* ts/daemon:
  Added support for dropping privileges to git-daemon.

* jc/apply:
  git-apply --reject: finishing touches.
  apply --reject: count hunks starting from 1, not 0
  git-apply --verbose
  git-apply --reject: send rejects to .rej files.
  git-apply --reject
  apply --reverse: tie it all together.
  diff.c: make binary patch reversible.
  builtin-apply --reverse: two bugfixes.

After a failed "git am" attempt:

	git apply --reject --verbose .dotest/patch

applies hunks that are applicable and leaves *.rej files the
rejected hunks, and it reports what it is doing.  With --index,
files with a rejected hunk do not get their index entries
updated at all, so "git diff" will show the hunks that
successfully got applied.

Without --verbose to remind the user that the patch updated some
other paths cleanly, it is very easy to lose track of the status
of the working tree, so --reject implies --verbose.
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>

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

I noticed that I was looking at the kernel gitweb output at some point
rather than just do "git log", simply because I liked seeing the
simplified date-format, ie the "5 days ago" rather than a full date.

This adds infrastructure to do that for "git log" too. It does NOT add the
actual flag to enable it, though, so right now this patch is a no-op, but
it should now be easy to add a command line flag (and possibly a config
file option) to just turn on the "relative" date format.

The exact cut-off points when it switches from days to weeks etc are
totally arbitrary, but are picked somewhat to avoid the "1 weeks ago"
thing (by making it show "10 days ago" rather than "1 week", or "70
minutes ago" rather than "1 hour ago").

[jc: with minor fix and tweak around "month" and "week" area.]
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

In the Windows world ZIP files are better supported than tar files.
Windows even includes built-in support for ZIP files nowadays.

git-zip-tree is similar to git-tar-tree; it creates ZIP files out of
git trees.  It stores the commit ID (if available) in a ZIP file comment
which can be extracted by unzip.

There's still quite some room for improvement: this initial version
supports no symlinks, calls write() way too often (three times per file)
and there is no unit test.

[jc: with a minor typefix to avoid void* arithmetic]
Signed-off-by: NRene Scharfe <rene.scharfe@lsrfire.ath.cx>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

Excessive use of global variables got me into trouble.
Signed-off-by: NEric Wong <normalperson@yhbt.net>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

Change places that use realloc, without a proper error path, to instead use
xrealloc. Drop an erroneous error path in the daemon code that used errno
in the die message in favour of the simpler xrealloc.
Signed-off-by: NJonas Fonseca <fonseca@diku.dk>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

According to sys/paramh.h it's a "BSD name" for values defined in
<limits.h>. Besides PATH_MAX seems to be more commonly used.
Signed-off-by: NJonas Fonseca <fonseca@diku.dk>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

Change unpack_entry_gently and its helper functions to use offsets
rather than addresses and left counts to supply pack position
information.  In most cases this makes the code easier to follow,
and it reduces the number of local variables in a few functions.
It also better prepares this code for mapping partial segments of
packs and altering what regions of a pack are mapped while unpacking
an entry.
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

If we're always incrementing both the offset and the pointer we
aren't gaining anything by keeping both.  Instead just use the
offset since that's what we were given and what we are expected
to return.  Also using offset is likely to make it easier to remap
the pack in the future should partial mapping of very large packs
get implemented.
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

[PATCH 3/5] Cleanup unpack_entry_gently and friends to use type_name array.

This change allows combining all of the non-delta entries into a
single case, as well as to remove an unnecessary local variable
in unpack_entry_gently.
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

[PATCH 2/5] Reuse compression code in unpack_compressed_entry.

This cleans up the code by reusing a perfectly good decompression
implementation at the expense of 1 extra byte of memory allocated in
temporary memory while the delta is being decompressed and applied
to the base.
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

This function was moved above unpack_delta_entry so we can call it
from within unpack_delta_entry without a forward declaration.

This change looks worse than it is.  Its really just a relocation
of unpack_non_delta_entry to earlier in the file and renaming the
function to unpack_compressed_entry.  No other changes were made.
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

* gl/cleanup:
  Convert memset(hash,0,20) to hashclr(hash).
  Convert memcpy(a,b,20) to hashcpy(a,b).

Signed-off-by: NJakub Narebski <jnareb@gmail.com>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

This is a high-level wrapper around the 'commit-diff' command
and used to produce cleaner history against the mirrored repository
through rebase/reset usage.

It's basically a more polished version of this:

for i in `git rev-list --no-merges remotes/git-svn..HEAD | tac`; do
	git-svn commit-diff $i~1 $i
done
git reset --hard remotes/git-svn
Signed-off-by: NEric Wong <normalperson@yhbt.net>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

Does this make sense to other git-svn users out there?

pull can give funky history unless you understand how git-svn works
internally, which users should not be expected to do.
Signed-off-by: NEric Wong <normalperson@yhbt.net>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

SVN seems to have a problem with https:// repositories from
time-to-time when doing multiple, sequential commits.  This
problem is not consistently reproducible without the patch,
but it should go away entirely with this patch...
Signed-off-by: NEric Wong <normalperson@yhbt.net>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

Signed-off-by: NJonas Fonseca <fonseca@diku.dk>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

Use list continuation to have better wrapping. This accounts for most of
the changes because it reindents a lot of text without applying other
changes.

Use cross-referencing for interlinking and the gitlink macro for pointing
to other tools in the git suite.
Signed-off-by: NJonas Fonseca <fonseca@diku.dk>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

Signed-off-by: NJonas Fonseca <fonseca@diku.dk>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

... so it wraps properly in small terminals.
Signed-off-by: NJonas Fonseca <fonseca@diku.dk>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

Add a short description and document a few selected options additionally to
the different "entities" in the standard calling convention. Advertise
other git repository browsers. Lastly, climb Mount Ego.
Signed-off-by: NJonas Fonseca <fonseca@diku.dk>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

Inspired by the cvs annotate documentation improve and expand the man page
to also mention the limitations of file annotations. Since people coming
from the SVN/CVS world might first look here, also briefly advertise how
the pickaxe interface makes it easy to go beyond these limitation.
Signed-off-by: NJonas Fonseca <fonseca@diku.dk>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

... and mention that '.' will list the local repo references.
Signed-off-by: NJonas Fonseca <fonseca@diku.dk>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

Signed-off-by: NJonas Fonseca <fonseca@diku.dk>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

I often find myself typing this but the common abbreviation "g" for
"again" has not been supported so far for some unknown reason.
Signed-off-by: NJunio C Hamano <junkio@cox.net>

date.c::approxidate_alpha() counts the number of alphabets
while moving the pointer but does not use the count.
Signed-off-by: NPierre Habouzit <madcoder@debian.org>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

builtin-apply.c defines a local variable 'c' which is used only
once and then later gets shadowed by another instance of 'c'.
Signed-off-by: NPierre Habouzit <madcoder@debian.org>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

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

builtin-mv.c and git.c has a nested loop that is governed by a
variable 'i', but they shadow it with another instance of 'i'.
Signed-off-by: NPierre Habouzit <madcoder@debian.org>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

builtin-tar-tree.c::git_tar_config() and http-push.c::add_one_object()
are not used outside their own files.
Signed-off-by: NPierre Habouzit <madcoder@debian.org>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

Signed-off-by: NPierre Habouzit <madcoder@debian.org>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

builtin-apply.c and builtin-push.c uses a local variable called 'error'
which shadows the error() function.
Signed-off-by: NPierre Habouzit <madcoder@debian.org>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

In the same spirit as hashcmp() and hashcpy().
Signed-off-by: NJunio C Hamano <junkio@cox.net>

This abstracts away the size of the hash values when copying them
from memory location to memory location, much as the introduction
of hashcmp abstracted away hash value comparsion.

A few call sites were using char* rather than unsigned char* so
I added the cast rather than open hashcpy to be void*.  This is a
reasonable tradeoff as most call sites already use unsigned char*
and the existing hashcmp is also declared to be unsigned char*.

[jc: Splitted the patch to "master" part, to be followed by a
 patch for merge-recursive.c which is not in "master" yet.

 Fixed the cast in the latter hunk to combine-diff.c which was
 wrong in the original.

 Also converted ones left-over in combine-diff.c, diff-lib.c and
 upload-pack.c ]
Signed-off-by: NShawn O. Pearce <spearce@spearce.org>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

(1 << i) < hspace is compared in the `int` space rather that in the
unsigned one.  the result will be wrong if hspace is between 0x40000000
and 0x80000000.
Signed-off-by: NPierre Habouzit <madcoder@debian.org>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

When an mbox-style patch contains a Cc: line in the header,
git-send-email will check the address against the sender specified
on the command line. If they don't match, sender_not_author will
be set to the address obtained from the Cc line.

When this happens, git-send-email inserts a From: line at the
beginning of the message body with the address obtained from the
Cc line in the header, and the sender might be accused of forging
patch authors.

This patch fixes this by only updating sender_not_author when
processing From: lines, not when processing Cc: lines.
Signed-off-by: NHaavard Skinnemoen <hskinnemoen@atmel.com>
Signed-off-by: NJunio C Hamano <junkio@cox.net>