This fixes a problem in safe_create_leading_directories() when the
argument starts with a '/' (i.e. the path is absolute).
Signed-off-by: NJohannes Schindelin <Johannes.Schindelin@gmx.de>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

Recent FAT workaround caused compilation trouble on OpenBSD;
different platforms use different error codes when we try to
hardlink the temporary file to its final location.  Existing
Coda hack also checks its own error code, but the thing is,
the case we care about is if link failed for a reason other
than that the final file has already existed (which would be
normal, or it could mean collision).  So just check the error
code against EEXIST.
Signed-off-by: NJunio C Hamano <junkio@cox.net>

FAT -- like Coda -- does not like cross-directory hard links. To be
precise, FAT does not like links at all. But links are not needed either.
So get rid of them.
Signed-off-by: NJohannes Schindelin <Johannes.Schindelin@gmx.de>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

If we want to re-pack just local packfiles, we need to know whether a
particular object is local or not.
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

This starts using the "user.name" and "user.email" config variables if
they exist as the default name and email when committing.  This means
that you don't have to use the GIT_COMMITTER_EMAIL environment variable
to override your email - you can just edit the config file instead.

The patch looks bigger than it is because it makes the default name and
email information non-static and renames it appropriately.  And it moves
the common git environment variables into a new library file, so that
you can link against libgit.a and get the git environment without having
to link in zlib and libcrypt.

In short, most of it is renaming and moving, the real change core is
just a few new lines in "git_default_config()" that copies the user
config values to the new base.

It also changes "git-var -l" to list the config variables.
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

The http commit walker cannot use the same temporary file
creation code because it needs to use predictable temporary
filename for partial fetch continuation purposes, but the code
to move the temporary file to the final location should be
usable from the ordinary object creation codepath.

Export move_temp_to_file from sha1_file.c and use it, while
losing the custom relink_or_rename function from http-fetch.c.

Also the temporary object file creation part needs to make sure
the leading path exists, in preparation of the really lazy
fan-out directory creation.
Signed-off-by: NJunio C Hamano <junkio@cox.net>

This makes it possible to have a "sparse" git object subdirectory
structure, something that has become much more attractive now that people
use pack-files all the time.

As a result of pack-files, a git object directory doesn't necessarily have
any individual objects lying around, and in that case it's just wasting
space to keep the empty first-level object directories around: on many
filesystems the 256 empty directories will be aboue 1MB of diskspace.

Even more importantly, after you re-pack a project that _used_ to be
unpacked, you could be left with huge directories that no longer contain
anything, but that waste space and take time to look through.

With this change, "git prune-packed" can just do an rmdir() on the
directories, and they'll get removed if empty, and re-created on demand.

This patch also tries to fix up "write_sha1_from_fd()" to use the new
common infrastructure for creating the object files, closing a hole where
we might otherwise leave half-written objects in the object database.

[jc: I unoptimized the part that really removes the fan-out directories
 to ease transition.  init-db still wastes 1MB of diskspace to hold 256
 empty fan-outs, and prune-packed rmdir()'s the grown but empty directories,
 but runs mkdir() immediately after that -- reducing the saving from 150KB
 to 146KB.  These parts will be re-introduced when everybody has the
 on-demand capability.]
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

This adds more cruft to diff --git header to record the blob SHA1 and
the mode the patch/diff is intended to be applied against, to help the
receiving end fall back on a three-way merge.  The new header looks
like this:

    diff --git a/apply.c b/apply.c
    index 7be5041..8366082 100644
    --- a/apply.c
    +++ b/apply.c
    @@ -14,6 +14,7 @@
     //    files that are being modified, but doesn't apply the patch
     //  --stat does just a diffstat, and doesn't actually apply
    +//  --show-index-info shows the old and new index info for...
    ...

Upon receiving such a patch, if the patch did not apply cleanly to the
target tree, the recipient can try to find the matching old objects in
her object database and create a temporary tree, apply the patch to
that temporary tree, and attempt a 3-way merge between the patched
temporary tree and the target tree using the original temporary tree
as the common ancestor.

The patch lifts the code to compute the hash for an on-filesystem
object from update-index.c and makes it available to the diff output
routine.
Signed-off-by: NJunio C Hamano <junkio@cox.net>

Signed-off-by: NJunio C Hamano <junio@twinsun.com>

The core part detected and died upon seeing a corrupted packfile, but
did not help the user by telling which packfile is corrupt and how.
Signed-off-by: NJunio C Hamano <junkio@cox.net>

Signed-off-by: NSven Verdoolaege <skimo@kotnet.org>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

An entry in the alternates file can name a directory relative to
the object store it describes.  A typical linux-2.6 maintainer
repository would have "../../../torvalds/linux-2.6.git/objects" there,
because the subsystem maintainer object store would live in

    /pub/scm/linux/kernel/git/$u/$system.git/objects/

and the object store of Linus tree is in

    /pub/scm/linux/kernel/git/torvalds/linux-2.6.git/objects/

This unfortunately is different from GIT_ALTERNATE_OBJECT_DIRECTORIES
which is relative to the cwd of the running process, but there is no
way to make it consistent with the behaviour of the environment
variable.  The process typically is run in $system.git/ directory for
a naked repository, or one level up for a repository with a working
tree, so we just define it to be relative to the objects/ directory
to be different from either ;-).

Later, the dumb transport could be updated to read from info/alternates
and make requests for the repository the repository borrows from.
Signed-off-by: NJunio C Hamano <junkio@cox.net>

We have deprecated the old environment variable names for quite a
while and now it's time to remove them.  Gone are:

    SHA1_FILE_DIRECTORIES AUTHOR_DATE AUTHOR_EMAIL AUTHOR_NAME
    COMMIT_AUTHOR_EMAIL COMMIT_AUTHOR_NAME SHA1_FILE_DIRECTORY
Signed-off-by: NJunio C Hamano <junkio@cox.net>

Hi. This patch contains the following possible cleanups:

 * Make some needlessly global functions in local-pull.c static
 * Change 'char *' to 'const char *' where appropriate
Signed-off-by: NPeter Hagervall <hager@cs.umu.se>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

Omitting the first branch in ?: is a GNU extension.  Cute,
but not supported by other compilers.  Replaced mostly
by explicit tests.  Calls to getenv() simply are repeated
on non-GNU compilers.
Signed-off-by: NJason Riedy <ejr@cs.berkeley.edu>

Yes, using the same format for the file and the environment variable
was a big mistake.  This uses LF as the path separator, and allows
lines that begin with '#' to be comments.  ':' is no longer a separator
in objects/info/alternates file.
Signed-off-by: NJunio C Hamano <junkio@cox.net>

Note that the pack file has to be in the usual location if it gets
installed later.
Signed-off-by: NDaniel Barkalow <barkalow@iabervon.org>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

It was a mistake to use GIT_ALTERNATE_OBJECT_DIRECTORIES
environment variable to specify what alternate object pools to
look for missing objects when working with an object database.
It is not a property of the process running the git commands,
but a property of the object database that is partial and needs
other object pools to complete the set of objects it lacks.

This patch allows you to have $GIT_OBJECT_DIRECTORY/info/alternates
whose contents is in exactly the same format as the environment
variable, to let an object database name alternate object pools
it depends on.
Signed-off-by: NJunio C Hamano <junkio@cox.net>

This patch fixes the only warning reported by gcc 4.0.1 on Fedora Core 4
for x86_64:

sha1_file.c:1391: warning: pointer targets in assignment differ in
signedness
Signed-off-by: NPavel Roskin <proski@gnu.org>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

If the object to write was packed, both its uncompressed and compressed
data were leaked.  If the object was not packed, its file was not unmapped.

[jc: I think it still leaks on the write error path of
write_sha1_to_fd(), but that should be fixable in a small separate
patch.]
Signed-off-by: NSergey Vlasov <vsu@altlinux.ru>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

When following a reference, read_object_with_reference() did not free the
intermediate object data.
Signed-off-by: NSergey Vlasov <vsu@altlinux.ru>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

Everybody envies rev-parse, who is the only one that can grok
the extended sha1 format.  Move the get_extended_sha1() out of
rev-parse, rename it to get_sha1() and make it available to
everybody else.

The one I posted earlier to the list had one bug where it did
not handle a name that ends with a digit correctly (it
incorrectly tried the "Nth parent" path).  This commit fixes it.
Signed-off-by: NJunio C Hamano <junkio@cox.net>

This causes ssh-pull to request objects in prefetch() and read then in
fetch(), such that it reduces the unpipelined round-trip time.

This also makes sha1_write_from_fd() support having a buffer of data
which it accidentally read from the fd after the object; this was
formerly not a problem, because it would always get a short read at
the end of an object, because the next object had not been
requested. This is no longer true.
Signed-off-by: NDaniel Barkalow <barkalow@iabervon.org>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

This adds support for reading an uninstalled index, and installing a
pack file that was added while the program was running, as well as
functions for determining where to put the file.
Signed-off-by: NDaniel Barkalow <barkalow@iabervon.org>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

Introduce a new file $GIT_DIR/info/grafts (or $GIT_GRAFT_FILE)
which is a list of "fake commit parent records".  Each line of
this file is a commit ID, followed by parent commit IDs, all
40-byte hex SHA1 separated by a single SP in between.  The
records override the parent information we would normally read
from the commit objects, allowing both adding "fake" parents
(i.e. grafting), and pretending as if a commit is not a child of
some of its real parents (i.e. cauterizing).
Signed-off-by: NJunio C Hamano <junkio@cox.net>

I have reviewed all occurrences of mmap() in git and fixed three types
of errors/defects:

1) The result is not checked.
2) The file descriptor is closed if mmap() succeeds, but not when it
fails.
3) Various casts applied to -1 are used instead of MAP_FAILED, which is
specifically defined to check mmap() return value.

[jc: This is a second round of Pavel's patch.  He fixed up the problem
that close() potentially clobbering the errno from mmap, which
the first round had.]
Signed-off-by: NPavel Roskin <proski@gnu.org>
Signed-off-by: NJunio C Hamano <junkio@cox.net>

This reverses the order of object lookup, to check pack index first and
then go to the filesystem to find .git/objects/??/ hierarchy.

When most of the objects are packed, this saves quite many stat() calls
and negative dcache entries; while the price this approach has to pay is
negligible, even when most of the objects are outside pack, because
checking pack index file is quite cheap.
Signed-off-by: NJunio C Hamano <junkio@cox.net>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Very irritating.  But "snprintf()" wants "char *", and zlib wants
"unsigned char *".

Remove map_sha1_file(), now unused.
Signed-off-by: NDaniel Barkalow <barkalow@iabervon.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Add write_sha1_to_fd(), which writes an object to a file descriptor. This
includes support for unpacking it and recompressing it.
Signed-off-by: NDaniel Barkalow <barkalow@iabervon.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

This patch makes the first half of write_sha1_file() and
index_fd() externally visible, to allow callers to compute the
object ID without actually storing it in the object database.

[JC demangled the whitespaces himself because he liked the patch
 so much, and reworked the interface to index_fd() slightly,
 taking suggestion from Linus and of his own.]
Signed-off-by: NBryan Larsen <bryan.larsen@gmail.com>
Signed-off-by: NJunio C Hamano <junkio@cox.net>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

I'm bored with doing it by hand all the time.

The function write_one_ref() is passed the list of refs received
from the other end, which was obtained by directory traversal
under $GIT_DIR/refs; this can contain paths other than what
git-init-db prepares and would fail to clone when there is
such.
Signed-off-by: NJunio C Hamano <junkio@cox.net>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

The function calls opendir() without a matching closedir().
Signed-off-by: NJunio C Hamano <junkio@cox.net>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

GIT_DIR=.  ends up being what some of the pack senders use, and we
sometimes messed up when cleaning up the path, ie a ".//HEAD" was
cleaned up into "/HEAD", not "HEAD" like it should be.

We should do some other cleanup, and probably also verify that symlinks
don't point to outside the git area.

"git_path()" returns a static pathname pointer into the git directory
using a printf-like format specifier.

"head_ref()" works like "for_each_ref()", except for just the HEAD.

We'll want this for incremental packing.

Mainly making a lot of local functions and variables be marked "static",
but there was a "zero as NULL" warning in there too.

A couple of bogus "might be used undefined" warnings are avoided
by moving the initializations unnecessarily early.

This implements show_pack_info() function used in verify-pack
command when -v flag is used to obtain something like
unpack-objects used to give when it was first written.

It shows the following for each non-deltified object found in
the pack:

    SHA1 type size offset

For deltified objects, it shows this instead:

    SHA1 type size offset depth base_sha1

In order to get the output in the order that appear in the pack
file for debugging purposes, you can do this:

 $ git-verify-pack -v packfile | sort -n -k 4,4
Signed-off-by: NJunio C Hamano <junkio@cox.net>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>