rebase-i-p: do not include non-first-parent commits touching UPSTREAM
This covers an odd boundary case found by Avi Kivity's script where a branch
coming off of UPSTREAM is merged into HEAD. Initially it show up in
UPSTREAM..HEAD, but technically UPSTREAM is not moving, the rest of head is, so
we should not need to rewrite the merge.
This adds a check saying we can keep `preserve=t` if `p=UPSTREAM`...unless this
is the first first-parent commit in our UPSTREAM..HEAD rev-list, which could
very well point to UPSTREAM, but we still need to consider it as rewritten so we
start pulling in the rest of the UPSTREAM..HEAD commits that point to it.
Signed-off-by: Stephen Haberman <stephen@exigencecorp.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
rebase-i-p: only list commits that require rewriting in todo
This is heavily based on Stephan Beyer's git sequencer rewrite of rebase-i-p.
Each commit is still found by rev-list UPSTREAM..HEAD, but a commit is only
included in todo if at least one its parents has been marked for rewriting.
Signed-off-by: Stephen Haberman <stephen@exigencecorp.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
rebase-i-p: fix 'no squashing merges' tripping up non-merges
Also only check out the first parent if this commit if not a squash--if it is a
squash, we want to explicitly ignore the parent and leave the wc as is, as
cherry-pick will apply the squash on top of it.
Signed-off-by: Stephen Haberman <stephen@exigencecorp.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
rebase-i-p: delay saving current-commit to REWRITTEN if squashing
If the current-commit was dumped to REWRITTEN, but then we squash the next
commit in to it, we have invalidated the HEAD was just written to REWRITTEN.
Instead, append the squash hash to current-commit and save both of them the next
time around.
Signed-off-by: Stephen Haberman <stephen@exigencecorp.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
rebase-i-p: use HEAD for updating the ref instead of mapping OLDHEAD
If OLDHEAD was reordered in the todo, and its mapped NEWHEAD was used to set the
ref, commits reordered after OLDHEAD in the todo would should up as un-committed
changes.
Signed-off-by: Stephen Haberman <stephen@exigencecorp.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
rebase--interactive: fix parent rewriting for dropped commits
`rebase -i -p` got its rev-list of commits to keep by --left-right and
--cherry-pick. Adding --cherry-pick would drop commits that duplicated changes
already in the rebase target.
The dropped commits were then forgotten about when it came to rewriting the
parents of their descendents, so the descendents would get cherry-picked with
their old, unwritten parents and essentially make the rebase a no-op.
This commit adds a $DOTEST/dropped directory to remember dropped commits and
rewrite their children's parent as the dropped commit's possibly-rewritten
first-parent.
Signed-off-by: Stephen Haberman <stephen@exigencecorp.com> Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
remote.c: correct the check for a leading '/' in a remote name
This test is supposed to disallow remote entries in the config file of the
form:
[remote "/foobar"]
...
The leading slash in '/foobar' is not acceptable.
Instead it was incorrectly testing that the subkey had no leading '/', which
had no effect since the subkey pointer was made to point at a '.' in the
preceding lines.
Signed-off-by: Brandon Casey <casey@nrlssc.navy.mil> Acked-by: Daniel Barkalow <barkalow@iabervon.org> Signed-off-by: Junio C Hamano <gitster@pobox.com>
We carefully verify that the input to git-apply is sane,
including cross-checking that the filenames we see in "+++"
headers match what was provided on the command line of "diff
--git". When --directory is used, however, we ended up
comparing the unadorned name to one with the prepended root,
causing us to complain about a mismatch.
We simply need to prepend the root directory, if any, when
pulling the name out of the git header.
Signed-off-by: Jeff King <peff@peff.net> Acked-by: Junio C Hamano <gitster@pobox.com> Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
rebase -i: do not fail when there is no commit to cherry-pick
In case there is no commit to apply (for example because you rebase to
upstream and all your local patches have been applied there), do not
fail. The non-interactive rebase already behaves that way.
Do this by introducing a new command, "noop", which is substituted for
an empty commit list, so that deleting the commit list can still abort
as before.
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
Since v1.6.0.2~13^2~ the completion of a thin pack uses sha1write() for
its ability to compute a SHA1 on the written data. This also provides
data buffering which, along with commit 92392b4a45, will confuse pread()
whenever an appended object is 1) freed due to memory pressure because
of the depth-first delta processing, and 2) needed again because it has
many delta children, and 3) its data is still buffered by sha1write().
Let's fix the issue by simply forcing cached data out when such an
object is written so it can be pread()'d at leisure.
Signed-off-by: Nicolas Pitre <nico@cam.org> Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
git-stash.sh: fix flawed fix of invalid ref handling (commit da65e7c1)
The referenced commit tried to fix a flaw in stash's handling of a user
supplied invalid ref. i.e. 'git stash apply fake_ref@{0}' should fail
instead of applying stash@{0}. But, it did so in a naive way by avoiding the
use of the --default option of rev-parse, and instead manually supplied the
default revision if the user supplied an empty command line. This prevented
a common usage scenario of supplying flags on the stash command line (i.e.
non-empty command line) which would be parsed by lower level git commands,
without supplying a specific revision. This should fall back to the default
revision, but now it causes an error. e.g. 'git stash show -p'
The correct fix is to use the --verify option of rev-parse, which fails
properly if an invalid ref is supplied, and still allows falling back to a
default ref when one is not supplied.
Convert stash-drop to use --verify while we're at it, since specifying
multiple revisions for any of these commands is also an error and --verify
makes it so.
Signed-off-by: Brandon Casey <casey@nrlssc.navy.mil> Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
But while we're at it, change the allocation to reference the
variable it is allocating memory for to try to prevent a similar
mistake, for example if the type is changed, in the future.
If we use pread() while at the end of the file, it will return 0, which is
not an error from the operating system point of view. In this case, errno
has not been set and must not be used.
Signed-off-by: Samuel Tardieu <sam@rfc1149.net> Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
If core.bare or core.sharedRepository are set in /etc/gitconfig or
~/.gitconfig, then 'git init' will read the values when constructing a
new config file; reading them, however, will override the values
specified on the command line. In the case of --bare, this ends up
causing a segfault, without the repository being properly initialised;
in the case of --shared, the permissions are set according to the
existing config settings, not what was specified on the command line.
This fix saves any specified values for --bare and --shared prior to
reading existing config settings, and restores them after reading but
before writing the new config file. core.bare is ignored in all
situations, while core.sharedRepository will only be used if --shared
is not specified to git init.
Also includes testcases which use a specified global config file
override, demonstrating the former failure scenario.
Signed-off-by: Deskin Miller <deskinm@umich.edu> Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
git-push.txt: Describe --repo option in more detail
The --repo option was described in a way that the reader would have to
assume that it is the same as the <repository> parameter. But it actually
servers a purpose, which is now written down.
Furthermore, the --mirror option was missing from the synopsis.
Signed-off-by: Johannes Sixt <johannes.sixt@telecom.at> Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
The original git-rebase honored pre-rebase hook so that public branches
can be protected from getting rebased, but rebase --interactive ignored
the hook entirely. This fixes it.
Signed-off-by: Nanako Shiraishi <nanako3@lavabit.com> Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
There are 9 places where prefix_path is called, and only in one of
them the returned pointer was checked to be non-zero and only to
call exit(128) as it is usually done by die(). In other 8 places,
the returned value was not checked and it caused SIGSEGV when a
path outside of the working tree was used. For instance, running
git update-index --add /some/path/outside
caused SIGSEGV.
This patch changes prefix_path() to die if the path is outside of
the repository, so it never returns NULL.
Signed-off-by: Dmitry Potapov <dpotapov@gmail.com> Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
fix bogus "diff --git" header from "diff --no-index"
When "git diff --no-index" is given an absolute pathname, it
would generate a diff header with the absolute path
prepended by the prefix, like:
diff --git a/dev/null b/foo
Not only is this nonsensical, and not only does it violate
the description of diffs given in git-diff(1), but it would
produce broken binary diffs. Unlike text diffs, the binary
diffs don't contain the filenames anywhere else, and so "git
apply" relies on this header to figure out the filename.
This patch just refuses to use an invalid name for anything
visible in the diff.
Now, this fixes the "git diff --no-index --binary a
/dev/null" kind of case (and we'll end up using "a" as the
basename), but some other insane cases are impossible to
handle. If you do
git diff --no-index --binary a /bin/echo
you'll still get a patch like
diff --git a/a b/bin/echo
old mode 100644
new mode 100755
index ...
and "git apply" will refuse to apply it for a couple of
reasons, and the diff is simply bogus.
And that, btw, is no longer a bug, I think. It's impossible
to know whethe the user meant for the patch to be a rename
or not. And as such, refusing to apply it because you don't
know what name you should use is probably _exactly_ the
right thing to do!
Original problem reported by Imre Deak. Test script and problem
description by Jeff King.
Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
This breaks my build on Solaris 8, as there is no separate
libiconv.
The history of this line is somewhat convoluted. In 2fd955c
(in November 2005), NEEDS_LIBICONV was turned on for all
Solaris builds, claiming to "fix an error in Solaris 10 by
setting NEEDS_LIBICONV".
Later, e15f545 (in February of 2006) claimed that "Solaris
9+ don't need iconv", and moved NEEDS_LIBICONV into a
section for Solaris 8.
that he does not set NEEDS_LIBICONV for Solaris 7.
So either one of those commits is totally wrong, or there is
some other magic going on where some Solaris installs need
it and others don't.
Given Brandon's statement and my problems on Solaris 8 with
NEEDS_LIBICONV, I am inclined to think the first commit was
bogus, and that NEEDS_LIBICONV shouldn't be set for Solaris
at all by default. If somebody wants to use iconv and has
installed it manually, they can set it in their config.mak.
Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
rebase -i: proper prepare-commit-msg hook argument when squashing
One would expect that the prepare-commit-msg hook gets 'squash' as the
second argument when squashing commits with 'rebase -i'. However,
that was not the case, as it got 'merge' instead. This patch fixes
the problem.
Signed-off-by: SZEDER Gábor <szeder@ira.uka.de> Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
gitweb: Add path_info tests to t/t9500-gitweb-standalone-no-errors.sh
Note that those tests only check that there are no errors nor
warnings from Perl; they do not check for example if gitweb doesn't
use ARRAY(0x8e3cc20) instead of correct value in links, etc.
Signed-off-by: Jakub Narebski <jnareb@gmail.com> Acked-by: Petr Baudis <pasky@suse.cz> Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
gitweb: Fix two 'uninitialized value' warnings in git_tree()
If we did try to access nonexistent directory or file, which means
that git_get_hash_by_path() returns `undef`, uninitialized $hash
variable was passed to 'open' call. Now we fail early with "404 Not
Found - No such tree" error. (If we try to access something which
does not resolve to tree-ish, for example a file / 'blob' object, the
error will be caught later, as "404 Not Found - Reading tree failed"
error).
If we tried to use 'tree' action without $file_name ('f' parameter)
set, which means either tree given by hash or a top tree (and we
currently cannot distinguish between those two cases), we cannot print
path breadcrumbs with git_print_page_path(). Fix this by moving call
to git_print_page_path() inside conditional.
Signed-off-by: Jakub Narebski <jnareb@gmail.com> Acked-by: Petr Baudis <pasky@suse.cz> Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
This patch fixes PATH_INFO handling by removing the relevant part from
$my_url and $my_uri, thus making it unnecessary to specify them by hand
in the gitweb configuration.
Signed-off-by: Giuseppe Bilotta <giuseppe.bilotta@gmail.com> Acked-by: Jakub Narebski <jnareb@gmail.com> Acked-by: Petr Baudis <pasky@suse.cz> Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
Documentation: remove '\' in front of short options
... because they show up in the man and html outputs.
This escaping is only needed for double dashes to be compatible with
older asciidoc versions; see commit e1ccf53 ([PATCH] Escape asciidoc's
built-in em-dash replacement, 2005-09-12).
Signed-off-by: SZEDER Gábor <szeder@ira.uka.de> Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
When doing a set-tree and there is no revision to commit to, the following unrelated error message is displayed: "Undefined subroutine &Git::SVN::fatal called at /opt/local/libexec/git-core/git-svn line 2575." The following patch fixes the problem and allows the real error message to be shown.
Signed-off-by: Luc Heinrich <luc@honk-honk.com> Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
Merge branch 'bc/maint-diff-hunk-header-fix' into maint
* bc/maint-diff-hunk-header-fix:
t4018-diff-funcname: test syntax of builtin xfuncname patterns
diff hunk pattern: fix misconverted "\{" tex macro introducers
diff: use extended regexp to find hunk headers
diff.*.xfuncname which uses "extended" regex's for hunk header selection
diff.c: associate a flag with each pattern and use it for compiling regex
diff.c: return pattern entry pointer rather than just the hunk header pattern
Currently, trying to use git-commit with unmerged files in the index
will show the message "Error building trees", which can be a bit
obscure to the end user. This patch makes the error message clearer, and
consistent with what git-write-tree reports in a similar situation.
Signed-off-by: Rafael Garcia-Suarez <rgarciasuarez@gmail.com> Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
Add remove_path: a function to remove as much as possible of a path
The function has two potential users which both managed to get wrong
their implementations (the one in builtin-rm.c one has a memleak, and
builtin-merge-recursive.c scribles over its const argument).
Signed-off-by: Alex Riesen <raa.lkml@gmail.com> Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
git-submodule: Fix "Unable to checkout" for the initial 'update'
Since commit 55218("checkout: do not lose staged removal"), in
cmd_add/cmd_update, "git checkout <commit>" following
"git clone -n" may fail if <commit> is different from HEAD.
So Use "git checkout -f <commit>" to fix this.
Signed-off-by: Ping Yin <pkufranky@gmail.com> Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
Documentation: clarify the details of overriding LESS via core.pager
The process of overriding the default LESS options using only
git-specific methods is rather obscure. Show the end user how
to do it in a step-by-step manner.
Signed-off-by: Chris Frey <cdfrey@foursquare.net> Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
git-stash.sh: don't default to refs/stash if invalid ref supplied
apply_stash() and show_stash() each call rev-parse with
'--default refs/stash' as an argument. This option causes rev-parse to
operate on refs/stash if it is not able to successfully operate on any
element of the command line. This includes failure to supply a "valid"
revision. This has the effect of causing 'stash apply' and 'stash show'
to operate as if stash@{0} had been supplied when an invalid revision is
supplied.
e.g. 'git stash apply stash@{1}' would fall back to
'git stash apply stash@{0}'
This patch modifies these two functions so that they avoid using the
--default option of rev-parse.
Signed-off-by: Brandon Casey <casey@nrlssc.navy.mil> Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
maint: check return of split_cmdline to avoid bad config strings
As the testcase demonstrates, it's possible for split_cmdline to return -1 and
deallocate any memory it's allocated, if the config string is missing an end
quote. In both the cases below, which are the only calling sites, the return
isn't checked, and using the pointer causes a pretty immediate segfault.
Signed-off-by: Deskin Miller <deskinm@umich.edu> Acked-by: Miklos Vajna <vmiklos@frugalware.org> Signed-off-by: Shawn O. Pearce <spearce@spearce.org>
Do not perform cross-directory renames when creating packs
A comment on top of create_tmpfile() describes caveats ('can have
problems on various systems (FAT, NFS, Coda)') that should apply
in this situation as well. This in the end did not end up solving
any of my personal problems, but it might be a useful cleanup patch
nevertheless.
Signed-off-by: Petr Baudis <pasky@suse.cz> Acked-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Junio C Hamano <gitster@pobox.com>
git-remote: do not use user input in a printf format string
'git remote show' substituted the remote name into a string that was later
used as a printf format string. If a remote name contains a printf format
specifier like this:
$ git remote add foo%sbar .
then the command
$ git remote show foo%sbar
would print garbage (if you are lucky) or crash. This fixes it.
Signed-off-by: Johannes Sixt <johannes.sixt@telecom.at> Signed-off-by: Junio C Hamano <gitster@pobox.com>
This patch makes "git remote -v" and "git remote show" report multiple URLs
rather than warn about them. Multiple URLs are OK for pushing into
multiple repos simultaneously. Without "-v" each repo is shown once only.
Signed-off-by: Michael J Gruber <git@drmicha.warpmail.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
diff.*.xfuncname which uses "extended" regex's for hunk header selection
Currently, the hunk headers produced by 'diff -p' are customizable by
setting the diff.*.funcname option in the config file. The 'funcname' option
takes a basic regular expression. This functionality was designed using the
GNU regex library which, by default, allows using backslashed versions of
some extended regular expression operators, even in Basic Regular Expression
mode. For example, the following characters, when backslashed, are
interpreted according to the extended regular expression rules: ?, +, and |.
As such, the builtin funcname patterns were created using some extended
regular expression operators.
Other platforms which adhere more strictly to the POSIX spec do not
interpret the backslashed extended RE operators in Basic Regular Expression
mode. This causes the pattern matching for the builtin funcname patterns to
fail on those platforms.
Introduce a new option 'xfuncname' which uses extended regular expressions,
and advertise it _instead_ of funcname. Since most users are on GNU
platforms, the majority of funcname patterns are created and tested there.
Advertising only xfuncname should help to avoid the creation of non-portable
patterns which work with GNU regex but not elsewhere.
Additionally, the extended regular expressions may be less ugly and
complicated compared to the basic RE since many common special operators do
not need to be backslashed.
For example, the GNU Basic RE:
^[ ]*\\(\\(public\\|static\\).*\\)$
becomes the following Extended RE:
^[ ]*((public|static).*)$
Signed-off-by: Brandon Casey <casey@nrlssc.navy.mil> Signed-off-by: Junio C Hamano <gitster@pobox.com>
diff.c: return pattern entry pointer rather than just the hunk header pattern
This is in preparation for associating a flag with each pattern which will
control how the pattern is interpreted. For example, as a basic or extended
regular expression.
Signed-off-by: Brandon Casey <casey@nrlssc.navy.mil> Signed-off-by: Junio C Hamano <gitster@pobox.com>
sha1_file: link() returns -1 on failure, not errno
5723fe7 (Avoid cross-directory renames and linking on object creation,
2008-06-14) changed the call to use link() directly instead of through a
custom wrapper, but forgot that it returns 0 or -1, not 0 or errno.
Signed-off-by: Thomas Rast <trast@student.ethz.ch> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Make git archive respect core.autocrlf when creating zip format archives
There is currently no call to git_config at the start of cmd_archive.
When creating tar archives the core config is read as a side-effect of
reading the tar specific config, but this doesn't happen for zip
archives.
The consequence is that in a configuration with core.autocrlf set,
although files in a tar archive are created with crlf line endings,
files in a zip archive retain unix line endings.
Signed-off-by: Charles Bailey <charles@hashpling.org> Acked-by: René Scharfe <rene.scharfe@lsrfire.ath.cx> Signed-off-by: Junio C Hamano <gitster@pobox.com>
git-svn: fix handling of even funkier branch names
Apparently do_switch() tolerates the lack of escaping in less
funky branch names. For the really strange and scary ones, we
need to escape them properly. It strangely maintains compatible
with the existing handling of branch names with spaces and
exclamation marks.
Reported-by: m.skoric@web.de ($gmane/94677) Signed-off-by: Eric Wong <normalperson@yhbt.net>
Textual diff output for unmerged paths was too eager to give condensed
combined diff. Even though "diff -c" (and "diff-files -c -p") is a
request to view combined diff without condensing (otherwise the user would
have explicitly asked for --cc, not -c), we showed "--cc" output anyway.
0fe7c1d (built-in diff: assorted updates, 2006-04-29) claimed to be
careful about doing this, but its breakage was hidden because back then
"git diff" was still a shell script that did not use the codepath it
introduced fully.
The object oriented version of File::Temp is a rather new incarnation it
seems. The File::Temp man page for v5.8.0 says "(NOT YET IMPLEMENTED)" in
the 'Objects' section. Instead of creating a file with a unique name in
the system TMPDIR, we can create our own temporary file with a static
name and use that instead.
Signed-off-by: Brandon Casey <casey@nrlssc.navy.mil> Tested-by: Tom G. Christensen <tgc@statsbiblioteket.dk> on RHEL 3, Perl 5.8.0 Signed-off-by: Junio C Hamano <gitster@pobox.com>
t9700/test.pl: avoid bareword 'STDERR' in 3-argument open()
Some versions of perl complain when 'STDERR' is used as the third argument
in the 3-argument form of open(). Convert to the 2-argument form which is
described for duping STDERR in my second edition camel book.
Signed-off-by: Brandon Casey <casey@nrlssc.navy.mil> Tested-by: Tom G. Christensen <tgc@statsbiblioteket.dk> on RHEL 3, Perl 5.8.0 Signed-off-by: Junio C Hamano <gitster@pobox.com>