df44483a (diff --stat: add config option to limit graph width,
2012-03-01) added the option diff.startGraphWidth to the list of
configuration variables in git-completion.bash, but failed to notice
that the list is sorted alphabetically.  Move it to its rightful place
in the list.
Signed-off-by: NRamkumar Ramachandra <artagnon@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

Signed-off-by: NFelipe Contreras <felipe.contreras@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

Commit d24fbca7 (Remove Git's support for smoke testing - 2011-12-23)
removed the smoke test support from the test suite but it was
re-added by commit 342e9ef2 (Introduce a performance testing
framework - 2012-02-17).  This appears to be the result of a
mis-rebase, since re-adding the smoke testing infrastructure does
not relate to the subject of that commit.

The current 'smoke' target is broken since the 'harness' script it
uses no longer exists, so just reapply this section of commit d24fbca7
and remove all of the smoke testing section in the makefile.
Signed-off-by: NJohn Keeping <john@keeping.me.uk>
Acked-by: NThomas Rast <trast@inf.ethz.ch>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

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

* jk/a-thread-only-dies-once:
  run-command: use thread-aware die_is_recursing routine
  usage: allow pluggable die-recursion checks

* jn/gitweb-install-doc:
  gitweb/INSTALL: GITWEB_CONFIG_SYSTEM is for backward compatibility
  gitweb/INSTALL: Simplify description of GITWEB_CONFIG_SYSTEM

* fc/untracked-zsh-prompt:
  prompt: fix untracked files for zsh

* jk/receive-pack-deadlocks-with-early-failure:
  receive-pack: close sideband fd on early pack errors

* jk/chopped-ident:
  blame: handle broken commit headers gracefully
  pretty: handle broken commit headers gracefully
  cat-file: print tags raw for "cat-file -p"

* rt/commentchar-fmt-merge-msg:
  t6200: avoid path mangling issue on Windows
  fmt-merge-msg: use core.commentchar in tag signatures completely
  fmt-merge-msg: respect core.commentchar in people credits

* rs/empty-archive:
  t5004: fix issue with empty archive test and bsdtar

* pe/pull-rebase-v-q:
  pull: Apply -q and -v options to rebase mode as well

The shell syntax "export X=Y A=B" is not understood by all shells.
Signed-off-by: NTorsten Bögershausen <tboegi@web.de>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

The shell syntax "export X=Y" is not understood by all shells.
Signed-off-by: NTorsten Bögershausen <tboegi@web.de>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

The shell syntax "export X=Y A=B" is not understood by all shells.
Signed-off-by: NTorsten Bögershausen <tboegi@web.de>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

The shell syntax "export X=Y" is not understood by all shells.
Signed-off-by: NTorsten Bögershausen <tboegi@web.de>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

The shell syntax "export X=Y" is not understood by all shells.
Signed-off-by: NTorsten Bögershausen <tboegi@web.de>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

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

"git diff --diff-algorithm=algo" was understood by the command line
parser, but "git diff --diff-algorithm algo" was not.

* jk/diff-algo-finishing-touches:
  diff: allow unstuck arguments with --diff-algorithm
  git-merge(1): document diff-algorithm option to merge-recursive

"git log -S/-G" started paying attention to textconv filter, but
there was no way to disable this.  Make it honor --no-textconv
option.

* sr/log-SG-no-textconv:
  diffcore-pickaxe: unify code for log -S/-G
  diffcore-pickaxe: fix leaks in "log -S<block>" and "log -G<pattern>"
  diffcore-pickaxe: port optimization from has_changes() to diff_grep()
  diffcore-pickaxe: respect --no-textconv
  diffcore-pickaxe: remove fill_one()
  diffcore-pickaxe: remove unnecessary call to get_textconv()

"git merge $(git rev-parse v1.8.2)" behaved quite differently from
"git merge v1.8.2", as if v1.8.2 were written as v1.8.2^0 and did
not pay much attention to the annotated tag payload.  Make the code
notice the type of the tag object, in addition to the dwim_ref()
based classification the current code uses (i.e. the name appears in
refs/tags/) to decide when to special case merging of tags.

* jc/merge-tag-object:
  t6200: test message for merging of an annotated tag
  t6200: use test_config/test_unconfig
  merge: a random object may not necssarily be a commit

Remove one of two consecutive, identical blocks for "git commit -c".

This was caused by a mechanical mismerge at d931e2fb (Merge
branch 'mp/complete-paths', 2013-02-08).  The side branch wanted to
add this block at fea16b47 but the same fix was done independently
at 68539758 already.
Signed-off-by: NMårten Kongstad <marten.kongstad@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

The usage string for cherry-pick and revert has never been updated to
reflect their ability to handle multiple commits. Other documentation is
already correct.
Signed-off-by: NKevin Bracey <kevin@bracey.fi>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

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

* ta/glossary:
  glossary: improve definitions of refspec and pathspec
  The name of the hash function is "SHA-1", not "SHA1"
  glossary: improve description of SHA-1 related topics
  glossary: remove outdated/misleading/irrelevant entries

Improve documentation to illustrate "push authenticated, fetch
anonymous" configuration for smart HTTP servers.

* jk/doc-http-backend:
  doc/http-backend: match query-string in apache half-auth example
  doc/http-backend: give some lighttpd config examples
  doc/http-backend: clarify "half-auth" repo configuration

* jk/test-trash:
  t/test-lib.sh: drop "$test" variable
  t/test-lib.sh: fix TRASH_DIRECTORY handling

* jk/daemon-user-doc:
  doc: clarify that "git daemon --user=<user>" option does not export HOME=~user

* jc/detached-head-doc:
  glossary: extend "detached HEAD" description

Conflicts:
	Documentation/glossary-content.txt

* jk/show-branch-strbuf:
  show-branch: use strbuf instead of static buffer

* js/rerere-forget-protect-against-NUL:
  rerere forget: do not segfault if not all stages are present
  rerere forget: grok files containing NUL

* jc/apply-ws-fix-tab-in-indent:
  test: resurrect q_to_tab
  apply --whitespace=fix: avoid running over the postimage buffer

* ap/combine-diff-ignore-whitespace:
  Allow combined diff to ignore white-spaces

* jk/suppress-clang-warning:
  fix clang -Wtautological-compare with unsigned enum

* tr/perl-keep-stderr-open:
  t9700: do not close STDERR
  perl: redirect stderr to /dev/null instead of closing

* lf/bundle-with-tip-wo-message:
  bundle: Accept prerequisites without commit messages

* jk/filter-branch-come-back-to-original:
  filter-branch: return to original dir after filtering

* rr/prompt-revert-head:
  bash: teach __git_ps1 about REVERT_HEAD

We signal presense of untracked files by adding a per-cent sign '%'
to the prompt.  But because '%' is used as an escape character to
introduce prompt customization in zsh (just like bash prompt uses
'\' to escape '\u', '\h', etc.), we need to say '%%' to get a
literal per-cent.
Helped-by: NAndreas Schwab <schwab@linux-m68k.org>
Signed-off-by: NFelipe Contreras <felipe.contreras@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

Since commit a22e6f85 (receive-pack: send pack-processing
stderr over sideband, 2012-09-21), receive-pack will start
an async sideband thread to copy the stderr from our
index-pack or unpack-objects child to the client. We hand
the thread's input descriptor to unpack(), which puts it in
the "err" member of the "struct child_process".

After unpack() returns, we use finish_async() to reap the
sideband thread. The thread is only ready to die when it
gets EOF on its pipe, which is connected to the err
descriptor. So we expect all of the write ends of that pipe
to be closed as part of unpack().

Normally, this works fine. After start_command forks, it
closes the parent copy of the descriptor. Then once the
child exits (whether it was successful or not), that closes
the only remaining writer.

However, there is one code-path in unpack() that does not
handle this. Before we decide which of unpack-objects or
index-pack to use, we read the pack header ourselves to see
how many objects it contains. If there is an error here, we
exit without running either sub-command, the pipe descriptor
remains open, and we are in a deadlock, waiting for the
sideband thread to die (which is in turn waiting for us to
close the pipe).

We can fix this by making sure that unpack() always closes
the pipe before returning.
Signed-off-by: NJeff King <peff@peff.net>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>