FS.list() is supposed to return an empty array if the File given is not
a directory.
Bug: 550111
Change-Id: I245da5f1f2bdafd9dfb38fb8d7eff27d900cd5a8
Signed-off-by: Thomas Wolf <thomas.wolf@paranor.ch>
Escape paths when writing them to a config file to ensure they work
with backslashes and unusual characters.
Bug: 550111
Change-Id: Iedc5c0f2c0c02ac6cadf43cdae0f0d19578aed91
Signed-off-by: Thomas Wolf <thomas.wolf@paranor.ch>
Ensure that this test works on Windows.
Bug: 550111
Change-Id: Ib3ca803f4d9521a5f2a0e1344bfbc0d2beccc9cc
Signed-off-by: Thomas Wolf <thomas.wolf@paranor.ch>
On Windows, java.io.File.canExecute is always true. Guard assertions
testing the executable bit with FS.DETECTED.supportsExecute().
Bug: 550111
Change-Id: I2704d122f5b1086d01a0503a8c047a02ecbc1d4a
Signed-off-by: Thomas Wolf <thomas.wolf@paranor.ch>
We want to save in the destination repository what manifest created its
structure. This helps to detect and debug failures in the manifest ->
superproject translations. The src commit should be easily readable from
the superproject tip.
Offer an API to write a file in the destination repository. RepoCommand
callers (e.g. gerrit supermanifest plugin) can use this to add a
file with the repo/ref/hash of the manifest.
Alternatives considered to write the source repo/ref/hash:
* .gitattributes of the .gitmodules file. Some updates in the manifest
don't touch the .gitmodules (e.g. a linkfile change), so it can fall
out of sync.
* commit message. Caller would need to follow the commit history to
find the latest modification by repo command. This is not helpful
e.g. for build bots that want to get the value in one call.
Change-Id: I113662734a7ccd39cbc60b46ad3f73038c807682
RepoCommand parses the manifest to get a list of projects, clears up
conflicts and then writes to the superproject. The first steps are
common but the writing is completely different for bare or "regular"
(with working dir) repository.
Split writing to bare and regular repos into its own classes. This
simplifies RepoCommand class and makes clearer what happens on each side
(e.g. many options apply only to bare repos).
Change-Id: I256e15729bd53ee15fc56de88bce86a2edb2417a
C git also does so. Note that currently the comment character is
hard-coded as the hash '#' throughout JGit.
Bug: 548529
Change-Id: I4a5597694082a9e5b07412b365cfaf41fa034cfa
Signed-off-by: Thomas Wolf <thomas.wolf@paranor.ch>
Respect git config commit.cleanup for rewording. Note that by default
this is CleanupMode.STRIP, whereas before this change, JGit would take
the reworded message verbatim.
Squashing was the only place in JGit where it automatically and
unconditionally removed comment lines from commit messages. In other
places it didn't do so, and client code needed to do so.
Unconditionally removing comments is problematic if the commit message
_should_ contain some line starting with a hash, which can easily occur
with the way Github, Gitlab, and other git web servers link to issues
or PRs: they all allow the short-hand "#<number>".
Introduce a new InteractiveHandler2 extension interface, which can
return the edited message _and_ a clean-up mode. This way, client code
can decide on its own how to clean the message, and if JGit shouldn't
do any further cleaning, it can return CleanupMode.VERBATIM. Or
CleanupMode.WHITESPACE. (In the case of SQUASH, it is then of course
the client's responsibility to remove the squash comment lines.)
If the old InteractiveHandler interface is used, CleanupMode.STRIP is
applied unconditionally for squashing, as before.
Bug: 578173
Change-Id: Ia0040c247884e684587dd45d6cb85f8b72a4b876
Signed-off-by: Thomas Wolf <thomas.wolf@paranor.ch>
Use CommitConfig.CleanupMode to implement git commit --cleanup. Add
setters for the clean-up mode, the comment character, and for the
default default clean-up mode.
Behavior of existing client code is unchanged as the default clean-up
mode is set to "verbatim". To use git config defaults, one can call
setCleanupMode(CleanupMode.DEFAULT). The default comment character
is hard-coded as '#' for now, as in other parts of JGit. Implementing
full support for core.commentChar shall be done in a separate change.
Bug: 553065
Change-Id: I470785e464a762d3f409f163f1cbdbb98dd81aaf
Signed-off-by: Thomas Wolf <thomas.wolf@paranor.ch>
Add an enumeration for the possible values, and a method to resolve the
"default" value. Give CommitConfig a static method to process a text
according to a given clean-up mode and comment character.
(The core.commentChar is not yet handled by JGit; it's hard-coded as #.)
Bug: 553065
Change-Id: If6e384522275f73b713fbc29ffcaa1753c239dea
Signed-off-by: Thomas Wolf <thomas.wolf@paranor.ch>
Fix tests failing on Windows because Repository instance is created but
not closed on tear down.
Fix repositories closed twice, except in tests that test this behavior
explicitly.
Name the temporary directories the tests run in after the test method;
that makes it easier to figure out in which tests repositories are
closed twice if it should occur again in the future.
Bug: 550111
Change-Id: I9398b58f0f36d2c29236d2a9a8599117d9083980
Signed-off-by: Nail Samatov <sanail@yandex.ru>
Signed-off-by: Thomas Wolf <thomas.wolf@paranor.ch>
With eol=native, we expect LF on Unixes, and CR-LF on Windows. One test
didn't account for this and always expected LF, and thus failed on
Windows.
Bug: 550111
Change-Id: I69354ac691c464d1b6003812ddb4510c5ab5e77a
Signed-off-by: Thomas Wolf <thomas.wolf@paranor.ch>
If the walk is created via ObjectWalk(Repository), it creates a new
ObjectReader. This reader was closed only on dispose(). If such an
ObjectWalk was used in a try-with-resource statement the reader might
not get closed.
Bug: 578458
Change-Id: I1be31829dc466530f23006a53c29b657fd5fb410
Signed-off-by: Thomas Wolf <thomas.wolf@paranor.ch>
Test.txt should not be in the "removed" list if it can't be deleted
but only in the "not deleted" list. The test was wrong.
Bug: 550111
Change-Id: I3ecede4278014c15015c8c24089647fa3db3742f
Signed-off-by: Thomas Wolf <thomas.wolf@paranor.ch>
Some tests checked whether or not a file is executable via
java.io.File.canExecute(). But that always returns true on Windows.
Use FS.DETECTED.canExecute() instead, and guard all such assertions to
run only of file systems that do support the "execute" flag.
Bug: 550111
Change-Id: Iacb9e414b612359fcecb61312c3dfb830801cd36
Signed-off-by: Thomas Wolf <thomas.wolf@paranor.ch>
The tests assumed that a path like "/tmp" was an absolute path, and
also compared against strings with forward slashes. On Windows, "/tmp"
is not an absolute path and thus resolved against the current directory,
and the separator is a backslash.
Change the tests to use ~/ notation, and test paths resolved against
the (mocked) user home directory. That way, the tests are independent
of the file system used.
Bug: 550111
Change-Id: I1c31608ca83c8d8586256d1586a792e4a33cfaa4
Signed-off-by: Thomas Wolf <thomas.wolf@paranor.ch>
Let's be lenient and accept hunk lines terminated by CR-LF, too, not
just lines terminated by LF.
Bug: 550111
Change-Id: I7f796df666300ab56cc6c07f22eda45fbf4c941e
Signed-off-by: Thomas Wolf <thomas.wolf@paranor.ch>
Explain SSH agent protocols, what transports are available and how to
choose them in ~/.ssh/config. For Windows, add some information on
which commonly used SSH agents can be used.
Change-Id: I0b08a95654fd76643512606edb1ed74d9980aa85
Signed-off-by: Thomas Wolf <thomas.wolf@paranor.ch>
Parse the value from the ssh config and if set use it when connecting.
Change-Id: I85b44c9468a5027602375706612c46ea7a99b2bd
Signed-off-by: Thomas Wolf <thomas.wolf@paranor.ch>
An SSH agent might contain keys that Apache MINA sshd cannot handle.
Pageant for instance can contain ed448 keys, which are not implemented
in OpenSSH or in Apache MINA sshd.
When an agent delivers such keys, simply skip (and log) them. That way,
we can work with the remaining keys. Otherwise a single unknown key in
the agent would break pubkey authentication.
Change-Id: I3945d932c7e64b628465004cfbaf10f4dc05f3e4
Signed-off-by: Thomas Wolf <thomas.wolf@paranor.ch>
Add parsing of the config. Implement the SSH agent protocol for adding
a key. In the pubkey authentication, add keys to the agent as soon as
they've been loaded successfully, before even attempting to use them
for authentication. OpenSSH does the same.
Bug: 577052
Change-Id: Id1c08d9676a74652256b22281c2f8fa0b6508fa6
Signed-off-by: Thomas Wolf <thomas.wolf@paranor.ch>
OpenSSH has (for legacy reasons?) the option of specifying the default
environment variable directly, instead of using ${SSH_AUTH_SOCK}. Make
sure the plain variable name is not taken as a relative path name.
Bug: 577053
Change-Id: If8f550dffc43887254f71aa0b487c50fa14d0627
Signed-off-by: Thomas Wolf <thomas.wolf@paranor.ch>
Win32-OpenSSH uses a named Windows pipe for communication. Implement
a connector for this mechanism using JNA. Choose the appropriate
connector based on the setting of the 'identityAgent' parameter.
Bug: 577053
Change-Id: I205f07fb33654aa18ca5db92706e65544ce38641
Signed-off-by: Thomas Wolf <thomas.wolf@paranor.ch>
If an SSH agent is used but "IdentitiesOnly yes" is set, only those
keys from the agent that correspond to one of the keys explicitly given
via an IdentityFile directive are to be used.
Implement this by filtering the list of keys obtained from the agent
against the list of IdentityFiles, each entry suffixed with ".pub".
Load the public keys from these files, and ignore all other keys from
the agent. Keys without ".pub" file are also ignored.
Apache MINA sshd has no operation to load only the public key from a
private key file, so we have to rely on *.pub files.
Bug: 577053
Change-Id: I75c2c0b3ce35781c933ec2944bd6da1b94f4caf9
Signed-off-by: Thomas Wolf <thomas.wolf@paranor.ch>
Handle the 'none' value, and change the value to select Pageant to
something that looks like an absolute UNC path name to avoid it's
handled as an relative path name.
Bug: 577053
Change-Id: I4ccf047abbc1def50e2782319e4fa7c744069401
Signed-off-by: Thomas Wolf <thomas.wolf@paranor.ch>
Fix TransportException occurring when deleting
a branch and push that change to remote repository
if BuiltinLFS is used to work with repository.
When finding LFS pointers in RemoteRepUpdate,
LfsPrePushHook fails to open ObjectReader
with new object id equal to ObjectId.zeroId().
If update is a deleting update (new object id is zero id),
we can assume that this update doesn't contain LFS Pointer
and we can skip step with extracting LFS pointer for that
RemoteRefUpdate.
Bug: 578313
Change-Id: Ic4367978338b8234d39d9af0d9674490f79fc22d
Signed-off-by: Nail Samatov <sanail@yandex.ru>
Checking the cancelled status and keeping a digest of the written data
is useful for other output streams. e.g. to write commit-graphs.
Pull up that functionality to a superclass, so it can be reused.
Change-Id: I177b50be09c4ea631e7a144cc6127085ec2ca411
Signed-off-by: kylezhao <kylezhao@tencent.com>
In I988f61e34 tools/remote-bazelrc was added, but this resource file
wasn't imported from main resource file workspace/.bazelrc.
Change-Id: Ie0c2f128fbccaa6f9e99ad29a3a76ac294f3cd29
Enhance cache performance monitoring for large data such as pack and
bitmap indexes. Provide details about what is loaded and evicted from
cache like total number of cache hits, time in cache before eviction.
Add a custom consumer to report loading events and eviction events when
enabled.
Signed-off-by: Alina Djamankulova <adjama@google.com>
Change-Id: I5739325db7ff7ec370e4defd8f7e46f1c3f5d2dd
The Bazel build can be used with Google's Remote Build Execution.
This needs the following setup steps:
$ gcloud auth application-default login
$ gcloud services enable remotebuildexecution.googleapis.com --project=${PROJECT}
Create a worker pool. The instances should have at least 4 CPUs each
for adequate performance.
$ gcloud alpha remote-build-execution worker-pools create default \
--project=${PROJECT} \
--instance=default_instance \
--worker-count=50 \
--machine-type=e2-standard-4 \
--disk-size=200
To use RBE, execute
$ bazel test --config=remote \
--remote_instance_name=projects/${PROJECT}/instances/default_instance \
//...
Change-Id: I988f61e342dab2136d8752ace945a4ed91a4189a
Bump Bazel version to release 5.0.0. In this new Bazel release,
--incompatible_use_toolchain_resolution_for_java_rules is flipped, that
means that the build must be adapted to toolchain resolution.
Specification toolchain resolution for java rules is here: [1]. Main
tracking Bazel issue is here: [2].
Given that new Bazel release also added support for remote JDK 17, add
experimental support for building with remote JDK 17 to produce major
byte code version 61.
Test Plan:
To build with remote JDK 11 run:
$ bazel build :all
To build with remote JDK 17 run:
$ bazel build --config java17 :all
[1] https://docs.google.com/document/d/1MVbBxbKVKRJJY7DnkptHpvz7ROhyAYy4a-TZ-n7Q0r4/edit?usp=sharing
[2] https://github.com/bazelbuild/bazel/issues/7849
Change-Id: I8d26dff722c6677ea7642913b61e416b23ea9041
In upcoming Bazel release 5.0 resource_jars attribute was removed: [1].
Replace resource_jars with runtime_deps as recommended in the release
notes.
[1] https://github.com/bazelbuild/bazel/issues/13221
Change-Id: Iab834d647cf63259ca182adc1f862b6001db0cce
and update
- com.google.gson to 2.8.9.v20220111-1409
- org.bouncycastle.bcpg to 1.70.0.v20220105-1522
- org.bouncycastle.bcpkix to 1.70.0.v20220105-1522
- org.bouncycastle.bcprov to 1.70.0.v20220105-1522
- org.bouncycastle.bcutil to 1.70.0.v20220105-1522
Change-Id: I7d9159e0e8524459a11da84cb2db709cb1f0823d
Matthias Sohn