replace direct calls to unlink(2) with unlink_or_warn
This helps to notice when something's going wrong, especially on
systems which lock open files.
I used the following criteria when selecting the code for replacement:
- it was already printing a warning for the unlink failures
- it is in a function which already printing something or is
called from such a function
- it is in a static function, returning void and the function is only
called from a builtin main function (cmd_)
- it is in a function which handles emergency exit (signal handlers)
- it is in a function which is obvously cleaning up the lockfiles
Signed-off-by: Alex Riesen <raa.lkml@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Introduce an unlink(2) wrapper which gives warning if unlink failed
This seem to be a very common pattern in the current code.
The function prints a generic removal failure message, the file name
which failed and readable errno presentation. The function preserves
errno and always returns the value unlink(2) returned, but prints
no message for ENOENT, as it was the most often filtered out in the
code calling unlink. Besides, removing a file is anyway the purpose of
calling unlink.
Signed-off-by: Alex Riesen <raa.lkml@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
On OS X (and maybe other unices), getaddrinfo(3) returns NULL
in the ai_canonname field if it's called with an IP address for
the hostname. We'll now use the IP address for the hostname if
ai_canonname was NULL, this also matches the behaviour on Linux.
merge-recursive: do not die on a conflicting submodule
We cannot represent the 3-way conflicted state in the work tree
for these entries, but it is normal not to have commit objects
for them in our repository. Just update the index and the life
will be good.
The combine diff logic knew only about blobs (and their checked-out form
in the work tree, either regular files or symlinks), and barfed when fed
submodules. This "externalizes" gitlinks in the same way as the normal
patch generation codepath does (i.e. "Subproject commit Xxx\n") to fix the
issue.
On platforms with $X, make removes any leftover scripts 'a' from
earlier builds if a new binary 'a.exe' is now built. However, on
cygwin 1.7.0, 'git' and 'git.exe' now consistently name the same file.
Test for file equality before attempting a remove, in order to avoid
nuking just-built binaries.
This repeats commit 0d768f7 for the installation destdir.
Signed-off-by: Eric Blake <ebb9@byu.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
* maint:
grep: fix segfault when "git grep '('" is given
Documentation: fix a grammatical error in api-builtin.txt
builtin-merge: fix a typo in an error message
* maint-1.6.1:
grep: fix segfault when "git grep '('" is given
Documentation: fix a grammatical error in api-builtin.txt
builtin-merge: fix a typo in an error message
* maint-1.6.0:
grep: fix segfault when "git grep '('" is given
Documentation: fix a grammatical error in api-builtin.txt
builtin-merge: fix a typo in an error message
SubmittingPatches: itemize and reflect upon well written changes
The SubmittingPatches file was trimmed down from a somewhat
overwhelming set of requirements from the Linux Kernel equivalent;
however perhaps a little of it can be returned without making the
text too long.
Signed-off-by: Sam Vilain <sam@vilain.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
When the user has defined NO_PERL, we want to skip building
gitweb entirely. However, the conditional to add
gitweb/gitweb.cgi to OTHER_PROGRAMS was evaluated before we
actually parsed the user's config.mak. This meant that "make
NO_PERL=NoThanks" worked fine, but putting "NO_PERL=NoThanks"
into your config.mak broke the build (it wanted gitweb.cgi
to satisfy "all", but the rule to build it was conditionally
ignored, so it complained).
Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Trying to be lazy and comparing files with fake-editor.sh to avoid
having to provide another example text does not work well: the blob
name changes when SHELL_PATH changes, and so does the 'index' line
in the diff.
Therefore provide a second example text.
Noticed by Mike Ralphson.
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> Signed-off-by: Junio C Hamano <gitster@pobox.com>
prune-packed: advanced progress even for non-existing fan-out directories
A progress indicator is used to count through the 256 object fan-out
directories while unused object files are removed. (However, it becomes
visible only if this process takes long enough.)
Previously, display_progress() was only called if object files were
actually removed. But if directories towards the end (fd/, fe/, ff/) did
not exist, this could leave a strange line
Removing duplicate objects: 99% (255/256), done.
in the terminal instead of the expected "100% (256/256)".
Signed-off-by: Johannes Sixt <j6t@kdbg.org> Signed-off-by: Junio C Hamano <gitster@pobox.com>
diff: do not color --stat output like patch context
The diffstat used the color.diff.plain slot (context text) for coloring
filenames and the whole summary line. This didn't look nice and the
affected text isn't patch context at all.
Signed-off-by: Markus Heidelberg <markus.heidelberg@web.de> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Add an option not to use link(src, dest) && unlink(src) when that is unreliable
It seems that accessing NTFS partitions with ufsd (at least on my EeePC)
has an unnerving bug: if you link() a file and unlink() it right away,
the target of the link() will have the correct size, but consist of NULs.
It seems as if the calls are simply not serialized correctly, as single-stepping
through the function move_temp_to_file() works flawlessly.
As ufsd is "Commertial software" (sic!), I cannot fix it, and have to work
around it in Git.
At the same time, it seems that this fixes msysGit issues 222 and 229 to
assume that Windows cannot handle link() && unlink().
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de> Acked-by: Johannes Sixt <j6t@kdbg.org> Signed-off-by: Junio C Hamano <gitster@pobox.com>
inet_ntop is not protocol independent.
getnameinfo(3) is part of POSIX and is available when getaddrinfo(3) is.
This code is only compiled when NO_IPV6 isn't defined.
The old method was buggy anyway, not every ipv6 address was converted
properly because the buffer (addr) was too small.
Signed-off-by: Benjamin Kramer <benny.kra@googlemail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
t7700-repack: repack -a now works properly, expect success from test
Since the recent rework of the object listing mechanism of
pack-objects/rev-list, git-repack now properly packs objects from alternate
repositories even when the local repository contains packs.
Signed-off-by: Brandon Casey <casey@nrlssc.navy.mil> Signed-off-by: Junio C Hamano <gitster@pobox.com>
t9001: use older Getopt::Long boolean prefix '--no' rather than '--no-'
The '--no-thread' option is a Getopt::Long boolean option. The '--no-'
prefix (as in --no-thread) for boolean options is not supported in
Getopt::Long version 2.32 which was released with Perl 5.8.0. This version
only supports '--no' as in '--nothread'. More recent versions of
Getopt::Long, such as version 2.34, support either prefix. So use the older
form in the tests.
Signed-off-by: Brandon Casey <casey@nrlssc.navy.mil> Signed-off-by: Junio C Hamano <gitster@pobox.com>
progress bar: round to the nearest instead of truncating down
Often the throughput output is requested when the data read so far is
one smaller than multiple of 1024; because 1023/1024 is ~0.999, it often
shows up as 0.99 because the code currently truncates.
remote.c: do not trigger remote.<name>.<var> codepath for two-level names
If the config file contains a section like this:
[remote]
default = foo
(it should be '[remotes]') then commands like
git status
git checkout
git branch -v
fail even though they are not obviously related to remotes. (These
commands write "ahead, behind" information and, therefore, access the
per-remote information).
Unknown configuration keys should be ignored, not trigger errors.
Signed-off-by: Johannes Sixt <j6t@kdbg.org> Signed-off-by: Junio C Hamano <gitster@pobox.com>
If for some reason the config file contains a key without a subkey like
[man]
foo = bar
then even a plain
git help
produces an error message. With this patch such an entry is ignored.
Additionally, the warning about unknown sub-keys is removed. It could
become annoying if new sub-keys are introduced in the future, and then
the configuration is read by an old version of git that does not know
about it.
Signed-off-by: Johannes Sixt <j6t@kdbg.org> Signed-off-by: Junio C Hamano <gitster@pobox.com>
git-format-patch.txt: general rewordings and cleanups
Clarify --no-binary description using some words from the original
commit 37c22a4b (add --no-binary, 2008-05-9). Cleanup --suffix
description. Add --thread style option to synopsis and reorganize it a
bit. Clarify renaming patches example and the configuration paragraph.
Signed-off-by: Stephen Boyd <bebarino@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
* jc/maint-read-tree-multi:
checkout branch: prime cache-tree fully
read-tree -m A B: prime cache-tree from the switched-to tree
Move prime_cache_tree() to cache-tree.c
read-tree A B: do not corrupt cache-tree
Documentation: git-svn: fix a grammatical error without awkwardness
The way the sentence is currently written, there needs to be an "its",
but this leads to: "however the remote wildcard may be anywhere as long
as it's its own" which is awkward to read.
Instead, this patch fixes he grammar in a simpler way.
Signed-off-by: Wesley J. Landaker <wjl@icecavern.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
In ActivetState Perl, exec does not wait for the started program. This
breaks difftool tests and may cause unexpected behaviour: git difftool
has returned, but the rest of code (diff and possibly the interactive
program are still running in the background.
Acked-by: David Aguilar <davvid@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Make it possible to color the status character ('*' '!' '+' '-') of each
commit corresponding to the branch it's in. This makes it easier to
follow a particular branch, especially if there are larger gaps in the
output.
Add the config option color.showbranch and the command line options
--color and --no-color to control the colored output.
Signed-off-by: Markus Heidelberg <markus.heidelberg@web.de> Signed-off-by: Junio C Hamano <gitster@pobox.com>
graph API: fix extra space during pre_commit_line state
An extra space is being inserted between the "commit" column and all of
the successive edges. Remove this space. This regression was
introduced by 427fc5b.
Signed-off-by: Allan Caffee <allan.caffee@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
test-lib.sh: Help test_create_repo() find the templates dir
Currently, test_create_repo() expects that templates can be found below
`pwd`/.. This assumption fails when tests are run against a git
installed somewhere else or test_create_repo() is called from
subdirectiories (several tests do this).
Therefore, use $TEST_DIRECTORY as introduced in 2d84e9fb and expect
templates to be present in $TEST_DIRECTORY/.. which should be the root
dir of the git checkout.
Signed-off-by: Michael J Gruber <git@drmicha.warpmail.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
gitk: Fix compare-commits function when we have local changes
This fixes a bug in the compare-commits function added in commit 010509f2 ("gitk: Add a command to compare two strings of commits")
where gitk would show an error dialog if the comparison of commits
got to a fake commit (one showing local changes). It extends
getpatchid to handle these fake commits by using [diffcmd] to get
the git diff command variant to use, and also handles the situation
where an error occurs.
Now that we can have the fake commit IDs showing up, which are
00..00 and 00..01, the short ID is ambiguous. To make sure the links
point to the right commit, this adds a new [appendshortlink] procedure
which takes the full link destination, and uses that rather than
appendwithlinks.
* mk/maint-apply-swap:
tests: make test-apply-criss-cross-rename more robust
builtin-apply: keep information about files to be deleted
tests: test applying criss-cross rename patch
fetch: Strip usernames from url's before storing them
When pulling from a remote, the full URL including username
is by default added to the commit message. Since it adds
very little value but could be used by malicious people to
glean valid usernames (with matching hostnames), we're far
better off just stripping the username before storing the
remote URL locally.
Note that this patch has no lasting visible effect when
"git pull" does not create a merge commit. It simply
alters what gets written to .git/FETCH_HEAD, which is used
by "git merge" to automagically create its messages.
Signed-off-by: Andreas Ericsson <ae@op5.se> Signed-off-by: Junio C Hamano <gitster@pobox.com>
grep: don't support "grep.color"-like config options
color.grep and color.grep.* is the official and documented way to
highlight grep matches. Comparable options like diff.color.* and
status.color.* exist for backward compatibility reasons only and are not
documented any more.
Signed-off-by: Markus Heidelberg <markus.heidelberg@web.de> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Commit 55f0566 (get_local_heads(): do not return random pointer if
there is no head, 2009-04-17) fixed a segfault for git push, this
patch adds a test-case to avoid future regressions.
Signed-off-by: Matthieu Moy <Matthieu.Moy@imag.fr> Signed-off-by: Junio C Hamano <gitster@pobox.com>
builtin-apply: keep information about files to be deleted
Example correct diff generated by `diff -M -B' might look like this:
diff --git a/file1 b/file2
similarity index 100%
rename from file1
rename to file2
diff --git a/file2 b/file1
similarity index 100%
rename from file2
rename to file1
Information about removing `file2' comes after information about creation
of new `file2' (renamed from `file1'). Existing implementation isn't able to
apply such patch, because it has to know in advance which files will be
removed.
This patch populates fn_table with information about removal of files
before calling check_patch() for each patch to be applied.
Signed-off-by: Michał Kiedrowicz <michal.kiedrowicz@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Windows: Skip fstat/lstat optimization in write_entry()
Commit e4c72923 (write_entry(): use fstat() instead of lstat() when file
is open, 2009-02-09) introduced an optimization of write_entry().
Unfortunately, we cannot take advantage of this optimization on Windows
because there is no guarantee that the time stamps are updated before the
file is closed:
"The only guarantee about a file timestamp is that the file time is
correctly reflected when the handle that makes the change is closed."
The failure of this optimization on Windows can be observed most easily by
running a 'git checkout' that has to update several large files. In this
case, 'git checkout' will report modified files, but infact only the
timestamps were incorrectly recorded in the index, as can be verified by a
subsequent 'git diff', which shows no change.
Dmitry Potapov reports the same fix needs on Cygwin; this commit contains
his updates for that.
Signed-off-by: Johannes Sixt <j6t@kdbg.org> Signed-off-by: Junio C Hamano <gitster@pobox.com>
When switching to another branch, the earlier code relied on incremental
invalidation of cache-tree entries to degrade it. While it is not wrong
per-se, we know that the resulting index must fully match the branch we
are switching to unless the -m (merge) option is used.
We should simply fully re-prime the cache-tree using the new tree object
in such a case. And for safety, invalidate the cache-tree as a whole in
other cases.
An earlier commit aab3b9a (read-tree A B C: do not create a bogus index
and do not segfault, 2009-03-12) resurrected the support for an obscure
(but useful) feature to read and overlay more than one tree into the index
without the -m (merge) option. But the fix was not enough.
Exercising this feature exposes a longstanding bug in the code that primes
the cache-tree in the index from the tree that was read. The intention
was that when we know that the index must exactly match the tree we just
read, we prime the entire cache-tree with it.
However, the logic to detect that case incorrectly triggered if you read
two trees without -m. This resulted in a corrupted cache-tree, and
write-tree would have produced an incorrect tree object out of such an
index.
Windows: Work around intermittent failures in mingw_rename
We have replaced rename() with a version that can rename a file to a
destination that already exists. Nevertheless, many users, the author
included, observe failures in the code that are not reproducible.
The theory is that the failures are due to some other process that happens
to have opened the destination file briefly at the wrong moment. (And there
is no way on Windows to delete or replace a file that is currently open.)
The most likely candidate for such a process is a virus scanner. The
failure is more often observed while there is heavy git activity (for
example while the test suite is running or during a rebase operation).
We work around the failure by retrying the rename operation if it failed
due to ERROR_ACCESS_DENIED. The retries are delayed a bit: The first only
by giving up the time slice, the next after the minimal scheduling
granularity, and if more retries are needed, then we wait some non-trivial
amount of time with exponential back-off.
Signed-off-by: Johannes Sixt <j6t@kdbg.org> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Fix the detection of the requested snapshot format, which failed for
PATH_INFO URLs since the references to the hashes which describe the
supported snapshot formats weren't dereferenced appropriately.
Signed-off-by: Holger Weiß <holger@zedat.fu-berlin.de> Acked-by: Jakub Narebski <jnareb@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
* ef/maint-fast-export:
builtin-fast-export.c: handle nested tags
builtin-fast-export.c: fix crash on tagged trees
builtin-fast-export.c: turn error into warning
test-suite: adding a test for fast-export with tag variants