This basically translates the man-page from 'git-developerish' to plain
english, adding some almost-sample output from git-status so users can
recognize what will happen.

Also mention explicitly that --mixed updates the index, while --soft
doesn't. I understood the old text to mean "--mixed is exactly like
--soft, but verbose".
Signed-off-by: NAndreas Ericsson <ae@op5.se>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

Also rearrange some path settings in the Makefile in the process.
Signed-off-by: NRyan Anderson <ryan@michonline.com>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

Make some functions static and convert func() function prototypes to to
func(void).  Fix declaration after statement, missing declaration and
redundant declaration warnings.
Signed-off-by: NTimo Hirvonen <tihirvon@gmail.com>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

Extend the regex syntax of value_regex so that prepending an exclamation
mark means non-match:

	[core]
		quetzal = "Dodo" for Brainf*ck
		quetzal = "T. Rex" for Malbolge
		quetzal = "cat"

You can match the third line with

	git-config-set --get quetzal '! for '
Signed-off-by: NJohannes Schindelin <Johannes.Schindelin@gmx.de>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

Plain except:s are evil as they will catch all kinds of exceptions
including NameError and AttrubiteError.
Signed-off-by: NFredrik Kuivinen <freku045@student.liu.se>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

The flag is universally available, even on VMS; use it.
Signed-off-by: NJunio C Hamano <junkio@cox.net>

Commits that weren't read from git-rev-list, i.e. the ones displayed
with an open circle, were displayed incorrectly: the headline was
null if there was only one line, and the commit comment was put all
on one line.  Also, the terminal commits weren't displayed when -r
was used.
Signed-off-by: NPaul Mackerras <paulus@samba.org>

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

... namely

--replace-all, to replace any amount of matching lines, not just 0 or 1,
--get, to get the value of one key,
--get-all, the multivar version of --get, and
--unset-all, which deletes all matching lines from .git/config
Signed-off-by: NJohannes Schindelin <Johannes.Schindelin@gmx.de>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

Signed-off-by: NAndreas Ericsson <ae@op5.se>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

With this patch, the client side passes identical paths for these two:
	ssh://host.xz/~junio/repo
	host.xz:~junio/repo
Signed-off-by: NAndreas Ericsson <ae@op5.se>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

This patch basically just removes the redundant code from
{receive,upload}-pack.c in favour of the library code in path.c.
Signed-off-by: NAndreas Ericsson <ae@op5.se>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

This should force git-daemon administrator's job a bit harder
because the exact paths need to be given in the whitelist, but
at the same time makes the auditing easier.

This moves validate_symref() from refs.c to path.c, because we
need to link path.c with git-daemon for its "enter_repo()", but
we do not want to link the daemon with the rest of git libraries
and its requirements.
Signed-off-by: NJunio C Hamano <junkio@cox.net>

This patch provides the work-horse of the user-relative paths feature,
using Linus' idea of a blind chdir() and getcwd() which makes it
remarkably simple.
Signed-off-by: NAndreas Ericsson <ae@op5.se>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

... includes the mean tests I mentioned on the list.
Signed-off-by: NJohannes Schindelin <Johannes.Schindelin@gmx.de>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

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

This is meant for the end user, who cannot be expected to edit
.git/config by hand.

Example:

	git-config-set core.filemode true

will set filemode in the section [core] to true,

	git-config-set --unset core.filemode

will remove the entry (failing if it is not there), and

	git-config-set --unset diff.twohead ^recar

will remove the unique entry whose value matches the regex "^recar"
(failing if there is no unique such entry).

It is just a light wrapper around git_config_set() and
git_config_set_multivar().
Signed-off-by: NJohannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

The function git_config_set() does exactly what you think it does.
Given a key (in the form "core.filemode") and a value, it sets the
key to the value. Example:

	git_config_set("core.filemode", "true");

The function git_config_set_multivar() is meant for setting variables which
can have several values for the same key. Example:

	[diff]
		twohead = resolve
		twohead = recarsive

the typo in the second line can be replaced by

	git_config_set_multivar("diff.twohead", "recursive", "^recar");

The third argument of the function is a POSIX extended regex which has to
match the value. If there is no key/value pair with a matching value, a new
key/value pair is added.

These commands are also capable of unsetting (deleting) entries:

	git_config_set_multivar("diff.twohead", NULL, "sol");

will delete the entry

		twohead = resolve
Signed-off-by: NJohannes Schindelin <Johannes.Schindelin@gmx.de>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

The decision about whether to build http-push or not belongs in the
Makefile.  This follows Junio's suggestion to determine whether curl
is new enough to support http-push.
Signed-off-by: NNick Hengeveld <nickh@reactrix.com>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

Move the static function get_curl_handle() around to make sure
its definition and declarations are seen by the compiler before
its first use.  Also remove an unused variable.
Signed-off-by: NJunio C Hamano <junkio@cox.net>

Clean up memory and file descriptor usage
Signed-off-by: NNick Hengeveld <nickh@reactrix.com>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

Improved XML parsing - replace specialized doc parser callbacks with generic
functions that track the parser context and use document-specific callbacks
to process that data.
Signed-off-by: NNick Hengeveld <nickh@reactrix.com>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

Better response handling for pack list requests - a 404 means we do have
the list but it happens to be empty.
Signed-off-by: NNick Hengeveld <nickh@reactrix.com>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

Rename object request functions and data to make it more clear which type
of request is being processed - this is a response to the introduction of
slot callbacks and the definition of different types of requests such as
alternates_request.
Signed-off-by: NNick Hengeveld <nickh@reactrix.com>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

Move shared HTTP request functionality out of http-fetch and http-push,
and replace the two fwrite_buffer/fwrite_buffer_dynamic functions with
one fwrite_buffer function that does dynamic buffering.  Use slot
callbacks to process responses to fetch object transfer requests and
push transfer requests, and put all of http-push into an #ifdef check
for curl multi support.
Signed-off-by: NNick Hengeveld <nickh@reactrix.com>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

When the last file in a directory is removed as the result of a
merge, try to rmdir the now-empty directory.
Signed-off-by: NJunio C Hamano <junkio@cox.net>

When the last file in a directory is removed as the result of a
merge, try to rmdir the now-empty directory.

[jc: We probably could use "rmdir -p", but for now we do that by
hand for portability.]
Signed-off-by: NJunio C Hamano <junkio@cox.net>

git-commit -v flag has been the default for quite some time, so
do not mention it.  Also a typofix.
Signed-off-by: NJunio C Hamano <junkio@cox.net>

With "-a", redundant pack removal is trivial, and otherwise
redundant pack removal is pointless; do not call
git-redundant-pack from this script.
Signed-off-by: NJunio C Hamano <junkio@cox.net>

Johannes suggested this earlier but I did not take it so
seriously because this command is not that important.  But this
probably matters on Cygwin which does not seem to come with
precompiled dc.  It is a mystery for me that anything that
mimics UNIX does not offer a dc, though.

I did the detection for the lack of dc command a bit differently
from the verison Johannes did.
Signed-off-by: NJunio C Hamano <junkio@cox.net>

Signed-off-by: NStefan-W. Hahn <stefan.hahn@s-hahn.de>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

gitk switched to use git-diff-tree with one argument in gettreediffs and
getblobdiffs.  git-diff-tree with one argument outputs commit ID in from
of the patch.  This causes an empty line after "Comments" in the lower
right pane.  Also, the diff in the lower left pane has the commit ID,
which is useless there.

This patch makes git use the newly added -no-commit-id option for
git-diff-tree to suppress commit ID.  It also removes the p variable in
both functions, since it has become useless after switching to the
one-argument invocation for git-diff-tree.
Signed-off-by: NPavel Roskin <proski@gnu.org>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

Hovering over a line in gitk displays the commit one-liner in a
box, but the text usually overflows the box.  The box size is
computed with a specified font, so this patch sets the text font
as well.
Signed-off-by: NFrank Sorenson <frank@tuxrocks.com>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

Instead of reading refs/heads/* and refs/tags/* files ourselves
and missing files in subdirectories of heads/ and tags/, use
ls-remote on local repository and grab all of them.  This lets us
also remove the procedure readotherrefs.
Signed-off-by: NJunio C Hamano <junkio@cox.net>

The current rebase implementation finds commits in our tree but
not in the upstream tree using git-cherry, and tries to apply
them using git-cherry-pick (i.e. always use 3-way) one by one.

Which is fine, but when some of the changes do not apply
cleanly, it punts, and punts badly.

Suppose you have commits A-B-C-D-E since you forked from the
upstream and submitted the changes for inclusion.  You fetch
from upstream head U and find that B has been picked up.  You
run git-rebase to update your branch, which tries to apply
changes contained in A-C-D-E, in this order, but replaying of C
fails, because the upstream got changes that touch the same area
from elsewhere.

Now what?

It notes that fact, and goes ahead to apply D and E, and at the
very end tells you to deal with C by hand.  Even if you somehow
managed to replay C on top of the result, you would now end up
with ...-B-...-U-A-D-E-C.

Breaking the order between B and others was the conscious
decision made by the upstream, so we would not worry about it,
and even if it were worrisome, it is too late for us to fix now.
What D and E do may well depend on having C applied before them,
which is a problem for us.

This rewrites rebase to use git-format-patch piped to git-am,
and when the patch does not apply, have git-am fall back on
3-way merge.  The updated diff/patch pair knows how to apply
trivial binary patches as long as the pre- and post-images are
locally available, so this should work on a repository with
binary files as well.

The primary benefit of this change is that it makes rebase
easier to use when some of the changes do not replay cleanly.
In the "unapplicable patch in the middle" case, this "rebase"
works like this:

 - A series of patches in e-mail form is created that records
   what A-C-D-E do, and is fed to git-am.  This is stored in
   .dotest/ directory, just like the case you tried to apply
   them from your mailbox.  Your branch is rewound to the tip of
   upstream U, and the original head is kept in .git/ORIG_HEAD,
   so you could "git reset --hard ORIG_HEAD" in case the end
   result is really messy.

 - Patch A applies cleanly.  This could either be a clean patch
   application on top of rewound head (i.e. same as upstream
   head), or git-am might have internally fell back on 3-way
   (i.e.  it would have done the same thing as git-cherry-pick).
   In either case, a rebased commit A is made on top of U.

 - Patch C does not apply.  git-am stops here, with conflicts to
   be resolved in the working tree.  Yet-to-be-applied D and E
   are still kept in .dotest/ directory at this point.  What the
   user does is exactly the same as fixing up unapplicable patch
   when running git-am:

   - Resolve conflict just like any merge conflicts.
   - "git am --resolved --3way" to continue applying the patches.

 - This applies the fixed-up patch so by definition it had
   better apply.  "git am" knows the patch after the fixed-up
   one is D and then E; it applies them, and you will get the
   changes from A-C-D-E commits on top of U, in this order.

I've been using this without noticing any problem, and as people
may know I do a lot of rebases.
Signed-off-by: NJunio C Hamano <junkio@cox.net>

A new usage, 'git-branch -f branch [start]', resets the branch head at
start (or current head).  Should be considered a dangerous operation,
but if you are like me to keep rewinding branches it is handy.
Signed-off-by: NJunio C Hamano <junkio@cox.net>

Truncate the result from readdir() in the exec-path if they end
with .exe, to make it a bit more readable on Cygwin.
Signed-off-by: NJunio C Hamano <junkio@cox.net>