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>
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>
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
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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.
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
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
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
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
Junio C Hamano [Sun, 22 Apr 2007 18:49:35 +0000 (11:49 -0700)]
pack-objects: quickfix for permission modes.
mkstemp() often creates the file in 0600 which means the
resulting packfile is not readable by anybody other than the
repository owner. Force 0644 for now, even though this is not
strictly correct.
Signed-off-by: Junio C Hamano <junkio@cox.net>
Alex Riesen [Sun, 22 Apr 2007 14:11:54 +0000 (16:11 +0200)]
Fix crash in t0020 (crlf conversion)
Reallocated wrong size.
Noticed on Ubuntu 7.04 probably because it has some malloc diagnostics in libc:
"git-read-tree --reset -u HEAD" aborted in the test. Valgrind sped up the
debugging greatly: took me 10 minutes.
Signed-off-by: Alex Riesen <raa.lkml@gmail.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>
Alex Riesen [Sun, 22 Apr 2007 14:12:22 +0000 (16:12 +0200)]
Fix a typo in crlf conversion code
Also, noticed by valgrind: the code caused a read out-of-bounds.
Some comments updated as well (they still reflected old calling
conventions).
Signed-off-by: Alex Riesen <raa.lkml@gmail.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>
Shawn O. Pearce [Sun, 22 Apr 2007 01:14:39 +0000 (21:14 -0400)]
Cleanup variables in cat-file
I want to add new command line options to cat-file, but
to do that we need to change how we handle argv[] first.
This is a simple cleanup that assigns names to the two
arguments we currently care about.
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
Junio C Hamano [Sun, 22 Apr 2007 07:26:56 +0000 (00:26 -0700)]
Update draft release notes for v1.5.2
Signed-off-by: Junio C Hamano <junkio@cox.net>
Junio C Hamano [Sun, 22 Apr 2007 07:11:41 +0000 (00:11 -0700)]
Documentation/Makefile: fix section (5) installation
Signed-off-by: Junio C Hamano <junkio@cox.net>
Junio C Hamano [Sun, 22 Apr 2007 06:51:27 +0000 (23:51 -0700)]
Update documentation links to point at v1.5.1.2
Signed-off-by: Junio C Hamano <junkio@cox.net>
Junio C Hamano [Sun, 22 Apr 2007 00:41:40 +0000 (17:41 -0700)]
Merge branch 'lt/objalloc'
* 'lt/objalloc':
Clean up object creation to use more common code
Use proper object allocators for unknown object nodes too
Junio C Hamano [Sun, 22 Apr 2007 00:40:48 +0000 (17:40 -0700)]
Merge branch 'jc/add'
* jc/add:
git-add -u: match the index with working tree.
Junio C Hamano [Sun, 22 Apr 2007 00:38:00 +0000 (17:38 -0700)]
Merge branch 'jc/attr'
* 'jc/attr': (28 commits)
lockfile: record the primary process.
convert.c: restructure the attribute checking part.
Fix bogus linked-list management for user defined merge drivers.
Simplify calling of CR/LF conversion routines
Document gitattributes(5)
Update 'crlf' attribute semantics.
Documentation: support manual section (5) - file formats.
Simplify code to find recursive merge driver.
Counto-fix in merge-recursive
Fix funny types used in attribute value representation
Allow low-level driver to specify different behaviour during internal merge.
Custom low-level merge driver: change the configuration scheme.
Allow the default low-level merge driver to be configured.
Custom low-level merge driver support.
Add a demonstration/test of customized merge.
Allow specifying specialized merge-backend per path.
merge-recursive: separate out xdl_merge() interface.
Allow more than true/false to attributes.
Document git-check-attr
Change attribute negation marker from '!' to '-'.
...
Junio C Hamano [Sun, 22 Apr 2007 00:21:10 +0000 (17:21 -0700)]
Merge branch 'lt/gitlink'
* lt/gitlink:
Tests for core subproject support
Expose subprojects as special files to "git diff" machinery
Fix some "git ls-files -o" fallout from gitlinks
Teach "git-read-tree -u" to check out submodules as a directory
Teach git list-objects logic to not follow gitlinks
Fix gitlink index entry filesystem matching
Teach "git-read-tree -u" to check out submodules as a directory
Teach git list-objects logic not to follow gitlinks
Don't show gitlink directories when we want "other" files
Teach git-update-index about gitlinks
Teach directory traversal about subprojects
Fix thinko in subproject entry sorting
Teach core object handling functions about gitlinks
Teach "fsck" not to follow subproject links
Add "S_IFDIRLNK" file mode infrastructure for git links
Add 'resolve_gitlink_ref()' helper function
Avoid overflowing name buffer in deep directory structures
diff-lib: use ce_mode_from_stat() rather than messing with modes manually
Junio C Hamano [Sun, 22 Apr 2007 00:20:50 +0000 (17:20 -0700)]
Merge branch 'np/pack'
* np/pack: (27 commits)
document --index-version for index-pack and pack-objects
pack-objects: remove obsolete comments
pack-objects: better check_object() performances
add get_size_from_delta()
pack-objects: make in_pack_header_size a variable of its own
pack-objects: get rid of create_final_object_list()
pack-objects: get rid of reuse_cached_pack
pack-objects: clean up list sorting
pack-objects: rework check_delta_limit usage
pack-objects: equal objects in size should delta against newer objects
pack-objects: optimize preferred base handling a bit
clean up add_object_entry()
tests for various pack index features
use test-genrandom in tests instead of /dev/urandom
simple random data generator for tests
validate reused pack data with CRC when possible
allow forcing index v2 and 64-bit offset treshold
pack-redundant.c: learn about index v2
show-index.c: learn about index v2
sha1_file.c: learn about index version 2
...
Junio C Hamano [Sun, 22 Apr 2007 00:19:34 +0000 (17:19 -0700)]
Merge branch 'jp/refs'
* jp/refs:
refs.c: add a function to sort a ref list, rather then sorting on add
Junio C Hamano [Sun, 22 Apr 2007 00:19:25 +0000 (17:19 -0700)]
Merge branch 'jc/quickfetch'
* jc/quickfetch:
Make sure quickfetch is not fooled with a previous, incomplete fetch.
git-fetch: use fetch--tool pick-rref to avoid local fetch from alternate
git-fetch--tool pick-rref
Junio C Hamano [Sun, 22 Apr 2007 00:16:48 +0000 (17:16 -0700)]
Merge branch 'maint'
* maint:
GIT 1.5.1.2
perl: install private Error.pm if the site version is older than our own
git-clone: fix dumb protocol transport to clone from pack-pruned ref
Junio C Hamano [Sat, 21 Apr 2007 20:57:07 +0000 (13:57 -0700)]
GIT 1.5.1.2
Signed-off-by: Junio C Hamano <junkio@cox.net>
Eric Wong [Sat, 21 Apr 2007 07:55:00 +0000 (00:55 -0700)]
perl: install private Error.pm if the site version is older than our own
bdash (on IRC) had a problem with Git.pm (via git-svn) when his
site installation of Error.pm was older than the version we
package.
Signed-off-by: Eric Wong <normalperson@yhbt.net>
Signed-off-by: Junio C Hamano <junkio@cox.net>
Junio C Hamano [Sat, 21 Apr 2007 10:11:10 +0000 (03:11 -0700)]
lockfile: record the primary process.
The usual process flow is the main process opens and holds the lock to
the index, does its thing, perhaps spawning children during the course,
and then writes the resulting index out by releaseing the lock.
However, the lockfile interface uses atexit(3) to clean it up, without
regard to who actually created the lock. This typically leads to a
confusing behaviour of lock being released too early when the child
exits, and then the parent process when it calls commit_lockfile()
finds that it cannot unlock it.
This fixes the problem by recording who created and holds the lock, and
upon atexit(3) handler, child simply ignores the lockfile the parent
created.
Signed-off-by: Junio C Hamano <junkio@cox.net>
Junio C Hamano [Sat, 21 Apr 2007 06:44:02 +0000 (23:44 -0700)]
convert.c: restructure the attribute checking part.
This separates the checkattr() call and interpretation of the
returned value specific to the 'crlf' attribute into separate
routines, so that we can run a single call to checkattr() to
check for more than one attributes, and then interprete what
the returned settings mean separately.
Signed-off-by: Junio C Hamano <junkio@cox.net>
Junio C Hamano [Sat, 21 Apr 2007 07:05:31 +0000 (00:05 -0700)]
Fix bogus linked-list management for user defined merge drivers.
ll_user_merge_tail is supposed to point at the pointer to be
updated to point at a newly created item.
Signed-off-by: Junio C Hamano <junkio@cox.net>
Alex Riesen [Thu, 19 Apr 2007 00:05:03 +0000 (02:05 +0200)]
Simplify calling of CR/LF conversion routines
Signed-off-by: Alex Riesen <raa.lkml@gmail.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>
Linus Torvalds [Sat, 21 Apr 2007 01:30:06 +0000 (18:30 -0700)]
Fix a copy-n-paste bug in the object decorator code.
Duh.
When I did the object decorator thing, I made the "loop over the hash"
function use the same logic for updating the hash, ie made them use
if (++j >= size)
j = 0;
for both the hash update for both "insert" and "lookup"
HOWEVER.
For some inexplicable reason I had an extraneous
j++;
in the insert path (probably just from the fact that the old code there
used
j++;
if (j >= size)
j = 0;
and when I made them use the same logic I just didn't remove the old
extraneous line properly.
This fixes it.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
Junio C Hamano [Sat, 21 Apr 2007 00:25:27 +0000 (17:25 -0700)]
git-clone: fix dumb protocol transport to clone from pack-pruned ref
This forward-ports a fix from
2986c022 to git-clone.
Signed-off-by: Junio C Hamano <junkio@cox.net>
Shawn O. Pearce [Fri, 20 Apr 2007 15:23:45 +0000 (11:23 -0400)]
Don't repack existing objects in fast-import
Some users of fast-import have been trying to use it to rewrite
commits and trees, an activity where the all of the relevant blobs
are already available from the existing packfiles. In such a case
we don't want to repack a blob, even if the frontend application
has supplied us the raw data rather than a mark or a SHA-1 name.
I'm intentionally only checking the packfiles that existed when
fast-import started and am always ignoring all loose object files.
We ignore loose objects because fast-import tends to operate on a
very large number of objects in a very short timespan, and it is
usually creating new objects, not reusing existing ones. In such
a situtation the majority of the objects will not be found in the
existing packfiles, nor will they be loose object files. If the
frontend application really wants us to look at loose object files,
then they can just repack the repository before running fast-import.
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
Junio C Hamano [Fri, 20 Apr 2007 08:39:39 +0000 (01:39 -0700)]
git-add -u: match the index with working tree.
This is a shorthand of what "git commit -a" does in preparation
for making a commit, which is:
git diff-files --name-only -z | git update-index --remove -z --stdin
Signed-off-by: Junio C Hamano <junkio@cox.net>
Aneesh Kumar K.V [Thu, 19 Apr 2007 16:56:03 +0000 (22:26 +0530)]
gitview: annotation support
List files modifed as a part of the commit in the diff window
Support annotation of the file listed in the diff window
Support history browsing in the annotation window.
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@gmail.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>
Shawn O. Pearce [Fri, 20 Apr 2007 06:37:18 +0000 (02:37 -0400)]
Kill the useless progress meter in merge-recursive
The mess known as the progress meter in merge-recursive was my own
fault; I put it in thinking that we might be spending a lot of time
resolving unmerged entries in the index that were not handled by
the simple 3-way index merge code.
Turns out we don't really spend that much time there, so the progress
meter was pretty much always jumping to "(n/n) 100%" as soon as
the program started. That isn't a very good indication of progress.
Since I don't have a great solution for how a progress meter should
work here, I'm proposing we back it out entirely.
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
Brian Gernhardt [Thu, 19 Apr 2007 04:51:06 +0000 (00:51 -0400)]
Remove case-sensitive file in t3030-merge-recursive.
Rename "A" to the unused "c"
Signed-off-by: Brian Gernhardt <benji@silverinsanity.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>
Alexandre Julliard [Thu, 19 Apr 2007 11:16:58 +0000 (13:16 +0200)]
git.el: Add a commit description to the reflog.
Add a description of the commit to the reflog using the first line of
the log message, the same way the git-commit script does it.
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
Shawn O. Pearce [Fri, 20 Apr 2007 06:08:47 +0000 (02:08 -0400)]
Contribute a fairly paranoid update hook
I'm using a variant of this update hook in a corporate environment
where we perform some validations of the commits and tags that
are being pushed. The model is a "central repository" type setup,
where users are given access to push to specific branches within
the shared central repository. In this particular installation we
run a specially patched git-receive-pack in setuid mode via SSH,
allowing all writes into the repository as the repository owner,
but only if this hook blesses it.
One of the major checks we perform with this hook is that the
'committer' line of a commit, or the 'tagger' line of a new annotated
tag actually correlates to the UNIX user who is performing the push.
Users can falsify these lines on their local repositories, but
the central repository that management trusts will reject all such
forgery attempts. Of course 'author' lines are still allowed to
be any value, as sometimes changes do come from other individuals.
Another nice feature of this hook is the access control lists for
all repositories on the system can also be stored and tracked in
a supporting Git repository, which can also be access controlled
by itself. This allows full auditing of who-had-what-when-and-why,
thanks to git-blame's data mining capabilities.
Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
Junio C Hamano [Fri, 20 Apr 2007 06:06:21 +0000 (23:06 -0700)]
Merge branch 'maint'
* maint:
Update git-config documentation
Fix unmatched emphasis tag in git-tutorial
Update git-cherry-pick documentation
Update git-archive documentation
Linus Torvalds [Fri, 20 Apr 2007 05:48:21 +0000 (22:48 -0700)]
Fix working directory errno handling when unlinking a directory
Alex Riesen noticed that the case where a file replaced a directory entry
in the working tree was broken on cygwin. It turns out that the code made
some Linux-specific assumptions, and also ignored errors entirely for the
case where the entry was a symlink rather than a file.
This cleans it up by separating out the common case into a function of its
own, so that both regular files and symlinks can share it, and by making
the error handling more obvious (and not depend on any Linux-specific
behaviour).
Acked-by: Alex Riesen <raa.lkml@gmail.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>
Junio C Hamano [Fri, 20 Apr 2007 03:48:03 +0000 (20:48 -0700)]
Document gitattributes(5)
Signed-off-by: Junio C Hamano <junkio@cox.net>
Junio C Hamano [Fri, 20 Apr 2007 05:37:19 +0000 (22:37 -0700)]
Update 'crlf' attribute semantics.
This updates the semantics of 'crlf' so that .gitattributes file
can say "this is text, even though it may look funny".
Setting the `crlf` attribute on a path is meant to mark the path
as a "text" file. 'core.autocrlf' conversion takes place
without guessing the content type by inspection.
Unsetting the `crlf` attribute on a path is meant to mark the
path as a "binary" file. The path never goes through line
endings conversion upon checkin/checkout.
Unspecified `crlf` attribute tells git to apply the
`core.autocrlf` conversion when the file content looks like
text.
Setting the `crlf` attribut to string value "input" is similar
to setting the attribute to `true`, but also forces git to act
as if `core.autocrlf` is set to `input` for the path.
Signed-off-by: Junio C Hamano <junkio@cox.net>