By default, libcurl will follow circular http redirects
forever. Let's put a cap on this so that somebody who can
trigger an automated fetch of an arbitrary repository (e.g.,
for CI) cannot convince git to loop infinitely.
The value chosen is 20, which is the same default that
Firefox uses.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Previously, libcurl would follow redirection to any protocol
it was compiled for support with. This is desirable to allow
redirection from HTTP to HTTPS. However, it would even
successfully allow redirection from HTTP to SFTP, a protocol
that git does not otherwise support at all. Furthermore
git's new protocol-whitelisting could be bypassed by
following a redirect within the remote helper, as it was
only enforced at transport selection time.
This patch limits redirects within libcurl to HTTP, HTTPS,
FTP and FTPS. If there is a protocol-whitelist present, this
list is limited to those also allowed by the whitelist. As
redirection happens from within libcurl, it is impossible
for an HTTP redirect to a protocol implemented within
another remote helper.
When the curl version git was compiled with is too old to
support restrictions on protocol redirection, we warn the
user if GIT_ALLOW_PROTOCOL restrictions were requested. This
is a little inaccurate, as even without that variable in the
environment, we would still restrict SFTP, etc, and we do
not warn in that case. But anything else means we would
literally warn every time git accesses an http remote.
This commit includes a test, but it is not as robust as we
would hope. It redirects an http request to ftp, and checks
that curl complained about the protocol, which means that we
are relying on curl's specific error message to know what
happened. Ideally we would redirect to a working ftp server
and confirm that we can clone without protocol restrictions,
and not with them. But we do not have a portable way of
providing an ftp server, nor any other protocol that curl
supports (https is the closest, but we would have to deal
with certificates).
[jk: added test and version warning]
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Some protocols (like git-remote-ext) can execute arbitrary
code found in the URL. The URLs that submodules use may come
from arbitrary sources (e.g., .gitmodules files in a remote
repository). Let's restrict submodules to fetching from a
known-good subset of protocols.
Note that we apply this restriction to all submodule
commands, whether the URL comes from .gitmodules or not.
This is more restrictive than we need to be; for example, in
the tests we run:
git submodule add ext::...
which should be trusted, as the URL comes directly from the
command line provided by the user. But doing it this way is
simpler, and makes it much less likely that we would miss a
case. And since such protocols should be an exception
(especially because nobody who clones from them will be able
to update the submodules!), it's not likely to inconvenience
anyone in practice.
Reported-by: Blake Burkhart <bburky@bburky.com>
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
If we are cloning an untrusted remote repository into a
sandbox, we may also want to fetch remote submodules in
order to get the complete view as intended by the other
side. However, that opens us up to attacks where a malicious
user gets us to clone something they would not otherwise
have access to (this is not necessarily a problem by itself,
but we may then act on the cloned contents in a way that
exposes them to the attacker).
Ideally such a setup would sandbox git entirely away from
high-value items, but this is not always practical or easy
to set up (e.g., OS network controls may block multiple
protocols, and we would want to enable some but not others).
We can help this case by providing a way to restrict
particular protocols. We use a whitelist in the environment.
This is more annoying to set up than a blacklist, but
defaults to safety if the set of protocols git supports
grows). If no whitelist is specified, we continue to default
to allowing all protocols (this is an "unsafe" default, but
since the minority of users will want this sandboxing
effect, it is the only sensible one).
A note on the tests: ideally these would all be in a single
test file, but the git-daemon and httpd test infrastructure
is an all-or-nothing proposition rather than a test-by-test
prerequisite. By putting them all together, we would be
unable to test the file-local code on machines without
apache.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Abandoning an already applied change in "git rebase -i" with
"--continue" left CHERRY_PICK_HEAD and confused later steps.
* js/rebase-i-clean-up-upon-continue-to-skip:
rebase -i: do not leave a CHERRY_PICK_HEAD file behind
t3404: demonstrate CHERRY_PICK_HEAD bug
Various fixes around "git am" that applies a patch to a history
that is not there yet.
* pt/am-abort-fix:
am --abort: keep unrelated commits on unborn branch
am --abort: support aborting to unborn branch
am --abort: revert changes introduced by failed 3way merge
am --skip: support skipping while on unborn branch
am -3: support 3way merge on unborn branch
am --skip: revert changes introduced by failed 3way merge
"git for-each-ref" reported "missing object" for 0{40} when it
encounters a broken ref. The lack of object whose name is 0{40} is
not the problem; the ref being broken is.
* mh/reporting-broken-refs-from-for-each-ref:
read_loose_refs(): treat NULL_SHA1 loose references as broken
read_loose_refs(): simplify function logic
for-each-ref: report broken references correctly
t6301: new tests of for-each-ref error handling
"git commit --cleanup=scissors" was not careful enough to protect
against getting fooled by a line that looked like scissors.
* sg/commit-cleanup-scissors:
commit: cope with scissors lines in commit message
A minor bugfix when pack bitmap is used with "rev-list --count".
* jk/rev-list-no-bitmap-while-pruning:
rev-list: disable --use-bitmap-index when pruning commits
An ancient test framework enhancement to allow color was not
entirely correct; this makes it work even when tput needs to read
from the ~/.terminfo under the user's real HOME directory.
* rh/test-color-avoid-terminfo-in-original-home:
test-lib.sh: fix color support when tput needs ~/.terminfo
Revert "test-lib.sh: do tests for color support after changing HOME"
"git format-patch --ignore-if-upstream A..B" did not like to be fed
tags as boundary commits.
* jc/do-not-feed-tags-to-clear-commit-marks:
format-patch: do not feed tags to clear_commit_marks()
The reachability bitmaps do not have enough information to
tell us which commits might have changed path "foo", so the
current code produces wrong answers for:
git rev-list --use-bitmap-index --count HEAD -- foo
(it silently ignores the "foo" limiter). Instead, we should
fall back to doing a normal traversal (it is OK to fall
back rather than complain, because --use-bitmap-index is a
pure optimization, and might not kick in for other reasons,
such as there being no bitmaps in the repository).
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When skipping commits whose changes were already applied via `git rebase
--continue`, we need to clean up said file explicitly.
The same is not true for `git rebase --skip` because that will execute
`git reset --hard` as part of the "skip" handling in git-rebase.sh, even
before git-rebase--interactive.sh is called.
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When rev-list's --cherry option does not detect that a patch has already
been applied upstream, an interactive rebase would offer to reapply it and
consequently stop at that patch with a failure, mentioning that the diff
is empty.
Traditionally, a `git rebase --continue` simply skips the commit in such a
situation.
However, as pointed out by Gábor Szeder, this leaves a CHERRY_PICK_HEAD
behind, making the Git prompt believe that a cherry pick is still going
on. This commit adds a test case demonstrating this bug.
Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The setup code used to die when core.bare and core.worktree are set
inconsistently, even for commands that do not need working tree.
* jk/die-on-bogus-worktree-late:
setup_git_directory: delay core.bare/core.worktree errors
Recent "git prune" traverses young unreachable objects to safekeep
old objects in the reachability chain from them, which sometimes
caused error messages that are unnecessarily alarming.
* jk/squelch-missing-link-warning-for-unreachable:
suppress errors on missing UNINTERESTING links
silence broken link warnings with revs->ignore_missing_links
add quieter versions of parse_{tree,commit}
"git rebase -i" fired post-rewrite hook when it shouldn't (namely,
when it was told to stop sequencing with 'exec' insn).
* mm/rebase-i-post-rewrite-exec:
t5407: use <<- to align the expected output
rebase -i: fix post-rewrite hook with failed exec command
rebase -i: demonstrate incorrect behavior of post-rewrite
If tput needs ~/.terminfo for the current $TERM, then tput will
succeed before HOME is changed to $TRASH_DIRECTORY (causing color to
be set to 't') but fail afterward.
One possible way to fix this is to treat HOME like TERM: back up the
original value and temporarily restore it before say_color() runs
tput.
Instead, pre-compute and save the color control sequences before
changing either TERM or HOME. Use the saved control sequences in
say_color() rather than call tput each time. This avoids the need to
back up and restore the TERM and HOME variables, and it avoids the
overhead of a subshell and two invocations of tput per call to
say_color().
Signed-off-by: Richard Hansen <rhansen@bbn.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
This reverts commit 102fc80d32.
There are two issues with that commit:
* It is buggy. In pseudocode, it is doing:
color is set || TERM != dumb && color works && color=t
when it should be doing:
color is set || { TERM != dumb && color works && color=t }
* It unnecessarily disables color when tput needs to read
~/.terminfo to get the control sequences.
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Communication between the HTTP server and http_backend process can
lead to a dead-lock when relaying a large ref negotiation request.
Diagnose the situation better, and mitigate it by reading such a
request first into core (to a reasonable limit).
* jk/http-backend-deadlock:
http-backend: spool ref negotiation requests to buffer
t5551: factor out tag creation
http-backend: fix die recursion with custom handler
The clean/smudge interface did not work well when filtering an
empty contents (failed and then passed the empty input through).
It can be argued that a filter that produces anything but empty for
an empty input is nonsense, but if the user wants to do strange
things, then why not?
* jh/filter-empty-contents:
sha1_file: pass empty buffer to index empty file
Make "git stash something --help" error out, so that users can
safely say "git stash drop --help".
* jk/stash-options:
stash: recognize "--help" for subcommands
stash: complain about unknown flags
This reverts commit ed178ef13a.
That commit was an attempt to improve the safety of applying
a stash, because the application process may create
conflicted index entries, after which it is hard to restore
the original index state.
Unfortunately, this hurts some common workflows around "git
stash -k", like:
git add -p ;# (1) stage set of proposed changes
git stash -k ;# (2) get rid of everything else
make test ;# (3) make sure proposal is reasonable
git stash apply ;# (4) restore original working tree
If you "git commit" between steps (3) and (4), then this
just works. However, if these steps are part of a pre-commit
hook, you don't have that opportunity (you have to restore
the original state regardless of whether the tests passed or
failed).
It's possible that we could provide better tools for this
sort of workflow. In particular, even before ed178ef, it
could fail with a conflict if there were conflicting hunks
in the working tree and index (since the "stash -k" puts the
index version into the working tree, and we then attempt to
apply the differences between HEAD and the old working tree
on top of that). But the fact remains that people have been
using it happily for a while, and the safety provided by
ed178ef is simply not that great. Let's revert it for now.
In the long run, people can work on improving stash for this
sort of workflow, but the safety tradeoff is not worth it in
the meantime.
Signed-off-by: Jeff King <peff@peff.net>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The test expects that "chmod -r ~/.git-credentials" would make it
unreadable to the user, and thus needs the SANITY prerequisite.
Reported-by: Jean-Yves LENHOF <jean-yves@lenhof.eu.org>
Signed-off-by: Paul Tan <pyokagan@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
The diff and submodule shortlog appended to the commit message template
by 'git commit --verbose' are not stripped when the commit message
contains an indented scissors line.
When cleaning up a commit message with 'git commit --verbose' or
'--cleanup=scissors' the code is careful and triggers only on a pure
scissors line, i.e. a line containing nothing but a comment character, a
space, and the scissors cut. This is good, because people can embed
scissors lines in the commit message while using 'git commit --verbose',
and the text they write after their indented scissors line doesn't get
deleted.
While doing so, however, the cleanup function only looks at the first
line matching the scissors pattern and if it doesn't start at the
beginning of the line, then the function just returns without performing
any cleanup. This is wrong, because a "real" scissors line added by
'git commit --verbose' might follow, and in that case the diff and
submodule shortlog get included in the commit message.
Fix this by changing the scissors pattern to match only at the beginning
of the line, yet be careful to catch scissors on the first line as well.
Helped-by: Junio C Hamano <gitster@pobox.com>
Signed-off-by: SZEDER Gábor <szeder@ira.uka.de>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Since 7b3b7e3 (am --abort: keep unrelated commits since the last failure
and warn, 2010-12-21), git-am would refuse to rewind HEAD if commits
were made since the last git-am failure. This check was implemented in
safe_to_abort(), which checked to see if HEAD's hash matched the
abort-safety file.
However, this check was skipped if the abort-safety file was empty,
which can happen if git-am failed while on an unborn branch. As such, if
any commits were made since then, they would be discarded. Fix this by
carrying on the abort safety check even if the abort-safety file is
empty.
Signed-off-by: Paul Tan <pyokagan@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When git-am is first run on an unborn branch, no ORIG_HEAD is created.
As such, any applied commits will remain even after a git am --abort.
To be consistent with the behavior of git am --abort when it is not run
from an unborn branch, we empty the index, and then destroy the branch
pointed to by HEAD if there is no ORIG_HEAD.
Signed-off-by: Paul Tan <pyokagan@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Even when a merge conflict occurs with am --3way, the index will be
modified with the results of any successfully merged files. These
changes to the index will not be reverted with a
"git read-tree --reset -u HEAD ORIG_HEAD", as git read-tree will not be
aware of how the current index differs from HEAD or ORIG_HEAD.
To fix this, we first reset any conflicting entries in the index. The
resulting index will contain the results of successfully merged files
introduced by the failed merge. We write this index to a tree, and then
use git read-tree to fast-forward this "index tree" back to ORIG_HEAD,
thus undoing all the changes from the failed merge.
When we are on an unborn branch, HEAD and ORIG_HEAD will not point to
valid trees. In this case, use an empty tree.
Signed-off-by: Paul Tan <pyokagan@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
When git am --skip is run, git am will copy HEAD's tree entries to the
index with "git reset HEAD". However, on an unborn branch, HEAD does not
point to a tree, so "git reset HEAD" will fail.
Fix this by treating HEAD as en empty tree when we are on an unborn
branch.
Signed-off-by: Paul Tan <pyokagan@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
While on an unborn branch, git am -3 will fail to do a threeway merge as
it references HEAD as "our tree", but HEAD does not point to a valid
tree.
Fix this by using an empty tree as "our tree" when we are on an unborn
branch.
Signed-off-by: Paul Tan <pyokagan@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Even when a merge conflict occurs with am --3way, the index will be
modified with the results of any succesfully merged files (such as a new
file). These changes to the index will not be reverted with a
"git read-tree --reset -u HEAD HEAD", as git read-tree will not be aware
of how the current index differs from HEAD.
To fix this, we first reset any conflicting entries from the index. The
resulting index will contain the results of successfully merged files.
We write the index to a tree, then use git read-tree -m to fast-forward
the "index tree" back to HEAD, thus undoing all the changes from the
failed merge.
Signed-off-by: Paul Tan <pyokagan@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
NULL_SHA1 is used to indicate an "invalid object name" throughout our
code (and the code of other git implementations), so it is vastly more
likely that an on-disk reference was set to this value due to a
software bug than that NULL_SHA1 is the legitimate SHA-1 of an actual
object. Therefore, if a loose reference has the value NULL_SHA1,
consider it to be broken.
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Test clean-up.
* jk/skip-http-tests-under-no-curl:
tests: skip dav http-push tests under NO_EXPAT=NoThanks
t/lib-httpd.sh: skip tests if NO_CURL is defined
The pull.ff configuration was supposed to override the merge.ff
configuration, but it didn't.
* pt/pull-ff-vs-merge-ff:
pull: parse pull.ff as a bool or string
pull: make pull.ff=true override merge.ff
Multi-ref transaction support we merged a few releases ago
unnecessarily kept many file descriptors open, risking to fail with
resource exhaustion. This is for 2.4.x track.
* mh/write-refs-sooner-2.4:
ref_transaction_commit(): fix atomicity and avoid fd exhaustion
ref_transaction_commit(): remove the local flags variable
ref_transaction_commit(): inline call to write_ref_sha1()
rename_ref(): inline calls to write_ref_sha1() from this function
commit_ref_update(): new function, extracted from write_ref_sha1()
write_ref_to_lockfile(): new function, extracted from write_ref_sha1()
t7004: rename ULIMIT test prerequisite to ULIMIT_STACK_SIZE
update-ref: test handling large transactions properly
ref_transaction_commit(): fix atomicity and avoid fd exhaustion
ref_transaction_commit(): remove the local flags variable
ref_transaction_commit(): inline call to write_ref_sha1()
rename_ref(): inline calls to write_ref_sha1() from this function
commit_ref_update(): new function, extracted from write_ref_sha1()
write_ref_to_lockfile(): new function, extracted from write_ref_sha1()
t7004: rename ULIMIT test prerequisite to ULIMIT_STACK_SIZE
update-ref: test handling large transactions properly
The ref API did not handle cases where 'refs/heads/xyzzy/frotz' is
removed at the same time as 'refs/heads/xyzzy' is added (or vice
versa) very well.
* mh/ref-directory-file:
reflog_expire(): integrate lock_ref_sha1_basic() errors into ours
ref_transaction_commit(): delete extra "the" from error message
ref_transaction_commit(): provide better error messages
rename_ref(): integrate lock_ref_sha1_basic() errors into ours
lock_ref_sha1_basic(): improve diagnostics for ref D/F conflicts
lock_ref_sha1_basic(): report errors via a "struct strbuf *err"
verify_refname_available(): report errors via a "struct strbuf *err"
verify_refname_available(): rename function
refs: check for D/F conflicts among refs created in a transaction
ref_transaction_commit(): use a string_list for detecting duplicates
is_refname_available(): use dirname in first loop
struct nonmatching_ref_data: store a refname instead of a ref_entry
report_refname_conflict(): inline function
entry_matches(): inline function
is_refname_available(): convert local variable "dirname" to strbuf
is_refname_available(): avoid shadowing "dir" variable
is_refname_available(): revamp the comments
t1404: new tests of ref D/F conflicts within transactions
The "log --decorate" enhancement in Git 2.4 that shows the commit
at the tip of the current branch e.g. "HEAD -> master", did not
work with --decorate=full.
* mg/log-decorate-HEAD:
log: do not shorten decoration names too early
log: decorate HEAD with branch name under --decorate=full, too
There was a commented-out (instead of being marked to expect
failure) test that documented a breakage that was fixed since the
test was written; turn it into a proper test.
* sb/t1020-cleanup:
subdirectory tests: code cleanup, uncomment test
core.excludesfile (defaulting to $XDG_HOME/git/ignore) is supposed
to be overridden by repository-specific .git/info/exclude file, but
the order was swapped from the beginning. This belatedly fixes it.
* jc/gitignore-precedence:
ignore: info/exclude should trump core.excludesfile
The connection initiation code for "ssh" transport tried to absorb
differences between the stock "ssh" and Putty-supplied "plink" and
its derivatives, but the logic to tell that we are using "plink"
variants were too loose and falsely triggered when "plink" appeared
anywhere in the path (e.g. "/home/me/bin/uplink/ssh").
* bc/connect-plink:
connect: improve check for plink to reduce false positives
t5601: fix quotation error leading to skipped tests
connect: simplify SSH connection code path
"git rebase -i" moved the "current" command from "todo" to "done" a
bit too prematurely, losing a step when a "pick" did not even start.
* ph/rebase-i-redo:
rebase -i: redo tasks that die during cherry-pick
Some time ago, "git blame" (incorrectly) lost the convert_to_git()
call when synthesizing a fake "tip" commit that represents the
state in the working tree, which broke folks who record the history
with LF line ending to make their project portabile across
platforms while terminating lines in their working tree files with
CRLF for their platform.
* tb/blame-resurrect-convert-to-git:
blame: CRLF in the working tree and LF in the repo
Code clean-up for xdg configuration path support.
* pt/xdg-config-path:
path.c: remove home_config_paths()
git-config: replace use of home_config_paths()
git-commit: replace use of home_config_paths()
credential-store.c: replace home_config_paths() with xdg_config_home()
dir.c: replace home_config_paths() with xdg_config_home()
attr.c: replace home_config_paths() with xdg_config_home()
path.c: implement xdg_config_home()
t0302: "unreadable" test needs POSIXPERM
t0302: test credential-store support for XDG_CONFIG_HOME
git-credential-store: support XDG_CONFIG_HOME
git-credential-store: support multiple credential files
If there is a loose reference file with invalid contents, "git
for-each-ref" incorrectly reports the problem as being a missing
object with name NULL_SHA1:
$ echo '12345678' >.git/refs/heads/nonsense
$ git for-each-ref
fatal: missing object 0000000000000000000000000000000000000000 for refs/heads/nonsense
With an explicit "--format" string, it can even report that the
reference validly points at NULL_SHA1:
$ git for-each-ref --format='%(objectname) %(refname)'
0000000000000000000000000000000000000000 refs/heads/nonsense
$ echo $?
0
This has been broken since
b7dd2d2 for-each-ref: Do not lookup objects when they will not be used (2009-05-27)
, which changed for-each-ref from using for_each_ref() to using
git_for_each_rawref() in order to avoid looking up the referred-to
objects unnecessarily. (When "git for-each-ref" is given a "--format"
string that doesn't include information about the pointed-to object,
it does not look up the object at all, which makes it considerably
faster. Iterating with DO_FOR_EACH_INCLUDE_BROKEN is essential to this
optimization because otherwise for_each_ref() would itself need to
check whether the object exists as part of its brokenness test.)
But for_each_rawref() includes broken references in the iteration, and
"git for-each-ref" doesn't itself reject references with REF_ISBROKEN.
The result is that broken references are processed *as if* they had
the value NULL_SHA1, which is the value stored in entries for broken
references.
Change "git for-each-ref" to emit warnings for references that are
REF_ISBROKEN but to otherwise skip them.
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Add tests that for-each-ref correctly reports broken loose reference
files and references that point at missing objects. In fact, two of
these tests fail, because (1) NULL_SHA1 is not recognized as an
invalid reference value, and (2) for-each-ref doesn't respect
REF_ISBROKEN. Fixes to come.
Note that when for-each-ref is run with a --format option that doesn't
require the object to be looked up, then we should still notice if a
loose reference file is corrupt or contains NULL_SHA1, but we don't
notice if it points at a missing object because we don't do an object
lookup. This is OK.
Signed-off-by: Michael Haggerty <mhagger@alum.mit.edu>
Signed-off-by: Junio C Hamano <gitster@pobox.com>
Junio C Hamano
Blake Burkhart
Jeff King
Johannes Schindelin
Richard Hansen
Paul Tan
SZEDER Gábor
Michael Haggerty