gitweb.git
Merge branch 'master' into sp/gfiShawn O. Pearce Tue, 30 Jan 2007 16:07:24 +0000 (11:07 -0500)

Merge branch 'master' into sp/gfi

git-fast-import requires use of inttypes.h, but the master branch has
added it to git-compat-util differently than git-fast-import originally
had used it. This merge back of master to the fast-import topic is to
get (and use) inttypes.h the way master is using it.

This is a partially evil merge to remove the call to setup_ident(),
as the master branch now contains a change which makes this implicit
and therefore removed the function declaration. (commit 01754769).

Conflicts:

git-compat-util.h

blameview: Use git-cat-file to read the file content.Aneesh Kumar K.V Tue, 30 Jan 2007 07:56:49 +0000 (13:26 +0530)

blameview: Use git-cat-file to read the file content.

Fix blameview to use git-cat-file to read the file content.
This make sure we show the right content when we have modified
file in the working directory which is not committed.

Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@gmail.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>

git-fetch: Allow fetching the remote HEADSanti Béjar Tue, 30 Jan 2007 09:36:24 +0000 (10:36 +0100)

git-fetch: Allow fetching the remote HEAD

... with:

$ git fetch ${remote} HEAD

Also

$ git fetch ${remote} :${localref}

worked, but

$ git fetch ${remote} HEAD:{localref}

didn't. Now both are equivalent.

Signed-off-by: Santi Béjar <sbejar@gmail.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>

git-send-email: remove debugging output.Junio C Hamano Tue, 30 Jan 2007 10:22:37 +0000 (02:22 -0800)

git-send-email: remove debugging output.

rfc2047 unquoter spitted out an annoying "- unquoted" which was
added during debugging but I forgot to remove.

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

Add a missing fork() error check.Johannes Sixt Wed, 24 Jan 2007 15:03:42 +0000 (16:03 +0100)

Add a missing fork() error check.

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

git-blame: somewhat better commenting.Junio C Hamano Tue, 30 Jan 2007 01:36:22 +0000 (17:36 -0800)

git-blame: somewhat better commenting.

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

Make fsck and fsck-objects be builtins.Mark Wooding Mon, 29 Jan 2007 15:48:06 +0000 (15:48 +0000)

Make fsck and fsck-objects be builtins.

The earlier change df391b192 to rename fsck-objects to fsck broke
fsck-objects. This should fix it again.

Signed-off-by: Mark Wooding <mdw@distorted.org.uk>
Signed-off-by: Junio C Hamano <junkio@cox.net>

git-commit -s: no extra space when sign-offs appear... Junio C Hamano Mon, 29 Jan 2007 09:06:27 +0000 (01:06 -0800)

git-commit -s: no extra space when sign-offs appear at the end already.

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

Replace perl code with pure shell codeSimon 'corecode' Schubert Mon, 29 Jan 2007 08:09:25 +0000 (09:09 +0100)

Replace perl code with pure shell code

Signed-off-by: Simon 'corecode' Schubert <corecode@fs.ei.tum.de>
Signed-off-by: Junio C Hamano <junkio@cox.net>

lock_any_ref_for_update(): do not accept malformatted... Junio C Hamano Mon, 29 Jan 2007 08:57:07 +0000 (00:57 -0800)

lock_any_ref_for_update(): do not accept malformatted refs.

We used to use lock_any_ref_for_update() because the command
needs to also update HEAD (which is not under refs/, so
lock_ref_sha1() cannot be used). The function however did not
check for refs with illegal characters in them.

Use check_ref_format() to catch malformed refs. For this check,
we specifically do not want to say having less than two levels
in the name is illegal to allow HEAD (and perhaps other special
refs in the future).

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

git-fsck-objects is now synonym to git-fsckJunio C Hamano Mon, 29 Jan 2007 00:33:58 +0000 (16:33 -0800)

git-fsck-objects is now synonym to git-fsck

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

[PATCH] Rename git-repo-config to git-config.Tom Prince Mon, 29 Jan 2007 00:16:53 +0000 (16:16 -0800)

[PATCH] Rename git-repo-config to git-config.

Signed-off-by: Tom Prince <tom.prince@ualberta.net>
Signed-off-by: Junio C Hamano <junkio@cox.net>

Heavily expanded update hook to send more useful emails... Andy Parkins Fri, 26 Jan 2007 09:01:04 +0000 (09:01 +0000)

Heavily expanded update hook to send more useful emails than the old hook

I know it's only an example, but having this might save someone else the
trouble of writing an enhanced version for themselves.

It basically does the same job as the old update hook, but with these
differences:
* The recipients list is read from the repository config file from
hooks.mailinglist
* Updating unannotated tags can be allowed by setting
hooks.allowunannotated
* Announcement emails (via annotated tag creation) can be sent to a
different mailing list by setting hooks.announcelist
* Output email is more verbose and generates specific content depending
on whether the ref is a tag, an annotated tag, a branch, or a
tracking branch
* The email is easier to filter; the subject line is prefixed with
[SCM] and a project description pulled from the "description" file
* It catches (and displays differently) branch updates that are
performed with a --force

Obviously, it's nothing that clever - it's the update hook I use on my
repositories but I've tried to keep it general, and tried to make the
output always relevant to the type of update.

Signed-off-by: Andy Parkins <andyparkins@gmail.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>

UNIX reference time of 1970-01-01 00:00 is UTC timezone... Andy Parkins Fri, 26 Jan 2007 08:58:48 +0000 (08:58 +0000)

UNIX reference time of 1970-01-01 00:00 is UTC timezone, not local time zone

I got bitten because in the UK (where one would expect 1970-01-01 00:00
to be UTC 0) some politicians decided to mess around with daylight
savings time from 1968 to 1971; it was permanently BST (+0100). That
means that on my computer the following is true:

$ date --date="1970-01-01 00:00" +"%F %T %z (%Z)"
1970-01-01 00:00:00 +0100 (BST)

This of course means that the --date argument to date is specified in
local time, not UTC. So when the hooks--update script does this:

date=$(date --date="1970-01-01 00:00:00 $ts seconds")

It's actually saying (in my timezone) "1970-01-01 01:00:00 UTC" + $ts.
Clearly this is wrong. The UNIX epoch started at midnight UTC not 1am
UTC.

This leads to the tagged time in hooks--update being shown as one hour
earlier than the true tagged time (in my timezone). The problem would
be worse for other timezones. For a +1300 timezone on 1970-01-01, the
tagged time would be 13 hours earlier. Oops.

The solution is to force the reference time to UTC, which is what this
patch does. In my timezone:

$ date --date="1970-01-01 00:00 +0000" +"%F %T %z (%Z)"
1970-01-01 01:00:00 +0100 (BST)

Much better.

Signed-off-by: Andy Parkins <andyparkins@gmail.com>

Teach for-each-ref about a little language called Tcl.Shawn O. Pearce Sun, 28 Jan 2007 07:39:13 +0000 (02:39 -0500)

Teach for-each-ref about a little language called Tcl.

Love it or hate it, some people actually still program in Tcl. Some
of those programs are meant for interfacing with Git. Programs such as
gitk and git-gui. It may be useful to have Tcl-safe output available
from for-each-ref, just like shell, Perl and Python already enjoy.

Thanks to Sergey Vlasov for pointing out the horrible flaws in the
first and second version of this patch, and steering me in the right
direction for Tcl value quoting.

Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>

Add a sample program 'blameview' to show how to use... Jeff King Sun, 28 Jan 2007 20:53:26 +0000 (12:53 -0800)

Add a sample program 'blameview' to show how to use git-blame --incremental

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

git-push through git protocolLinus Torvalds Sun, 21 Jan 2007 19:04:13 +0000 (11:04 -0800)

git-push through git protocol

This allows pushing over the git:// protocol, and while it's not
authenticated, it could make sense from within a firewalled
setup where nobody but trusted internal people can reach the git
port. git-daemon is possibly easier and faster to set up in the
kind of situation where you set up git instead of CVS inside a
company.

"git-receive-pack" is disabled by default, so you need to enable it
explicitly by starting git-daemon with the "--enable=receive-pack"
command line argument, or by having your config enable it automatically.

Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>

Document 'git-blame --incremental'Junio C Hamano Sun, 28 Jan 2007 20:21:53 +0000 (12:21 -0800)

Document 'git-blame --incremental'

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

Documentation/config.txt: Fix documentation of colour... Mark Wooding Sun, 28 Jan 2007 15:17:36 +0000 (15:17 +0000)

Documentation/config.txt: Fix documentation of colour config tweaks.

* The description of valid colour specifications was rather
incomplete, so fix it so that it actually describes colour specs as
accepted by color_parse().

* The list of colour items allowed in color.diff.BLAH was missing the
`commit' and `whitespace' entries.

Signed-off-by: Mark Wooding <mdw@distorted.org.uk>
Signed-off-by: Junio C Hamano <junkio@cox.net>

wt-status: Actually accept `color.status.BLAH' configur... Mark Wooding Sun, 28 Jan 2007 14:55:03 +0000 (14:55 +0000)

wt-status: Actually accept `color.status.BLAH' configuration variables.

A stupid typo stopped this from working.

Signed-off-by: Mark Wooding <mdw@distorted.org.uk>
Signed-off-by: Junio C Hamano <junkio@cox.net>

git-blame --incremental: don't use pagerRen\e,Ai\e(B Scharfe Sun, 28 Jan 2007 14:25:55 +0000 (15:25 +0100)

git-blame --incremental: don't use pager

Starting a pager defeats the purpose of the incremental output
mode. This changes git-blame to only paginate if --incremental
was not given.

git -p blame --incremental still starts the pager, though.

Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx>
Signed-off-by: Junio C Hamano <junkio@cox.net>

Compute accurate distances in git-describe before output.Shawn O. Pearce Sat, 27 Jan 2007 06:54:21 +0000 (01:54 -0500)

Compute accurate distances in git-describe before output.

My prior change to git-describe attempts to print the distance
between the input commit and the best matching tag, but this distance
was usually only an estimate as we always aborted revision walking
as soon as we overflowed the configured limit on the number of
possible tags (as set by --candidates).

Displaying an estimated distance is not very useful and can just be
downright confusing. Most users (heck, most Git developers) don't
immediately understand why this distance differs from the output
of common tools such as `git rev-list | wc -l`. Even worse, the
estimated distance could change in the future (including decreasing
despite no rebase occuring) if we find more possible tags earlier
on during traversal. (This could happen if more tags are merged
into the branch between queries.) These factors basically make an
estimated distance useless.

Fortunately we are usually most of the way through an accurate
distance computation by the time we abort (due to reaching the
current --candidates limit). This means we can simply finish
counting out the revisions still in our commit queue to present
the accurate distance at the end. The number of commits remaining
in the commit queue is probably less than the number of commits
already traversed, so finishing out the count is not likely to take
very long. This final distance will then always match the output of
`git rev-list | wc -l`.

We can easily reduce the total number of commits that need to be
walked at the end by stopping as soon as all of the commits in the
commit queue are flagged as being merged into the already selected
best possible tag. If that's true then there are no remaining
unseen commits which can contribute to our best possible tag's
depth counter, so further traversal is useless.

Basic testing on my Mac OS X system shows there is no noticable
performance difference between this accurate distance counting
version of git-describe and the prior version of git-describe,
at least when run on git.git.

Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>

Update describe documentation.Junio C Hamano Sat, 27 Jan 2007 07:24:07 +0000 (23:24 -0800)

Update describe documentation.

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

Teach git-describe to display distances from tags.Shawn O. Pearce Thu, 25 Jan 2007 17:39:54 +0000 (12:39 -0500)

Teach git-describe to display distances from tags.

If you get two different describes at different
times from a non-rewinding branch and they both come up with the same
tag name, you can tell which is the 'newer' one by distance. This is
rather common in practice, so its incredibly useful.

[jc: still needs documentation and fixups when traversal gives up
early.]

Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>

git-blame --porcelain: quote filename in c-style when... Junio C Hamano Sun, 28 Jan 2007 09:42:31 +0000 (01:42 -0800)

git-blame --porcelain: quote filename in c-style when needed.

Otherwise a pathname that has funny characters such as LF would
screw up the parsing programs of the output.

Strictly speaking, this is not backward compatible, but the
current output for pathnames that have embedded LF and such
cannot be sanely parsed anyway, and pathnames that only use
characters from the portable pathname character set won't be
affected.

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

git-blame --incrementalLinus Torvalds Sun, 28 Jan 2007 09:34:06 +0000 (01:34 -0800)

git-blame --incremental

This adds --incremental option to help GUI porcelains to show
the result from git-blame incrementally. The output gives the
origin information in the same format as the porcelain format.
The first line has commit object name, the line number of the
first line in the group in the original file, the line number of
that file in the final image, and number of lines in the group.
Then subsequent lines show the metainformation for the commit
when the commit is shown for the first time, except the filename
information is always shown (we cannot even make it conditional
to -C option as blame always follows the renaming of the file
wholesale).

Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>

Don't force everybody to call setup_ident().Junio C Hamano Sun, 28 Jan 2007 08:50:53 +0000 (00:50 -0800)

Don't force everybody to call setup_ident().

Back when only handful commands that created commit and tag were
the only users of committer identity information, it made sense
to explicitly call setup_ident() to pre-fill the default value
from the gecos information. But it is much simpler for programs
to make the call automatic when get_ident() is called these days,
since many more programs want to use the information when updating
the reflog.

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

git-log -g --pretty=oneline should display the reflog... Nicolas Pitre Sun, 28 Jan 2007 03:40:36 +0000 (22:40 -0500)

git-log -g --pretty=oneline should display the reflog message

In the context of reflog output the reflog message is more useful than
the commit message's first line. When relevant the reflog message
will contain that line anyway.

Signed-off-by: Nicolas Pitre <nico@cam.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>

Document --check option to git diff.Bill Lear Sat, 27 Jan 2007 13:21:53 +0000 (07:21 -0600)

Document --check option to git diff.

Signed-off-by: Bill Lear <rael@zopyra.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>

Allow the tag signing key to be specified in the config... Andy Parkins Fri, 26 Jan 2007 14:13:46 +0000 (14:13 +0000)

Allow the tag signing key to be specified in the config file

I did this:

$ git tag -s test-sign
gpg: skipped "Andy Parkins <andyparkins@gmail.com>": secret key not available
gpg: signing failed: secret key not available
failed to sign the tag with GPG.

The problem is that I have used the comment field in my key's UID
definition.

$ gpg --list-keys andy
pub 1024D/4F712F6D 2003-08-14
uid Andy Parkins (Google) <andyparkins@gmail.com>

So when git-tag looks for "Andy Parkins <andyparkins@gmail.com>";
obviously it's not going to be found.

There shouldn't be a requirement that I use the same form of my name in
my git repository and my gpg key - I might want to be formal (Andrew) in
my gpg key and informal (Andy) in the repository. Further I might have
multiple keys in my keyring, and might want to use one that doesn't
match up with the address I use in commit messages.

This patch adds a configuration entry "user.signingkey" which, if
present, will be passed to the "-u" switch for gpg, allowing the tag
signing key to be overridden. If the entry is not present, the fallback
is the original method, which means existing behaviour will continue
untouched.

Signed-off-by: Andy Parkins <andyparkins@gmail.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>

If abbrev is set to zero in git-describe, don't add... Andy Parkins Fri, 26 Jan 2007 14:28:55 +0000 (14:28 +0000)

If abbrev is set to zero in git-describe, don't add the unique suffix

When on a non-tag commit, git-describe normally outputs descriptions of
the form
v1.0.0-g1234567890
Some scripts (for example the update hook script) might just want to
know the name of the nearest tag, so they then have to do
x=$(git-describe HEAD | sed 's/-g*//')
This is costly, but more importantly is fragile as it is relying on the
output format of git-describe, which we would then have to maintain
forever.

This patch adds support for setting the --abbrev option to zero. In
that case git-describe does as it always has, but outputs only the
nearest found tag instead of a completely unique name. This means that
scripts would not have to parse the output format and won't need
changing if the git-describe suffix is ever changed.

Signed-off-by: Andy Parkins <andyparkins@gmail.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>

fix suggested branch creation command when detaching... Nicolas Pitre Fri, 26 Jan 2007 16:50:06 +0000 (11:50 -0500)

fix suggested branch creation command when detaching head

Doing:

$ git checkout HEAD^

Generates the following message:

|warning: you are not on ANY branch anymore.
|If you meant to create a new branch from the commit, you need -b to
|associate a new branch with the wanted checkout. Example:
| git checkout -b <new_branch_name> HEAD^

Of course if the user does as told at this point the created branch
won't be located at the expected commit. Reword this message a bit to
avoid such confusion.

Signed-off-by: Nicolas Pitre <nico@cam.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>

write_in_full: size_t is unsigned.Junio C Hamano Sat, 27 Jan 2007 01:39:03 +0000 (17:39 -0800)

write_in_full: size_t is unsigned.

It received the return value from xwrite() in a size_t variable
'written' and expected comparison with 0 would catch an error
from xwrite().

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

create_symref: check error return from open().Junio C Hamano Sat, 27 Jan 2007 01:00:57 +0000 (17:00 -0800)

create_symref: check error return from open().

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

vc-git.el: Take into account the destination name in... Alexandre Julliard Fri, 26 Jan 2007 10:57:50 +0000 (11:57 +0100)

vc-git.el: Take into account the destination name in vc-checkout.

This is necessary for vc-version-other-window. Based on a patch by Sam
Vilain <sam.vilain@catalyst.net.nz>.

Currently, the vc-git-checkout function uses `git checkout' to fetch a
file from the git repository to the working copy. However, it is
completely ignoring the input argument that specifies the destination
file. `git-checkout' does not support specifying this, so we have to
use `git-cat-file', capture the output in a buffer and then save it.

Signed-off-by: Alexandre Julliard <julliard@winehq.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>

git-merge: leave sensible reflog message when used... Junio C Hamano Fri, 26 Jan 2007 23:09:02 +0000 (15:09 -0800)

git-merge: leave sensible reflog message when used as the first level UI.

It used to throw potentially multi-line log message at reflog.
Just record the heads that were given to be merged at the command
line and the action.

Revert the removal of the check in "git-update-ref -m" I made earlier
which was only a work-around for this.

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

Make sure we do not write bogus reflog entries.Junio C Hamano Fri, 26 Jan 2007 10:26:04 +0000 (02:26 -0800)

Make sure we do not write bogus reflog entries.

The file format dictates that entries are LF terminated so
the message cannot have one in it. Chomp the message to make
sure it only has a single line if necessary, while removing the
leading whitespace.

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

Remove unnecessary found variable from describe.Shawn O. Pearce Thu, 25 Jan 2007 17:40:03 +0000 (12:40 -0500)

Remove unnecessary found variable from describe.

Junio added the found variable to enforce commit date order when two
tags have the same distance from the requested commit. Except it is
unnecessary as match_cnt is already used to record how many possible
tags have been identified thus far.

Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>

Use inttypes.h rather than stdint.h.Jason Riedy Thu, 25 Jan 2007 21:11:40 +0000 (13:11 -0800)

Use inttypes.h rather than stdint.h.

Older Solaris machines lack stdint.h but have inttypes.h.
The standard has inttypes.h including stdint.h, so at worst
this pollutes the namespace a bit.

Signed-off-by: Jason Riedy <ejr@cs.berkeley.edu>
Signed-off-by: Junio C Hamano <junkio@cox.net>

Documentation: pack-refs --all vs default behaviourJunio C Hamano Fri, 26 Jan 2007 06:51:49 +0000 (22:51 -0800)

Documentation: pack-refs --all vs default behaviour

Document the recommended way to prime a repository with tons of
references with 'pack-refs --all -prune'.

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

show-branch -g: default to HEADJunio C Hamano Fri, 26 Jan 2007 06:14:45 +0000 (22:14 -0800)

show-branch -g: default to HEAD

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

Add dangling objects tips.Linus Torvalds Fri, 26 Jan 2007 05:55:34 +0000 (21:55 -0800)

Add dangling objects tips.

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

parse-remote: do not barf on a remote shorthand without... Junio C Hamano Fri, 26 Jan 2007 05:50:49 +0000 (21:50 -0800)

parse-remote: do not barf on a remote shorthand without any refs to fetch.

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

diffcore-pickaxe: fix infinite loop on zero-length... Jeff King Fri, 26 Jan 2007 04:48:58 +0000 (23:48 -0500)

diffcore-pickaxe: fix infinite loop on zero-length needle

The "contains" algorithm runs into an infinite loop if the needle string
has zero length. The loop could be modified to handle this, but it makes
more sense to simply have an empty needle return no matches. Thus, a
command like
git log -S
produces no output.

We place the check at the top of the function so that we get the same
results with or without --pickaxe-regex. Note that until now,
git log -S --pickaxe-regex
would match everything, not nothing.

Arguably, an empty pickaxe string should simply produce an error
message; however, this is still a useful assertion to add to the
algorithm at this layer of the code.

Noticed by Bill Lear.

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

Allow non-developer to clone, checkout and fetch more... Junio C Hamano Fri, 26 Jan 2007 03:05:01 +0000 (19:05 -0800)

Allow non-developer to clone, checkout and fetch more easily.

The code that uses committer_info() in reflog can barf and die
whenever it is asked to update a ref. And I do not think
calling ignore_missing_committer_name() upfront like recent
receive-pack did in the aplication is a reasonable workaround.

What the patch does.

- git_committer_info() takes one parameter. It used to be "if
this is true, then die() if the name is not available due to
bad GECOS, otherwise issue a warning once but leave the name
empty". The reason was because we wanted to prevent bad
commits from being made by git-commit-tree (and its
callers). The value 0 is only used by "git var -l".

Now it takes -1, 0 or 1. When set to -1, it does not
complain but uses the pw->pw_name when name is not
available. Existing 0 and 1 values mean the same thing as
they used to mean before. 0 means issue warnings and leave
it empty, 1 means barf and die.

- ignore_missing_committer_name() and its existing caller
(receive-pack, to set the reflog) have been removed.

- git-format-patch, to come up with the phoney message ID when
asked to thread, now passes -1 to git_committer_info(). This
codepath uses only the e-mail part, ignoring the name. It
used to barf and die. The other call in the same program
when asked to add signed-off-by line based on committer
identity still passes 1 to make sure it barfs instead of
adding a bogus s-o-b line.

- log_ref_write in refs.c, to come up with the name to record
who initiated the ref update in the reflog, passes -1. It
used to barf and die.

The last change means that git-update-ref, git-branch, and
commit walker backends can now be used in a repository with
reflog by somebody who does not have the user identity required
to make a commit. They all used to barf and die.

I've run tests and all of them seem to pass, and also tried "git
clone" as a user whose GECOS is empty -- git clone works again
now (it was broken when reflog was enabled by default).

But this definitely needs extra sets of eyeballs.

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

contrib/emacs/vc-git.el: support vc-version-other-windowSam Vilain Thu, 25 Jan 2007 23:41:23 +0000 (12:41 +1300)

contrib/emacs/vc-git.el: support vc-version-other-window

Currently, the vc-git-checkout function uses `git checkout' to fetch a
file from the git repository to the working copy. However, it is
completely ignoring the input argument that specifies the destination
file. `git-checkout' does not support specifying this, so we have to
use `git-cat-file', capture the output in a buffer and then save it.

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

Fix seriously broken "git pack-refs"Linus Torvalds Fri, 26 Jan 2007 00:51:21 +0000 (16:51 -0800)

Fix seriously broken "git pack-refs"

Do *NOT* try this on a repository you care about:

git pack-refs --all --prune
git pack-refs

because while the first "pack-refs" does the right thing, the second
pack-refs will totally screw you over.

This is because the second one tries to pack only tags; we should
also pack what are already packed -- otherwise we would lose them.

[jc: with an additional test]

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

New files in git weren't being downloaded during CVS... Andy Parkins Mon, 22 Jan 2007 10:56:27 +0000 (10:56 +0000)

New files in git weren't being downloaded during CVS update

If a repository was checked out via git-cvsserver and then later a new
file is added to the git repository via some other method; a CVS update
wasn't fetching the new file.

It would be reported as a new file as
A some/dir/newfile.c
but would never appear in the directory.

The problem seems to be that git-cvsserver was treating these two cases
identically, as "A" type results.

1. New file in repository
2. New file locally

In fact, traditionally, case 1 is treated as a "U" result, and case 2
only is treated as an "A" result. "A", should just report that the file
is added locally and then skip that file during an update as there is
(of course) nothing to send.

In both these cases there is no working revision, so the checking for
"is there no working revision" will return true. The test for case 2
needs refining to say "if there is no working revision and no upstream
revision". This patch does just that, leaving case 1 to be handled by
the normal "U" handler.

I've also updated the log message to more accurately describe the
operation. i.e. that "A" means that content is scheduled for addition;
not that it actually has been added.

Signed-off-by: Andy Parkins <andyparkins@gmail.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>

make --upload-pack option to git-fetch configurableUwe Kleine-König Thu, 25 Jan 2007 04:45:39 +0000 (05:45 +0100)

make --upload-pack option to git-fetch configurable

This introduces the config item remote.<name>.uploadpack to override the
default value (which is "git-upload-pack").

Signed-off-by: Uwe Kleine-König <ukleinek@informatik.uni-freiburg.de>
Signed-off-by: Junio C Hamano <junkio@cox.net>

Consolidate {receive,fetch}.unpackLimitJunio C Hamano Thu, 25 Jan 2007 01:02:15 +0000 (17:02 -0800)

Consolidate {receive,fetch}.unpackLimit

This allows transfer.unpackLimit to specify what these two
configuration variables want to set.

We would probably want to deprecate the two separate variables,
as I do not see much point in specifying them independently.

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

fetch-pack: remove --keep-auto and make it the default.Junio C Hamano Thu, 25 Jan 2007 00:47:24 +0000 (16:47 -0800)

fetch-pack: remove --keep-auto and make it the default.

This makes git-fetch over git native protocol to automatically
decide to keep the downloaded pack if the fetch results in more
than 100 objects, just like receive-pack invoked by git-push
does. This logic is disabled when --keep is explicitly given
from the command line, so that a very small clone still keeps
the downloaded pack as before.

The 100 threshold can be adjusted with fetch.unpacklimit
configuration. We might want to introduce transfer.unpacklimit
to consolidate the two unpacklimit variables, which will be a
topic for the next patch.

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

Allow fetch-pack to decide keeping the fetched pack... Junio C Hamano Tue, 23 Jan 2007 06:37:33 +0000 (22:37 -0800)

Allow fetch-pack to decide keeping the fetched pack without exploding

With --keep-auto option, fetch-pack decides to keep the pack
without exploding it just like receive-pack does.

We may want to later make this the default.

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

Refactor the pack header reading function out of receiv... Junio C Hamano Tue, 23 Jan 2007 05:55:18 +0000 (21:55 -0800)

Refactor the pack header reading function out of receive-pack.c

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

Allow default core.logallrefupdates to be overridden... Alex Riesen Tue, 23 Jan 2007 15:51:18 +0000 (16:51 +0100)

Allow default core.logallrefupdates to be overridden with template's config

Signed-off-by: Alex Riesen <raa.lkml@gmail.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>

ls-remote and clone: accept --upload-pack=<path> as... Junio C Hamano Tue, 23 Jan 2007 08:51:53 +0000 (00:51 -0800)

ls-remote and clone: accept --upload-pack=<path> as well.

This makes them consistent with other commands that take the
path to the upload-pack program. We also pass --upload-pack
instead of --exec to the underlying fetch-pack, although it is
not strictly necessary.

[jc: original motivation from Uwe]

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

rename --exec to --upload-pack for fetch-pack and peek... Uwe Kleine-König Tue, 23 Jan 2007 08:20:17 +0000 (09:20 +0100)

rename --exec to --upload-pack for fetch-pack and peek-remote

Just some option name disambiguation. This is the counter part to
commit d23842fd which made a similar change for push and send-pack.

--exec continues to work.

Signed-off-by: Uwe Kleine-König <ukleinek@informatik.uni-freiburg.de>
Signed-off-by: Junio C Hamano <junkio@cox.net>

Documentation: --amend cannot be combined with -c/... Peter Eriksen Wed, 24 Jan 2007 19:54:46 +0000 (20:54 +0100)

Documentation: --amend cannot be combined with -c/-C/-F.

We used to get the following confusing error message:

$ git commit --amend -a -m foo
Option -m cannot be combined with -c/-C/-F

This is because --amend cannot be combined with -c/-C/-F, which makes
sense, because they try to handle the same log message in different ways.
So update the documentation to reflect this.

Signed-off-by: Peter Eriksen <s022018@student.dtu.dk>
Signed-off-by: Junio C Hamano <junkio@cox.net>

Documentation/config.txt: Correct info about subsection... Jakub Narebski Wed, 24 Jan 2007 14:14:33 +0000 (15:14 +0100)

Documentation/config.txt: Correct info about subsection name

Contrary to variable values, in subsection names parsing character
escape codes (besides literal escaping of " as \", and \ as \\)
is not performed; subsection name cannot contain newlines.

Signed-off-by: Jakub Narebski <jnareb@gmail.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>

git-daemon documentation on enabling services.Junio C Hamano Wed, 24 Jan 2007 23:29:07 +0000 (15:29 -0800)

git-daemon documentation on enabling services.

Noticed by Franck Bui-Huu.

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

reflog inspection: introduce shortcut "-g"Johannes Schindelin Wed, 24 Jan 2007 14:05:16 +0000 (15:05 +0100)

reflog inspection: introduce shortcut "-g"

A short-hand "-g" for "git log --walk-reflogs" and "git
show-branch --reflog" makes it easier to access the reflog
info.

[jc: added -g to show-branch for symmetry]

Signed-off-by: Johannes Schindelin <Johannes.Schindelin@gmx.de>
Signed-off-by: Junio C Hamano <junkio@cox.net>

annotate: use pagerJohannes Schindelin Wed, 24 Jan 2007 14:04:37 +0000 (15:04 +0100)

annotate: use pager

Signed-off-by: Johannes Schindelin <Johannes.Schindelin@gmx.de>
Signed-off-by: Junio C Hamano <junkio@cox.net>

t/t1300-repo-config.sh: value continued on next lineJakub Narebski Tue, 23 Jan 2007 12:37:25 +0000 (13:37 +0100)

t/t1300-repo-config.sh: value continued on next line

Documentation/config.txt:
Variable value ending in a '`\`' is continued on the next line in the
customary UNIX fashion.

Test it.

Signed-off-by: Jakub Narebski <jnareb@gmail.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>

git-checkout -m: fix merge caseJunio C Hamano Wed, 24 Jan 2007 00:51:22 +0000 (16:51 -0800)

git-checkout -m: fix merge case

Commit c1a4278e switched the "merging checkout" implementation
from 3-way read-tree to merge-recursive, but forgot that
merge-recursive will signal an unmerged state with its own exit
status code. This prevented the clean-up phase (paths cleanly
merged should not be updated in the index) from running.

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

reflog gc: a tag that does not point at a commit is... Junio C Hamano Tue, 23 Jan 2007 05:39:03 +0000 (21:39 -0800)

reflog gc: a tag that does not point at a commit is not a crime.

Although unusual, tags can point at any object. Warning only
once is fine, but warning every time about the same tag gets
annoying.

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

contrib/vim: update syntax for changed commit templateJeff King Tue, 23 Jan 2007 03:21:15 +0000 (22:21 -0500)

contrib/vim: update syntax for changed commit template

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

format-patch: fix bug with --stdout in a subdirectoryJeff King Tue, 23 Jan 2007 03:38:28 +0000 (22:38 -0500)

format-patch: fix bug with --stdout in a subdirectory

We set the output directory to the git subdirectory prefix if one has
not already been specified. However, in the case of --stdout, we
explicitly _don't_ want the output directory to be set. The result was
that "git-format-patch --stdout" in a directory besides the project root
produced the "standard output, or directory, which one?" error message.

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

[PATCH] honor --author even with --amend, -C, and -c.Junio C Hamano Mon, 22 Jan 2007 21:03:31 +0000 (13:03 -0800)

[PATCH] honor --author even with --amend, -C, and -c.

Earlier code discarded GIT_AUTHOR_DATE taken from the base
commit when --author was specified. This was often wrong as
that use is likely to fix the spelling of author's name.

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

.mailmap: fix screw-ups in Uwe's nameJunio C Hamano Tue, 23 Jan 2007 00:25:15 +0000 (16:25 -0800)

.mailmap: fix screw-ups in Uwe's name

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

git-svn: remove leading slash when printing removed... Eric Wong Mon, 22 Jan 2007 20:25:30 +0000 (12:25 -0800)

git-svn: remove leading slash when printing removed directories

Not sure why it was there in the first place, we always do our
work relative to the URL we're connected to; even if that URL is
the root of the repository, so the leading slash is pointless...
Lets be consistent when printing things for the user to see.

Signed-off-by: Eric Wong <normalperson@yhbt.net>
Signed-off-by: Junio C Hamano <junkio@cox.net>

sha1_file.c: Avoid multiple calls to find_pack_entry().Peter Eriksen Mon, 22 Jan 2007 20:29:45 +0000 (21:29 +0100)

sha1_file.c: Avoid multiple calls to find_pack_entry().

We used to call find_pack_entry() twice from read_sha1_file() in order
to avoid printing an error message, when the object did not exist. This
is fixed by moving the call to error() to the only place it really
could be called.

Signed-off-by: Peter Eriksen <s022018@student.dtu.dk>
Signed-off-by: Junio C Hamano <junkio@cox.net>

Documentation/config.txt: Document config file syntax... Jakub Narebski Mon, 22 Jan 2007 15:25:47 +0000 (16:25 +0100)

Documentation/config.txt: Document config file syntax better

Separate part of Documentation/config.txt which deals with git config file
syntax into "Syntax" subsection, and expand it. Add information about
subsections, boolean values, escaping and escape sequences in string
values, and continuing variable value on the next line.

Add also proxy settings to config file example to show example of
partially enclosed in double quotes string value.

Parts based on comments by Junio C Hamano, Johannes Schindelin,
config.c, and the smb.conf(5) man page.

Signed-off-by: Jakub Narebski <jnareb@gmail.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>

cvsimport: activate -a option, really.Junio C Hamano Mon, 22 Jan 2007 20:20:14 +0000 (12:20 -0800)

cvsimport: activate -a option, really.

An earlier commit ded9f400 added $opt_a support to disable the
cvsps grace period mechanism, but forgot to tell the option
parser about it.

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

Cleanup uninitialized value in chompAlex Riesen Mon, 22 Jan 2007 14:58:03 +0000 (15:58 +0100)

Cleanup uninitialized value in chomp

which happens if you use ActiveState Perl and a
pipe workaround specially for it.

Signed-off-by: Alex Riesen <raa.lkml@gmail.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>

Force Activestate Perl to tie git command pipe handle... Alex Riesen Mon, 22 Jan 2007 16:16:05 +0000 (17:16 +0100)

Force Activestate Perl to tie git command pipe handle to a handle class

Otherwise it tries to tie it to a scalar and complains about missing
method. Dunno why, may be ActiveState brokenness again.

Signed-off-by: Alex Riesen <raa.lkml@gmail.com>
Acked-by: Petr Baudis <pasky@suse.cz>
Signed-off-by: Junio C Hamano <junkio@cox.net>

Insert ACTIVESTATE_STRING in Git.pmAlex Riesen Mon, 22 Jan 2007 16:14:56 +0000 (17:14 +0100)

Insert ACTIVESTATE_STRING in Git.pm

Also add "git" to the pipe parameters, otherwise it does not work at all, as
no git commands are usable out of git context.

Signed-off-by: Alex Riesen <raa.lkml@gmail.com>
Signed-off-by: Junio C Hamano <junkio@cox.net>

fsck-objects: refactor checking for connectivityLinus Torvalds Mon, 22 Jan 2007 06:26:41 +0000 (22:26 -0800)

fsck-objects: refactor checking for connectivity

This separates the connectivity check into separate codepaths,
one for reachable objects and the other for unreachable ones,
while adding a lot of comments to explain what is going on.

When checking an unreachable object, unlike a reachable one, we
do not have to complain if it does not exist (we used to
complain about a missing blob even when the only thing that
references it is a tree that is dangling). Also we do not have
to check and complain about objects that are referenced by an
unreachable object.

This makes the messages from fsck-objects a lot less noisy and
more useful.

Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>

git-gc: do not run git-prune by default.Junio C Hamano Mon, 22 Jan 2007 07:28:28 +0000 (23:28 -0800)

git-gc: do not run git-prune by default.

git-prune is not safe when run uncontrolled in parallel while
other git operations are creating new objects. To avoid
mistakes, do not run git-prune by default from git-gc.

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

shallow repository: disable unsupported operations... Junio C Hamano Mon, 22 Jan 2007 06:23:58 +0000 (22:23 -0800)

shallow repository: disable unsupported operations for now.

We currently do not support fetching/cloning from a shallow repository
nor pushing into one. Make sure these are not attempted so that we
do not have to worry about corrupting repositories needlessly.

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

is_repository_shallow(): prototype fix.Junio C Hamano Mon, 22 Jan 2007 06:22:23 +0000 (22:22 -0800)

is_repository_shallow(): prototype fix.

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

Make sure git_connect() always give two file descriptors.Junio C Hamano Mon, 22 Jan 2007 01:10:51 +0000 (17:10 -0800)

Make sure git_connect() always give two file descriptors.

Earlier, git_connect() returned the same fd twice or two
separate fds, depending on the way the connection was made (when
we are talking to the other end over a single socket, we used
the same fd twice, and when our end is connected to a pipepair
we used two).

This forced callers who do close() and dup() to really care
which was which, and most of the existing callers got this
wrong, although without much visible ill effect. Many were
closing the same fd twice when we are talking over a single
socket, and one was leaking a fd.

This fixes it to uniformly use two separate fds, so if somebody
wants to close only reader side can just do close() on it
without worrying about it accidentally also closing the writer
side or vice versa.

Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Signed-off-by: Junio C Hamano <junkio@cox.net>

Revert "prune: --grace=time"Junio C Hamano Mon, 22 Jan 2007 05:29:44 +0000 (21:29 -0800)

Revert "prune: --grace=time"

This reverts commit 9b088c4e394df84232cfd37aea78349a495b09c1.

Protecting 'mature' objects does not make it any safer. We should
admit that git-prune is inherently unsafe when run in parallel with
other operations without involving unwarranted locking overhead,
and with the latest git, even rebase and reset would not immediately
create crufts anyway.

Documentation/tutorial-2: Fix interesting typo in an... Junio C Hamano Mon, 22 Jan 2007 05:24:05 +0000 (21:24 -0800)

Documentation/tutorial-2: Fix interesting typo in an example.

Marco Candrian noticed that one cat-file example refers to a
blob object that is never used in the example sequence.

The bug is interesting in that the output from the botched
sample command is consistent with the incorrect blob object
name ;-).

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

GIT v1.5.0-rc2 v1.5.0-rc2Junio C Hamano Sun, 21 Jan 2007 07:44:55 +0000 (23:44 -0800)

GIT v1.5.0-rc2

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

prune: --grace=timeMatthias Lederhofer Fri, 19 Jan 2007 10:49:35 +0000 (11:49 +0100)

prune: --grace=time

This option gives grace period to objects that are unreachable
from the refs from getting pruned.

The default value is 24 hours and may be changed using
gc.prunegrace.

Signed-off-by: Matthias Lederhofer <matled@gmx.net>
Signed-off-by: Junio C Hamano <junkio@cox.net>

--walk-reflogs: do not crash with cyclic reflog ancestryJohannes Schindelin Sat, 20 Jan 2007 21:28:16 +0000 (22:28 +0100)

--walk-reflogs: do not crash with cyclic reflog ancestry

Since you can reset --hard to any revision you already had, when
traversing the reflog ancestry, we may not free() the commit buffer.

Signed-off-by: Johannes Schindelin <Johannes.Schindelin@gmx.de>
Signed-off-by: Junio C Hamano <junkio@cox.net>

--walk-reflogs: actually find the right commit by date.Johannes Schindelin Sat, 20 Jan 2007 09:49:15 +0000 (10:49 +0100)

--walk-reflogs: actually find the right commit by date.

Embarassing thinko.

Signed-off-by: Johannes Schindelin <Johannes.Schindelin@gmx.de>

Fix --walk-reflog with --pretty=onelineJunio C Hamano Sat, 20 Jan 2007 08:51:41 +0000 (00:51 -0800)

Fix --walk-reflog with --pretty=oneline

Now, "git log --abbrev-commit --pretty=o --walk-reflogs HEAD" is
reasonably pleasant to use.

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

reflog-walk: build fixesJunio C Hamano Sat, 20 Jan 2007 08:47:34 +0000 (00:47 -0800)

reflog-walk: build fixes

Dependency on reflog-walk.h was missing in the Makefile, and
reflog-walk.c did not even include it.

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

log --walk-reflog: documentationJunio C Hamano Sat, 20 Jan 2007 07:21:32 +0000 (23:21 -0800)

log --walk-reflog: documentation

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

--walk-reflogs: disallow uninteresting commitsJohannes Schindelin Sat, 20 Jan 2007 02:28:19 +0000 (03:28 +0100)

--walk-reflogs: disallow uninteresting commits

Do not allow uninteresting commits with --walk-reflogs, since it is
not clear what should be shown in these cases:

$ git log --walk-reflogs master..next
$ git log --walk-reflogs ^master

Signed-off-by: Johannes Schindelin <Johannes.Schindelin@gmx.de>

Teach the revision walker to walk by reflogs with ... Johannes Schindelin Thu, 11 Jan 2007 10:47:48 +0000 (11:47 +0100)

Teach the revision walker to walk by reflogs with --walk-reflogs

When called with "--walk-reflogs", as long as there are reflogs
available, the walker will take this information into account, rather
than the parent information in the commit object.

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <junkio@cox.net>

git-rebase: allow rebasing a detached HEAD.Junio C Hamano Sun, 21 Jan 2007 03:11:29 +0000 (19:11 -0800)

git-rebase: allow rebasing a detached HEAD.

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

branch -f: no reason to forbid updating the current... Junio C Hamano Sat, 20 Jan 2007 18:51:37 +0000 (10:51 -0800)

branch -f: no reason to forbid updating the current branch in a bare repo.

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

git-tag -d: allow deleting multiple tags at once.Junio C Hamano Sat, 20 Jan 2007 18:47:41 +0000 (10:47 -0800)

git-tag -d: allow deleting multiple tags at once.

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

Do not verify filenames in a bare repositoryJohannes Schindelin Sat, 20 Jan 2007 02:09:34 +0000 (03:09 +0100)

Do not verify filenames in a bare repository

For example, it makes no sense to check the presence of a file
named "HEAD" when calling "git log HEAD" in a bare repository.

Noticed by Han-Wen Nienhuys.

Signed-off-by: Johannes Schindelin <Johannes.Schindelin@gmx.de>

Stop ignoring Documentation/READMEJunio C Hamano Sat, 20 Jan 2007 07:52:06 +0000 (23:52 -0800)

Stop ignoring Documentation/README

We do not copy this file from elsewhere anymore.

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

apply --cached: fix crash in subdirectoryJohannes Schindelin Sun, 21 Jan 2007 01:17:19 +0000 (02:17 +0100)

apply --cached: fix crash in subdirectory

The static variable "prefix" was shadowed by an unused parameter
of the same name. In case of execution in a subdirectory, the
static variable was accessed, leading to a crash.

Signed-off-by: Johannes Schindelin <Johannes.Schindelin@gmx.de>

show-branch --reflog: fix show_date() callJunio C Hamano Sun, 21 Jan 2007 02:57:06 +0000 (18:57 -0800)

show-branch --reflog: fix show_date() call

Not passing tz to show_date() is not a fix.

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

show_date(): fix relative datesJohannes Schindelin Sat, 20 Jan 2007 21:21:38 +0000 (22:21 +0100)

show_date(): fix relative dates

We pass a timestamp (i.e. number of seconds elapsed since Jan 1 1970,
00:00:00 GMT) to the function. So there is no need to "fix" the
timestamp according to the timezone.

Signed-off-by: Johannes Schindelin <Johannes.Schindelin@gmx.de>

show-branch --reflog: tighten input validation.Junio C Hamano Sat, 20 Jan 2007 06:51:49 +0000 (22:51 -0800)

show-branch --reflog: tighten input validation.

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