setup_git_directory_gently: do not barf when GIT_DIR is given.
Earlier we barfed when GIT_DIR environment variable points at a
directory yet to be created, which made it impossible to use
configuration mechanism in "git-init-db".
Signed-off-by: Matthias Lederhofer <matled@gmx.net> Signed-off-by: Junio C Hamano <junkio@cox.net>
* lt/web:
gitweb.cgi: git_blame2: slight optimization reading the blame lines
gitweb.cgi: git_blame2: Revision blocks now have alternating colors
gitweb.cgi: git_blame2: Allow back-trekking through commits
gitweb.cgi: git_blame2: an alternative simple working git blame
* jn/make:
Set datarootdir in config.mak.in
Quote all calls to GIT_CONF_APPEND_LINE
Typofix in configure.ac comment.
configure.ac vertical whitespace usage cleanup
autoconf: Checks for some programs
autoconf: Checks for libraries
autoconf: Checks for some library functions.
autoconf: Checks for typedefs, structures, and compiler characteristics.
autoconf: Preparing the way for autodetection
Copy description of build configuration variables to configure.ac
Teach make clean about configure and autoconf
autoconf: Use autoconf to write installation directories to config.mak.autogen
This changes the calling convention of built-in commands and
passes the "prefix" (i.e. pathname of $PWD relative to the
project root level) down to them.
Signed-off-by: Linus Torvalds <torvalds@osdl.org> Signed-off-by: Junio C Hamano <junkio@cox.net>
Any git command that expects to work in a subdirectory of a project, and
that reads the git config files (which is just about all of them) needs to
make sure that it does the "setup_git_directory()" call before it tries to
read the config file.
This means, among other things, that we need to move the call out of
"init_revisions()", and into the caller.
This does the mostly trivial conversion to do that.
Signed-off-by: Linus Torvalds <torvalds@osdl.org> Signed-off-by: Junio C Hamano <junkio@cox.net>
Display an error from update-ref if target ref name is invalid.
Alex Riesen (raa.lkml@gmail.com) recently observed that git branch
would fail with no error message due to unexpected situations with
regards to refs. For example, if .git/refs/heads/gu is a file but
"git branch -b refs/heads/gu/fixa HEAD" was invoked by the user
it would fail silently due to refs/heads/gu being a file and not
a directory.
This change adds a test for trying to create a ref within a directory
that is actually currently a file, and adds error printing within
the ref locking routine should the resolve operation fail.
The error printing code probably belongs at this level of the library
as other failures within the ref locking, writing and logging code
are also currently at this level of the code.
Signed-off-by: Shawn O. Pearce <spearce@spearce.org> Signed-off-by: Junio C Hamano <junkio@cox.net>
* pb/multi-fetch:
Teach git-http-fetch the --stdin switch
Teach git-local-fetch the --stdin switch
Make pull() support fetching multiple targets at once
Make pull() take some implicit data as explicit arguments
The generated binary patch was _not_ binary -- earlier I made
the --full-index flag to imply binary patch generation to the diff
machinery, but later we made it independent from --binary (although
the latter implies the former).
log and diff family: honor config even from subdirectories
There currently is an unfortunate circular dependency between
what init_revisions (the command line revision specification
parser) does and setting up the log and diff options. The
function uses setup_git_directory() to find the root of the
project relative to the current directory and calls diff_setup()
to prepare diff generation. However, some of the things that
diff_setup() does needs to depend on the configuration variable,
which needs to be read after setup_git_directory() is called.
This patch is a low impact workaround. It first lets
init_revisions() to run and do its thing, then uses git_config()
and diff_setup() after it returns, so that configuration
variables that affects the diff operation can be used from
subdirectories.
Make pull() take some implicit data as explicit arguments
Currently it's a bit weird that pull() takes a single argument
describing the commit but takes the write_ref from a global variable.
This makes it take that as a parameter as well, which might be nicer
for the libification in the future, but especially it will make for
nicer code when we implement pull()ing multiple commits at once.
Signed-off-by: Petr Baudis <pasky@suse.cz> Signed-off-by: Junio C Hamano <junkio@cox.net>
git-update-ref writes into the lockfile, and renames it afterwards. Like
commit v1.3.0-rc3~22, it is not only cleaner, but also helps with shared
setups: every developer can have a different primary group; what matters
is that $GIT_DIR/refs/heads has to be writable by a group you are in.
Signed-off-by: Johannes Schindelin <Johannes.Schindelin@gmx.de> Signed-off-by: Junio C Hamano <junkio@cox.net>
This fixes the builtin mv for the test which Josef provided, and also
fixes moving directories into existing directories, as noted by Jon Smirl.
In case the destination exists, fail early (this cannot be overridden
by -f).
Signed-off-by: Johannes Schindelin <Johannes.Schindelin@gmx.de> Signed-off-by: Junio C Hamano <junkio@cox.net>
git wrapper: add --git-dir=<path> and --bare options
With this, you can say
git --bare repack -a -d
inside a bare repository, and it will actually work. While at it,
also move the --version, --help and --exec-path options to the
handle_options() function.
While at documenting the new options, also document the --paginate
option.
Signed-off-by: Johannes Schindelin <Johannes.Schindelin@gmx.de> Signed-off-by: Junio C Hamano <junkio@cox.net>
Turns out that DBD::SQLite does not favour preparing statements which are
never executed. So, turn all 4 statements, which were prepared _always_,
into methods, like the other 12 prepared statements.
Signed-off-by: Johannes Schindelin <Johannes.Schindelin@gmx.de> Signed-off-by: Junio C Hamano <junkio@cox.net>
This patch defines $state->{prependdir} as the empty string, so that
quite a few warnings are avoided.
Signed-off-by: Johannes Schindelin <Johannes.Schindelin@gmx.de> Acked-by: Martin Langhoff <martin@catalyst.net.nz> Signed-off-by: Junio C Hamano <junkio@cox.net>
I have a tag with a comment which includes an & character. Firefox wouldn't
display my gitweb summary page due to malformed XML. This solves the problem.
Signed-off-by: Daniel Drake <dsd@gentoo.org> Signed-off-by: Junio C Hamano <junkio@cox.net>
* jt/format-patch:
builtin-log: typefix for recent format-patch changes.
Add option to set initial In-Reply-To/References
Add option to enable threading headers
git-format-patch: Make the second and subsequent mails replies to the first
* ew/apply:
Fix t4114 on cygwin
apply: handle type-changing patch correctly.
apply: split out removal and creation into different phases.
apply: check D/F conflicts more carefully.
typechange tests for git apply (currently failing)
gitweb.cgi: git_blame2: Revision blocks now have alternating colors
A revision block is the largest number of adjacent
lines of code originating from the same revision.
This patch adds color to git_blame2(), in that no two
adjacent revision blocks have the same color. The color
alternates between light and dark.
As we annotate the code lines, we alternate the color
(light, dark) of code lines _per revision_. This makes it
easier to see line conglomerations per revision.
Signed-off-by: Luben Tuikov <ltuikov@yahoo.com> Signed-off-by: Junio C Hamano <junkio@cox.net>
gitweb.cgi: git_blame2: Allow back-trekking through commits
This patch adds the capability of back-trekking through
commits from git_blame2() as follows:
blame2->commit->blame2->commit->blame2->...->initial commit.
Signed-off-by: Luben Tuikov <ltuikov@yahoo.com> Signed-off-by: Junio C Hamano <junkio@cox.net>
gitweb.cgi: git_blame2: an alternative simple working git blame
This patch adds an alternative simple working git-blame
called git_blame2(). Simple, because it displays just
three columns: the commit, the line number and the line
of code. Alternative, because the original git_blame()
is left untouched. Lines of code are printed html
escaped, but as-is.
git_blame2() uses git-blame as opposed to git-annotate
used by git_blame().
Signed-off-by: Luben Tuikov <ltuikov@yahoo.com> Signed-off-by: Junio C Hamano <junkio@cox.net>
gitweb.cgi: Include direct link to "raw" files from "history"
In "history" view, the "page_path" is now also a URL link to
the "raw" format of the file, which will always give you
the latest version in the repository.
This is helpful for externally linking files, such that
the latest version is always referenced and in "raw" format.
Signed-off-by: Luben Tuikov <ltuikov@yahoo.com> Signed-off-by: Junio C Hamano <junkio@cox.net>
gitweb.cgi: Teach git_history() to read hash from $hash_base
Teach git_history() to take its hash argument from
the hb parameter, i.e. from $hash_base. Also change
all "a=history" actions to pass "hb=" instead of "h=".
Signed-off-by: Luben Tuikov <ltuikov@yahoo.com> Signed-off-by: Junio C Hamano <junkio@cox.net>
Linus:
get_pathspec() does turn '.' into an empty string (which is
correct - git internally does _not_ ever understand the notion of
"." as the current working directory), but it doesn't ever do the
optimization of noticing that a pathspec that consists solely of
an empty string is "equivalent" to an empty pathspec.
The test is to ensure that this behaviour stays.
Signed-off-by: Alex Riesen <raa.lkml@gmail.com> Signed-off-by: Junio C Hamano <junkio@cox.net>
When paging through the output of git-whatchanged, the color cues help to
visually navigate within a diff. However, it is difficult to notice when a
new commit starts, because the commit and log are shown in the "normal"
color. This patch colorizes the 'commit' line, customizable through
diff.colors.commit and defaulting to yellow.
As a side effect, some of the diff color engine (slot enum, get_color) has
become accessible outside of diff.c.
Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <junkio@cox.net>
# git checkout apply
# git format-patch -k -3 master..conflict | git am -k -3
=> git-am fails with a conflict message
# git reset --hard
# git format-patch -k -3 master..ok | git am -k -3
=> git am fails with the same conflict message as above,
=> since it's trying to apply the old .dotest directory
With the patch it complains about an old .dotest
directory instead.
Signed-off-by: Lukas Sandström <lukass@etek.chalmers.se> Signed-off-by: Junio C Hamano <junkio@cox.net>
When naming commits, stop walking the parent chain as soon as we find
a commit that already has a name. The parent chain of that commit will
be walked later on in any case (or may even have been walked already).
This avoids O(n^2) behavior; on a tree where show-branch displays 6800
commits, the total run time drops from 77 seconds to 5 seconds.
Signed-off-by: Alexandre Julliard <julliard@winehq.org> Signed-off-by: Junio C Hamano <junkio@cox.net>
pack-objects: check pack.window for default window size
For some repositories, deltas simply don't make sense. One can disable
them for git-repack by adding --window, but git-push insists on making
the deltas which can be very CPU-intensive for little benefit.
Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <junkio@cox.net>
It may be desirable for the compiler to disable linking against Fink
or DarwinPorts, especially if both are installed on the system and
the user wants GIT to be linked specifically to only one of them.
Signed-off-by: Shawn O. Pearce <spearce@spearce.org> Signed-off-by: Junio C Hamano <junkio@cox.net>
git.el: Try to reuse an existing buffer when running git-status.
By default, running git-status again will now reuse an existing buffer
that displays the same directory. The old behavior of always creating
a new buffer can be obtained by customizing the git-reuse-status-buffer
option.
Signed-off-by: Alexandre Julliard <julliard@winehq.org> Signed-off-by: Junio C Hamano <junkio@cox.net>
git-svn: fix fetching new directories copies when using SVN:: libs
Log output from SVN doesn't list all the new files that were
added if a new directory was copied from an existing place in
the repository. This means we'll have to do some extra work and
traverse new directories ourselves.
This has been updated from the original patch to defer traversed
adds until all removals have been done. Please disregard the
original.
Thanks to Ben Williamson for the excellent bug report and
testing.
Signed-off-by: Eric Wong <normalperson@yhbt.net> Signed-off-by: Junio C Hamano <junkio@cox.net>
By default, git-tar-tree(1) sets file and directories modes to 0666
or 0777. While this is both useful and acceptable for projects such
as the Linux Kernel, it might be excessive for other projects. With
this variable, it becomes possible to tell git-tar-tree(1) to apply
a specific umask to the modes above. The special value "user"
indicates that the user's current umask will be used. This should be
enough for most projects, as it will lead to the same permissions as
git-checkout(1) would use. The default value remains 0, which means
world read-write.
Signed-off-by: Willy Tarreau <w@1wt.eu> Acked-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <junkio@cox.net>
unpack-objects: remove stale and confusing comment
The very initial version of unpack-objects.c::unpack_all() used
to unpack from the end of the pack, but since end of June last
year it was changed to stream from the front and the comment
does not reflect the reality anymore.
When --keep is specified there is no reason to pass --thin to git-fetch-pack,
which are mutually exclusive. This does not hurt because fetch-pack disables
thin transfer when both are given internally, but still is confusing.
git-diff A...B to (usually) mean "git-diff `git-merge-base A B` B"
This tweaks the argument parser of "git diff" to allow "git-diff
A...B" to show diffs leading to B since their merge-base, when
there is only one sensible merge base between A and B.
Currently nonsense cases are thrown at combined-diff to produce
nonsense results, which would eventually need to be fixed.
* sp/reflog:
Record rebase changes as 'rebase' in the reflog.
Log ref changes made by resolve.
Log ref changes made by quiltimport.
Log ref changes made by git-merge and git-pull.
A type-change diff is always split into a patch to delete old,
immediately followed by a patch to create new. check_patch()
routine noticed that the path to be created already exists in
the working tree and/or in the index when looking at the
creation patch and mistakenly thought it to be an error.
When creating a new file where a directory used to be (or the user had
an empty directory) the code did not check the result from lstat() closely
enough, and mistakenly thought the path already existed in the working tree.
This does not fix the problem where you have a patch that creates a file
at "foo" and removes a file at "foo/bar" (which presumably is the last file
in "foo/" directory in the original). For that, we would need to restructure
write_out_results() loop.
checkout -f failed to check out a file if an existing directory interfered.
When path foo/bar existed in the working tree, checkout -f to switch to
a branch that has a file foo silently did a wrong thing. It failed to
remove the directory foo, did not check out the file foo, and the worst
of all it did not report any errors.
typechange tests for git apply (currently failing)
I've found that git apply is incapable of handling patches
involving object type changes to the same path.
Of course git itself is perfectly capable of making commits that
generate these changes, as it only tracks trees states. It's
just that the diffs between them are less useful if they can't
be applied.
Some of these are rare, but I've hit one of them (file becoming
a symlink) recently in real-world usage, and was inspired to
find more potential breakages :)
I'm not sure when I'll have time to fix these myself and I'm not
very familiar with the apply code. So if someone could get
some or all of these cases working, they would be my hero :)
Some of these are what I would refer to as corner-cases from
hell. Most (if not all) other systems fail some of these. In
fact, they aren't even capable of representing most of these
changes in their histories; much less being able to handle
patches to that effect.
Signed-off-by: Eric Wong <normalperson@yhbt.net> Signed-off-by: Junio C Hamano <junkio@cox.net>
The core function used in show-branch, join_revs(), was supposed
to be exactly the same algorithm as merge_bases(), except that
it was a version enhanced for use with more than two heads.
However, it needed to mark and keep a list of all the commits it
has seen, because it needed them for its semi-graphical output.
The function to implement this list, mark_seen(), stupidly used
insert_by_date(), when it did not need to keep the list sorted
during its processing. This made "show-branch --merge-base"
more than 20x slower compared to "merge-base --all" in some
cases (e.g. between b5032a5 and 48ce8b0 in the Linux 2.6 kernel
archive). The performance of "show-branch --independent"
suffered from the same reason.
This patch sorts the resulting list after the list traversal
just once to fix these problems.
Autoconf 2.60 expresses datadir in terms of datarootdir. If datarootdir
is not substituted, configure issues a warning and uses a compatibility
substitution for datadir.
Signed-off-by: Pavel Roskin <proski@gnu.org> Signed-off-by: Junio C Hamano <junkio@cox.net>