.

Note: I left merge-recursive.c out since it has a strange notion of
progress as it apparently increase the expected total number as it goes.
Someone with more intimate knowledge of what that is supposed to mean
might look at converting it to the common progress interface.

Signed-off-by: Nicolas Pitre <nico@cam.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoSupport 'diff=pgm' attribute
Junio C Hamano [Mon, 23 Apr 2007 00:52:55 +0000 (17:52 -0700)]
Support 'diff=pgm' attribute

This enhances the attributes mechanism so that external programs
meant for existing GIT_EXTERNAL_DIFF interface can be specifed
per path.

To configure such a custom diff driver, first define a custom
diff driver in the configuration:

[diff "my-c-diff"]
command = <<your command string comes here>>

Then mark the paths that you want to use this custom driver
using the attribute mechanism.

*.c diff=my-c-diff

The intent of this separation is that the attribute mechanism is
used for specifying the type of the contents, while the
configuration mechanism is used to define what needs to be done
to that type of the contents, which would be specific to both
platform and personal taste.

Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agopack-objects: make generated packfile read-only
Junio C Hamano [Mon, 23 Apr 2007 02:00:16 +0000 (19:00 -0700)]
pack-objects: make generated packfile read-only

Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoUpdate tests not to assume that generated packfiles are writable.
Junio C Hamano [Mon, 23 Apr 2007 01:59:34 +0000 (18:59 -0700)]
Update tests not to assume that generated packfiles are writable.

Signed-off-by: Junio C Hamano <junkio@cox.net>
17 years agoFix 'quickfix' on pack-objects.
Junio C Hamano [Sun, 22 Apr 2007 19:28:34 +0000 (12:28 -0700)]
Fix 'quickfix' on pack-objects.

The earlier quickfix forced world-readable permission bits.  This
updates it to honor umask and core.sharedrepository settings.

Signed-off-by: Junio C Hamano <junkio@cox.net>