* gp/pack-refs-remove-empty-dirs:
  pack-refs: remove newly empty directories

* git://git.bogomips.org/git-svn:
  git svn: fix dcommit to work with touched files
  git svn: add an option to recode pathnames

When printing an error message saying a ref was requested that we do not
have, only print that ref, rather than the ref and everything sent to us
on the same packet line (e.g. protocol support specifications).
Signed-off-by: NElijah Newren <newren@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

The flags --start, --stop, and --restart can be used without the "--".
Document this feature.
Signed-off-by: NJared Hance <jaredhance@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

Run $post_dispatch_hook->() not $pre_dispatch_hook->() after each
request.
Signed-off-by: NJakub Narebski <jnareb@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

The dcommit command fails if an otherwise unmodified file has
been touched in the working directory:

    Cannot dcommit with a dirty index.  Commit your changes
    first, or stash them with `git stash'.

This happens because "git diff-index" reports a difference
between the index and the filesystem:

    :100644 100644 d00491...... 000000...... M      file

The fix is to run "git update-index --refresh" before
"git diff-index" as is done in git-rebase and
git-rebase--interactive before "git diff-files".

This changes dcommit to display a list of modified files before
exiting.

Also add a similar test case for "git svn rebase".

[ew: rearranged commit message subject]
Signed-off-by: NDavid D. Kilzer <ddkilzer@kilzer.net>
Acked-by: NEric Wong <normalperson@yhbt.net>

Teach git-ls-files a new option --debug that just tacks all available
data from the cache onto each file's line.
Signed-off-by: NThomas Rast <trast@student.ethz.ch>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

* maint:
  test-lib: Remove 3 year old no-op --no-python option
  test-lib: Ignore --quiet under a TAP harness
  Documentation/rev-parse: quoting is required with --parseopt
  Documentation: reporting bugs
  Fix git rebase --continue to work with touched files
  Document ls-files -t as semi-obsolete.

* ab/tap:
  test-lib: Remove 3 year old no-op --no-python option
  test-lib: Ignore --quiet under a TAP harness

The --no-python option was added to test-lib.sh by Johannes Schindelin
in early 2006 in abb7c7b3. It was later turned into a no-op by Junio C
Hamano in 7cdbff14 the same year.

Over three years is long enough before removing this old wart which
was retained for backwards compatibility. Our tests have been using
NO_PYTHON and "test_have_prereq PYTHON" for a long time now.
Signed-off-by: NÆvar Arnfjörð Bjarmason <avarab@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

Running the tests with --quiet under a TAP harness will always fail,
since a TAP harness always needs actual test output to go along with
the plan that's being emitted.

Change the test-lib.sh to ignore the --quiet option under
HARNESS_ACTIVE to work around this. Then users that have --quiet in
their GIT_TEST_OPTS can run tests under prove(1) without everything
breaking.
Signed-off-by: NÆvar Arnfjörð Bjarmason <avarab@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

Creating a variable nr here to use throughout the function only to change
refspec_nr to nr at the end, having not used refspec_nr the entire time,
is rather pointless. Instead, simply increment refspec_nr.

While at it, use ALLOC_GROW() instead of xrealloc().
Signed-off-by: NJared Hance <jaredhance@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

When calling rev-parse --parseopt, as in the (now fixed) documented
example

  eval "$(echo "$OPTS_SPEC" | git rev-parse --parseopt -- "$@" || echo exit $?)"

the outermost quoting is required, as otherwise all runs of arbitrary
whitespace inside the resulting 'set -- ...' call would be collapsed
into a single space.

This was exposed as a result of our new use of cat <<\EOF since
47e9cd28 (parseopt: wrap rev-parse --parseopt usage for eval
consumption, 2010-06-12), but has always been a problem when handling
arguments containing e.g. newlines.

Point this out in the documentation, and in particular correct the
example that did not have the quotes.
Noticed-by: NJoshua Jensen <jjensen@workspacewhiz.com>
Signed-off-by: NThomas Rast <trast@student.ethz.ch>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

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

Introduce a new option 'svn.pathnameencoding' that instructs git svn to
recode pathnames to a given encoding.  It can be used by windows users
and by those who work in non-utf8 locales to avoid corrupted file names
with non-ascii characters.

[rp: renamed the option and added manpage documentation]
Signed-off-by: NDmitry Statyvka <dstatyvka@tmsoft-ltd.kiev.ua>
Signed-off-by: NRobert Pollak <robert.pollak@gmail.com>
Acked-by: NEric Wong <normalperson@yhbt.net>

Move highlight config out of guess_file_syntax() so that it can be
extended/overridden by system/user configuration.
Signed-off-by: NAlejandro R. Sedeño <asedeno@mit.edu>
Acked-by: NJakub Narebski <jnareb@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

When performing a non-interactive rebase, sometimes
"git rebase --continue" will fail if an unmodified file is
touched in the working directory:

    You must edit all merge conflicts and then
    mark them as resolved using git add

This is caused by "git diff-files" reporting a difference
between the index and the filesystem:

    :100644 100644 d00491...... 000000...... M	file

The fix is to run "git update-index --refresh" before
"git diff-files" as is done in git-rebase--interactive.
Signed-off-by: NDavid D. Kilzer <ddkilzer@kilzer.net>
Acked-by: NJohannes Schindelin <Johannes.Schindelin@gmx.de>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

The behavior of "git ls-files -t" is very misleading (see
http://thread.gmane.org/gmane.comp.version-control.git/126516 and
http://thread.gmane.org/gmane.comp.version-control.git/144394/focus=144397
for examples of mislead users) and badly documented, hence we point the
users to superior alternatives.

The feature is marked as "semi-obsolete" but not "scheduled for removal"
since it's a plumbing command, scripts might use it, and Git testsuite
already uses it to test the state of the index.
Signed-off-by: NMatthieu Moy <Matthieu.Moy@imag.fr>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

Change the SubmittingPatches recommendations to mention the 50
character soft limit on patch subject lines. 50 characters is the soft
limit mentioned in git-commit(1) and gittutorial(7), it's also the
point at which Gitweb, GitHub and various other Git front ends start
abbreviating the commit message.
Signed-off-by: NÆvar Arnfjörð Bjarmason <avarab@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

The wording of the Signed-off-by rules could be read as stating that
S-O-B should only be added when the submitter considered the patch
ready for inclusion in git.git.

We also want Signed-off-by to be used for e.g. RFC patches, in case
someone wants to dig an old patch out of the archive and improve
it. Change the wording to recommend a Signed-off-by for all submitted
patches.

The problem with the wording came up in the "[PATCH/RFC] Hacky version
of a glob() driven config include" thread[1]. Bert Wesarg suggested[2]
that it be removed to avoid confusion, which this change implements.

  1. <1273180440-8641-1-git-send-email-avarab@gmail.com>
  2. <AANLkTimziTKL13VKIOcaS1TX1F_xvTVjH8Q398Yx36Us@mail.gmail.com>
Suggested-by: NBert Wesarg <bert.wesarg@googlemail.com>
Signed-off-by: NÆvar Arnfjörð Bjarmason <avarab@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

A bit of history in chronological order, the newest at bottom:

- 80ccaa78 (upload-pack: Move the revision walker into a separate function.)
   do_rev_list was introduced with create_full_pack argument

- 21edd3f1 (upload-pack: Run rev-list in an asynchronous function.)
   do_rev_list was now called by start_async, create_full_pack was
   passed by rev_list.data

- f0cea83f (Shift object enumeration out of upload-pack)
   rev_list.data was now zero permanently. Creating full pack was
   done by passing --all to pack-objects

- ae6a5609 (run-command: support custom fd-set in async)
   rev_list.data = 0 was found out redudant and got rid of.

Get rid of the code as well, for less headache while reading do_rev_list.

[jc: noticed by Elijah Newren]
Signed-off-by: NNguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

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

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

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

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

If $HOME is unset (as in some automated build situations),
currently

	git config --path path.home "~"
	git config --path --get path.home

segfaults.  Error out with

	Failed to expand user dir in: '~/'

instead.
Reported-by: NJulien Cristau <jcristau@debian.org>
Signed-off-by: NJonathan Nieder <jrnieder@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

From api-parse-options.txt:

  `description` is a short string to describe the effect of the option.
  It shall begin with a lower-case letter and a full stop (`.`) shall be
  omitted at the end.

It also makes it less confusing if the argument is 'no.' or 'no'.
Signed-off-by: NStephen Boyd <bebarino@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

The message is especially confusing when "git fetch" is ran from "git
pull", for users not aware of "git fetch". The new message makes it clear
that "fetch" means "fetch new revisions", and gives hint on the solution.

We don't add a advice.* configuration option since this message doesn't
appear in normal use, and shouldn't disturb advanced users.
Signed-off-by: NMatthieu Moy <Matthieu.Moy@imag.fr>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

The manpage was added in 1ed6f2c5 (Documentation: gitrevisions,
2010-07-05), but since it does not have a corresponding git command,
it needs an exception for check-docs.
Signed-off-by: NThomas Rast <trast@student.ethz.ch>
Acked-by: NMichael J Gruber <git@drmicha.warpmail.net>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

8b1fa778 (Allow passing of configuration parameters in the command
line, 2010-03-26) forgot the closing ']' for the -c option.

While we're there, also rewrap.  Instead of folding the last two lines
together, try to highlight that COMMAND is required by starting a line
with it.
Signed-off-by: NThomas Rast <trast@student.ethz.ch>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

We document how to run prove with the --state option in t/README. This
produces a .prove YAML file in the current directory. Change the t/
gitignore to ignore it, and clean it up on `make clean'.
Signed-off-by: NÆvar Arnfjörð Bjarmason <avarab@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

Mention the effects of the receive.deny* family of options for the
"remote rejected" case.  While there, also split up the explanation
into an easier-to-parse list format.
Signed-off-by: NThomas Rast <trast@student.ethz.ch>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

This option was introduced by 747ca245 (receive-pack:
receive.denyDeleteCurrent, 2009-02-08) but never documented.
Signed-off-by: NThomas Rast <trast@student.ethz.ch>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

The NULL sentinel argument to the execl*() family of calls must be
cast to (char *), as otherwise:

- platforms where NULL is just 0 (not (void *)) would pass an int

- (admittedly esoteric) platforms where NULL is (void *)0 and (void *)
  and (char *) have different memory layouts would pass the wrong kind
  of pointer
Signed-off-by: NThomas Rast <trast@student.ethz.ch>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

Signed-off-by: NNguyễn Thái Ngọc Duy <pclouds@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

Without additional configuration steps, the documentation build on Cygwin
fails because the XML catalog is missing required rewrites for certain
docbook resources.

This patch documents the required configuration.
Signed-off-by: NJon Seymour <jon.seymour@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

One test case checked the stdout and stderr of 'git add' by constructing a
single 'expect' file that contained both streams. But when the command
runs, the order of stdout and stderr output is unpredictable because it
depends on how the streams are buffered. At least on Windows, the buffering
is different from what the test case expected. Hence, check the two output
texts separately.
Signed-off-by: NJohannes Sixt <j6t@kdbg.org>
Acked-by: NJens Lehmann <Jens.Lehmann@web.de>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

On systems with an echo which defaults to the XSI-conformant behavior
(Solaris, or others using Ksh), echo will interpret certain backslashed
characters as control sequences.  This can cause a problem for interactive
rebase when it is used to rebase commits whose commit "subject" (the first
line) contains any of these backslashed sequences.  In this case, echo will
substitute the control sequence for the backslashed characters and either
the rebased commit message will differ from the original, or the rebase
process will fail.  Neither is desirable.

So work around this issue by replacing the echo statements used to print
out portions of the commit message, with printf.

Also, add a test to test for this breakage.
Signed-off-by: NBrandon Casey <casey@nrlssc.navy.mil>
Acked-by: NJonathan Nieder <jrnieder@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>