When the user exports a non-default IFS without HT, scripts that
rely on being able to parse "ls-files -s | while read a b c..."
start to fail.  Protect them from such a misconfiguration.

* jc/maint-protect-sh-from-ifs:
  sh-setup: protect from exported IFS

When "git push" triggered the automatic gc on the receiving end, a
message from "git prune" that said it was removing cruft leaked to
the standard output, breaking the communication protocol.

* bc/receive-pack-stdout-protection:
  receive-pack: do not leak output from auto-gc to standard output
  t/t5400: demonstrate breakage caused by informational message from prune

"git diff" had a confusion between taking data from a path in the
working tree and taking data from an object that happens to have
name 0{40} recorded in a tree.

* jk/maint-null-in-trees:
  fsck: detect null sha1 in tree entries
  do not write null sha1s to on-disk index
  diff: do not use null sha1 as a sentinel value

"git send-email" did not unquote encoded words that appear on the
header correctly, and lost "_" from strings.

* tr/maint-send-email-2047:
  send-email: improve RFC2047 quote parsing

When the user gives an argument that can be taken as both a
revision name and a pathname without disambiguating with "--", we
used to give a help message "Use '--' to separate".  The message
has been clarified to show where that '--' goes on the command
line.

* mm/die-with-dashdash-help:
  setup: clarify error messages for file/revisions ambiguity

"gitweb" when used with PATH_INFO failed to notice directories with
SP (and other characters that need URL-style quoting) in them.

* js/gitweb-path-info-unquote:
  gitweb: URL-decode $my_url/$my_uri when stripping PATH_INFO

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

A minor documentation update.

* mv/pull-r-for-rebase:
  man: git pull -r is a short for --rebase

We did not document that many commands take unique prefix
abbreviations of long options (e.g. "--option" may be the only flag
that the command accepts that begin with "--opt", in which case you
can give "--opt") anywhere easy to find for new people.

* jc/maint-abbrev-option-cli:
  gitcli: describe abbreviation of long options

It was unclear what "--topo-order" was really about in the
documentation. It is not just about "children before parent", but
also about "don't mix lineages".

* jc/maint-rev-list-topo-doc:
  rev-list docs: clarify --topo-order description

In earlier days, "imitate the style in the neibouring code" was
sufficient to keep the coherent style, but over time some parts of
the codebase have drifted enough to make it ineffective.

* hv/coding-guidelines:
  Documentation/CodingGuidelines: spell out more shell guidelines

Our documentation used to assume having files in .git/refs/*
directories was the only to have branches and tags, but that is not
true for quite some time.

* jc/tag-doc:
  Documentation: do not mention .git/refs/* directories

Add a compatibility/utility function to the test framework.

* mk/test-seq:
  tests: Introduce test_seq

* lp/no-cmd-http-fetch:
  builtin.h: remove unused cmd_<foo> declarations

* bw/maint-1.7.9-solaris-getpass:
  Enable HAVE_DEV_TTY for Solaris
  terminal: seek when switching between reading and writing

* jk/maint-commit-check-committer-early:
  commit: check committer identity more strictly

Signed-off-by: NEric S. Raymond <esr@thyrsus.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

Signed-off-by: NEric S. Raymond <esr@thyrsus.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

These changes remove all need to modify the ciabot scripts for installation.
Instead, per-project configuration can be dome via variables in a [ciabot]
section of the config file.

Also, correct for the new server address.
Signed-off-by: NEric S. Raymond <esr@thyrsus.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

Letting the "--rebase" option squat on the short-and-sweet single
letter option "-r" was an unintended accident and was not even
documented, but the short option seems to be already used in the
wild. Let's document it so that other options that begin with "r"
would not be tempted to steal it.
Signed-off-by: NMiklos Vajna <vmiklos@suse.cz>
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>

* rj/maint-grep-remove-redundant-test:
  t7810-*.sh: Remove redundant test

* hv/link-alt-odb-entry:
  link_alt_odb_entry: fix read over array bounds reported by valgrind

It was unclear what "--topo-order" was really about in the
documentation.  It is not just about "children before parent", but
also about "don't mix lineages".

Reword the description for both "--date-order" and "--topo-order",
and add an illustration to it.
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

When gitweb is used as a DirectoryIndex, it attempts to strip
PATH_INFO on its own, as $cgi->url() fails to do so.

However, it fails to account for the fact that PATH_INFO has
already been URL-decoded by the web server, but the value
returned by $cgi->url() has not been. This causes the stripping
to fail whenever the URL contains encoded characters.

To see this in action, setup gitweb as a DirectoryIndex and
then use it on a repository with a directory containing a
space in the name. Navigate to tree view, examine the gitweb
generated html and you'll see a link such as:

  <a href="/test.git/tree/HEAD:/directory with spaces">directory with spaces</a>

When clicked on, the browser will URL-encode this link, giving
a $cgi->url() of the form:

   /test.git/tree/HEAD:/directory%20with%20spaces

While PATH_INFO is:

   /test.git/tree/HEAD:/directory with spaces

Fix this by calling unescape() on both $my_url and $my_uri before
stripping PATH_INFO from them.
Signed-off-by: NJay Soffian <jaysoffian@gmail.com>
Acked-by: NJakub Narebski <jnareb@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

In earlier days, "imitate the style in the neibouring code" was
sufficient to keep the coherent style, but over time some parts of
the codebase have drifted enough to make it ineffective.

Spell some of the guidelines out.
Signed-off-by: NHeiko Voigt <hvoigt@hvoigt.net>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

Many scripted Porcelains rely on being able to split words at the
default $IFS characters, i.e. SP, HT and LF.  If the user exports a
non-default IFS to the environment, what they read from plumbing
commands such as ls-files that use HT to delimit fields may not be
split in the way we expect.

Protect outselves by resetting it, just like we do so against CDPATH
exported to the environment.

Noticed by Andrew Dranse <adranse@oanda.com>.
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

The standard output channel of receive-pack is a structured protocol
channel, and subprocesses must never be allowed to leak anything
into it by writing to their standard output.

Use RUN_COMMAND_STDOUT_TO_STDERR option to run_command_v_opt() just
like we do when running hooks to prevent output from "gc" leaking to
the standard output.
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

When receive-pack triggers 'git gc --auto' and 'git prune' is called to
remove a stale temporary object, 'git prune' prints an informational
message to stdout about the file that it will remove.  Since this message
is written to stdout, it is sent back over the transport channel to the git
client which tries to interpret it as part of the pack protocol and then
promptly terminates with a complaint about a protocol error.

Introduce a test which exercises the auto-gc functionality of receive-pack
and demonstrates this breakage.
Signed-off-by: NBrandon Casey <drafnel@gmail.com>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

Now that git_terminal_prompt can cleanly interact with /dev/tty on
Solaris, enable HAVE_DEV_TTY so that this code path is used for
credential reading instead of relying on the crippled getpass().
Signed-off-by: NBen Walton <bwalton@artsci.utoronto.ca>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

When a stdio stream is opened in update mode (e.g., "w+"),
the C standard forbids switching between reading or writing
without an intervening positioning function. Many
implementations are lenient about this, but Solaris libc
will flush the recently-read contents to the output buffer.
In this instance, that meant writing the non-echoed password
that the user just typed to the terminal.

Fix it by inserting a no-op fseek between the read and
write.

The opposite direction (writing followed by reading) is also
disallowed, but our intervening fflush is an acceptable
positioning function for that alternative.
Signed-off-by: NJeff King <peff@peff.net>
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

Hopefully that will be the final 1.7.11.x maintenance release.
Signed-off-by: NJunio C Hamano <gitster@pobox.com>

* jn/block-sha1:
  Makefile: BLK_SHA1 does not require fast htonl() and unaligned loads
  block-sha1: put expanded macro parameters in parentheses
  block-sha1: avoid pointer conversion that violates alignment constraints

* jn/make-assembly-in-right-directory:
  Makefile: fix location of listing produced by "make subdir/foo.s"

* ms/daemon-doc-typo:
  Documentation/git-daemon: add missing word

* lm/git-blame-el:
  git-blame.el: Do not use bare 0 to mean (point-min)
  git-blame.el: Use with-current-buffer where appropriate
  git-blame.el: Do not use goto-line in lisp code

* rs/ipv6-ssh-url:
  git: Wrong parsing of ssh urls with IPv6 literals ignores port

* rs/git-blame-mapcar-mapc:
  git-blame.el: use mapc instead of mapcar

* rr/doc-commit:
  commit: document a couple of options