Way back when Junio developed the 64 bit index topic he came up
with a means of changing the .idx file format so that older Git
clients would recognize that they don't understand the file and
refuse to read it, while newer clients could tell the difference
between the old-style and new-style .idx files.  Unfortunately
this wasn't recorded anywhere.

This change documents how we might go about changing the .idx
file format by using a special signature in the first four bytes.
Credit (and possible blame) goes completely to Junio for thinking
up this technique.

The change also modifies the error message of the current Git code
so that users get a recommendation to upgrade their Git software
should this version or later encounter a new-style .idx which it
cannot process.  We already do this for the .pack files, but since
we usually process the .idx files first its important that these
files are recognized and encourage an upgrade.
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

The revision specification syntax (sometimes referred to as
SHA1-expressions) is accepted almost everywhere in Git by
almost every tool.  Unfortunately it is only documented in
git-rev-parse.txt, and most users don't know to look there.
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

In ab2a1a32 Junio improved the reflog query logic to support
obtaining the n-th prior value of a ref, but this was never
documented in git-rev-parse.  Now it is.
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

... but we never documented it.
Signed-off-by: NJunio C Hamano <junkio@cox.net>

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

Instead of keeping the confused end user reading low-level
documentation, suggest the higher level commands that implement
what the user may want to do using them upfront.
Signed-off-by: NJunio C Hamano <junkio@cox.net>

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

Current README content is way too esoteric for someone looking at GIT
for the first time. Instead it should provide a quick summary of what
GIT is with a few pointers to other resources.

The bulk of the previous README content is moved to
Documentation/core-intro.txt.
Signed-off-by: NNicolas Pitre <nico@cam.org>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

An incorrect config file can say:

	[format]
		headers

and crash the parsing.
Signed-off-by: NJunio C Hamano <junkio@cox.net>

The default can also be changed with "format.suffix" configuration.
Leaving it empty would not add any suffix.
Signed-off-by: NJunio C Hamano <junkio@cox.net>

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

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

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

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

Add discussion section to git-checkout documentation and mention
detached HEAD in repository-layout document.
Signed-off-by: NJunio C Hamano <junkio@cox.net>

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

Clarify that this is not meant for end users, and list what
shell functions are defined.
Signed-off-by: NJunio C Hamano <junkio@cox.net>

... to match the change we did earlier to git-log documentation.
Signed-off-by: NJunio C Hamano <junkio@cox.net>

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

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

In general list at Wiki seems to be maintained a lot better than
this list.
Signed-off-by: NJunio C Hamano <junkio@cox.net>

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

Also warn about format=flowed (aka 'flawed').
Signed-off-by: NJunio C Hamano <junkio@cox.net>

Take it from the tutorial, since not everybody necessarily reads it.
Signed-off-by: NJunio C Hamano <junkio@cox.net>

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

Talking about what the files contain without talking about where
they are does not help new users.
Signed-off-by: NJunio C Hamano <junkio@cox.net>

This allows "git checkout -m <other-branch>" to notice renames and
carry local changes in the working tree forward.
Signed-off-by: NJunio C Hamano <junkio@cox.net>

... which was fixed since then.
Signed-off-by: NJunio C Hamano <junkio@cox.net>

* Promiscous pull shows the distributed nature of git better.
* Add a new step after that to teach "remote add".
* Highlight that with the shorthand defined you will get
  remote tracking branches for free.
* Fix Alice's workflow.
Signed-off-by: NSanti Béjar <sbejar@gmail.com>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

Then the newbies only have to understand one layout.
Signed-off-by: NSanti Béjar <sbejar@gmail.com>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

From time to time, I would get this error:

[...]
sed: -e expression #8, char 41: Unterminated `s' command
make: *** [git-add--interactive] Error 1

Turns out that the function WriteMakefile() called in Makefile.PL
outputs the message "Writing perl.mak for Git" to stdout! Thus,
the output of "make -C perl -s --no-print-directory instlibdir"
would be prefixed by that message whenever Makefile.PL was newer
than perl.mak.

This is fixed by redirecting stdout to stderr in Makefile.PL.
Signed-off-by: NJohannes E. Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

The example section were talking about the old broken default
behaviour.  Correct it.
Signed-off-by: NJunio C Hamano <junkio@cox.net>

Mention --thin, --no-thin, --repo and -v.
Signed-off-by: NJunio C Hamano <junkio@cox.net>

The text is just copied from git-send-pack.txt.
Signed-off-by: NUwe Kleine-K,Av(Bnig <zeisberg@informatik.uni-freiburg.de>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

People that redirect STDOUT output should always see STDERR
prompts interactively.

STDERR should always be flushed without buffering, so
they should always show up.  If that is unset, we still
explicitly flush by calling STDERR->flush.

The svn command-line client prompts to STDERR, too.
Signed-off-by: NEric Wong <normalperson@yhbt.net>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

The reflog code clears empty directories when rename returns
either EISDIR or ENOTDIR.  Seems to be the only place.
Signed-off-by: NJason Riedy <ejr@cs.berkeley.edu>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

Not all echos know -n.  This was causing a test failure in
t5401-update-hooks.sh, but not t3800-mktag.sh for some reason.
Signed-off-by: NJason Riedy <ejr@cs.berkeley.edu>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

AIX 5.3 seems to need _ALL_SOURCE for struct addrinfo, but that
introduces a struct list in grp.h.
Signed-off-by: NJason Riedy <ejr@cs.berkeley.edu>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

My bash refused to run the two scripts missing a #!, and it's
better to use the same line for all the scripts.
Signed-off-by: NJason Riedy <ejr@cs.berkeley.edu>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

Instead, we complain to the user and suggest that they explicitly
specify the remote and branch. We depend on the exit status of
git-symbolic-ref, so let's go ahead and document that.
Signed-off-by: NJeff King <peff@peff.net>
Signed-off-by: NJunio C Hamano <junkio@cox.net>