* stable-4.10:
Bazel: Update bazlets to the latest master revision
Bazel: Remove FileTreeIteratorWithTimeControl from BUILD file
BatchRefUpdate: repro racy atomic update, and fix it
Delete unused FileTreeIteratorWithTimeControl
Fix RacyGitTests#testRacyGitDetection
Change RacyGitTests to create a racy git situation in a stable way
Silence API warnings
Change-Id: If672b4f0c350f4e8ff7e1e706485cffd8137236d
Signed-off-by: Matthias Sohn <matthias.sohn@sap.com>
* stable-4.9:
BatchRefUpdate: repro racy atomic update, and fix it
Delete unused FileTreeIteratorWithTimeControl
Fix RacyGitTests#testRacyGitDetection
Change RacyGitTests to create a racy git situation in a stable way
Silence API warnings
Change-Id: Id5bf44645655fca40ad22bb1f1ad20a7c2e8f6db
Signed-off-by: Matthias Sohn <matthias.sohn@sap.com>
The latest revision includes various fixes to allow the build
to work with recent versions of Bazel.
Change-Id: I72c100b99762010946d9b2784286af560bbdf185
Signed-off-by: David Pursehouse <david.pursehouse@gmail.com>
FileTreeIteratorWithTimeControl was deleted in a024759, but was
not removed from the BUILD file, thus causing the bazel build to
fail.
Change-Id: I892c0ffcac947298d0d6009374ee2c5d9afefb66
Signed-off-by: David Pursehouse <david.pursehouse@gmail.com>
(cherry picked from commit e54fde8616)
PackedBatchRefUpdate was creating a new packed-refs list that was
potentially unsorted. This would be papered over when the list was
read back from disk in parsePackedRef, which detects unsorted ref
lists on reading, and sorts them. However, the BatchRefUpdate also
installed the new (unsorted) list in-memory in
RefDirectory#packedRefs.
With the timestamp granularity code committed to stable-5.1, we can
more often accurately decide that the packed-refs file is clean, and
will return the erroneous unsorted data more often. Unluckily timed
delays also cause the file to be clean, hence this problem was
exacerbated under load.
The symptom is that refs added by a BatchRefUpdate would stop being
visible directly after they were added. In particular, the Gerrit
integration tests uses BatchRefUpdate in its setup for creating the
Admin group, and then tries to read it out directly afterward.
The tests recreates one failure case. A better approach would be to
revise RefList.Builder, so it detects out-of-order lists and
automatically sorts them.
Fixes https://bugs.eclipse.org/bugs/show_bug.cgi?id=548716 and
https://bugs.chromium.org/p/gerrit/issues/detail?id=11373.
Bug: 548716
Change-Id: I613c8059964513ce2370543620725b540b3cb6d1
Signed-off-by: Han-Wen Nienhuys <hanwen@google.com>
Signed-off-by: Matthias Sohn <matthias.sohn@sap.com>
The only usage of this test iterator was removed in df637928d. Hence
delete this iterator and associated test.
Change-Id: I47710133ec3edc675c21db210960c024982668c6
Signed-off-by: Matthias Sohn <matthias.sohn@sap.com>
(cherry picked from commit a024759cf5)
This test case assumed file system timestamp resolution of 1 second. On
filesystems with a finer resolution this test fails since the index
entry is only smudged if the file index entry's lastModified and the
lastModified of the git index itself are within the same filesystem
timer tick. Fix this by ensuring that these timestamps are identical
which should work for any filesystem timer resolution.
Bug: 548188
Change-Id: Id84d59e1cfeb48fa008f8f27f2f892c4f73985de
Signed-off-by: Matthias Sohn <matthias.sohn@sap.com>
Signed-off-by: Christian Halstrick <christian.halstrick@sap.com>
(cherry picked from commit 1159f9dd7c)
By using File#setLastModified, we can create a racy git situation
stably.
Tested with --runs_per_test=100
Bug: 526111
Change-Id: I60b3632d353e19f335668325aa603640be423f58
Signed-off-by: Masaya Suzuki <masayasuzuki@google.com>
(cherry picked from commit df637928d2)
The java.net.ProxySelector is quite a bit different from the one
in Eclipse. Eclipse (and the OS) uses "socks" as URI scheme to
look up a SOCKS proxy. java.net.ProxySelector needs "socket" as
scheme (and internally maps that to "socks" if and when it asks
the OS about the proxies).
Moreover, java.net.ProxySelector may return unresolved addresses,
whereas the Eclipse proxy selector always returns resolved addresses.
Fix both by explicitly resolving unresolved proxy addresses and using
scheme "socket" in the DefaultProxyDataFactory.
Tested manually with the jgit command-line tool using ssh -vvv -D7020
localhost and 3proxy as SOCKS5 proxies on localhost (3proxy with
user/password authentication). Start jgit with _JAVA_OPTIONS set to
"-DsocksProxyHost=<host> -DsocksProxyPort=<port>
-Djava.net.useSystemProxies=false" to test manually.
Bug: 548965
Change-Id: Ib81ae8255ac2f9c48268f172e7d8ebb4a792b66d
Signed-off-by: Thomas Wolf <thomas.wolf@paranor.ch>
Add the constant, and implement hashing of known host names in
OpenSshServerKeyDatabase. Add a test verifying that the hashing
works.
Bug: 548492
Change-Id: Iabe82b666da627bd7f4d82519a366d166aa9ddd4
Signed-off-by: Thomas Wolf <thomas.wolf@paranor.ch>
Provide a wrapper interface and change the implementation such that
a client can substitute its own database of known hosts keys instead
of the default file-based mechanism.
Bug: 547619
Change-Id: Ifc25a4519fa5bcf7bb8541b9f3e2de15215e3d66
Signed-off-by: Thomas Wolf <thomas.wolf@paranor.ch>
The same effect could already be obtained if the ~/.ssh/config file
did not exist. But that is more difficult to control by clients,
since JGit would pick up the config if it was then created. Therefore
allow specifying a null config explicitly to permanently switch off
config file handling.
Change-Id: Iedf8a7f4d5c1ca08e0a513ed28301d8e5261b22a
Signed-off-by: Thomas Wolf <thomas.wolf@paranor.ch>
Reduce the dependency on the ClientSession in preparation to
remove it altogether. Remove the internal helper, re-implement
the needed bits. We have not implemented any configuration
possibility in JGit for creating hashed host names in known hosts
files, so we don't need the sshd code that theoretically would
enable this.
Change-Id: I295f5106b60e1cc3a9d085b0cb7ff747daae88be
Signed-off-by: Thomas Wolf <thomas.wolf@paranor.ch>
We're actually interested only in the known public keys, we don't need
the corresponding host entry from the config.
Change-Id: Ibde6dffe9e3f87bfbb7c70d1f733b2b0e28cad71
Signed-off-by: Thomas Wolf <thomas.wolf@paranor.ch>
* changes:
dfs: Take size as long instead of int
dfs: Read at the aligned position
dfs: Add a position argument
dfs: Move the deeply nested code to its own method
* stable-5.5:
Return a new instance from openSystemConfig and openUserConfig
Change-Id: I3ba9fa8ceced6e9f883fc23afb0e2c812391beac
Signed-off-by: Jonathan Nieder <jrn@google.com>
* stable-5.4:
Return a new instance from openSystemConfig and openUserConfig
Change-Id: If576513ebbd80c7f01b83be5ab2edd436f5739e3
Signed-off-by: Jonathan Nieder <jrn@google.com>
* stable-5.3:
Return a new instance from openSystemConfig and openUserConfig
Change-Id: I5f2f957ff27b7941ab29b2dc1c885fbf0093705a
Signed-off-by: Jonathan Nieder <jrn@google.com>
* stable-5.2:
Return a new instance from openSystemConfig and openUserConfig
Change-Id: I6491549ab13aad2a4c3f8444a090a94a378eccdb
Signed-off-by: Jonathan Nieder <jrn@google.com>
* stable-5.1:
Return a new instance from openSystemConfig and openUserConfig
Change-Id: I93afa16b14e3e2d127705e3378fb703d6c329694
Signed-off-by: Jonathan Nieder <jrn@google.com>
Move the handling of cached user and system config to getSystemConfig
and getUserConfig methods and revert the implementation of
openSystemConfig and openUserConfig to the old stateless
implementation.
This ensures the open methods respect the passed-in parent config, which
may be different on each invocation. Additionally, returning a new
instance matches the behavior of the previous implementation of the
default system reader, which downstream callers may be depending on.
Move the implementation of the new caching methods getSystemConfig and
getUserConfig up to SystemReader. This avoids that we break the ABI for
subclasses of SystemReader.
Also see [1] which fixed a similar problem with Gerrit's custom
SystemReader.
[1] https://gerrit-review.googlesource.com/c/gerrit/+/225458
Change-Id: If54a2491932d8fc914d4649cb73c9e837c5b8ad0
Signed-off-by: Matthias Sohn <matthias.sohn@sap.com>
These warnings were missed to address in a0048208 which introduced them.
Change-Id: Ia2d15fdce72c10378d020682b80fe7fc548c0d4c
Signed-off-by: Matthias Sohn <matthias.sohn@sap.com>
In order to support implementations of CachedPackUriProvider (which need
to supply, among other things, the checksum of the packfile
corresponding to a CachedPack), in a004820858 ("UploadPack: support
custom packfile-to-URI mapping", 2019-08-20), the method getCheckSum()
was added to PackIndex. However, there is no way to access the PackIndex
from a DfsCachedPack.
Therefore, add an accessor for the DfsPackFile stored in the
DfsCachedPack. Now, a user who has a DfsCachedPack can use
DfsCachedPack#getPackFile then DfsPackFile#getPackIndex then
PackIndex#getCheckSum to obtain the checksum of a pack.
Change-Id: Ia010c016f6cac0f058ee20eff4c10f57338bfefc
Signed-off-by: Jonathan Tan <jonathantanmy@google.com>
When firstParent is set, RevWalk traverses only the first parent of a
commit, even though that commit is UNINTERESTING. Since we want the
maximal UNINTERESTING set, we shouldn't prune any parents here. This
issue is apparent only when some of the commits being traversed are
unparsed, since walker.carryFlagsImpl() propagates the UNINTERESTING
flag to all parsed ancestors, masking the issue.
Therefore teach RevWalk to traverse all parents when a commit is
UNINTERESTING and not only the first parent. Since this issue is
masked by commit parsing, also test situations when the commits
involved are unparsed.
Signed-off-by: Alex Spradlin <alexaspradlin@google.com>
Change-Id: I95e2ad9ae8f1f50fbecae674367ee7e0855519b1
Factor out the test parameterization to use both connection factories
into a common super class and use it in more tests.
This made HttpClientTests.testV2HttpSubsequentResponse() fail for
Apache HTTP. The test used the pattern
- create POST connection
- setDoOutput(true)
- connect()
- write output stream
- get & read input stream
This pattern is never used in JGit, which actually calls connect() only
in one case in LFS, and that's on a HEAD request.
The above pattern works on JDK, but fails on Apache HTTP because with
Apache HTTP a connect() actually executes the full request including
writing the entity. To work with Apache HTTP, the pattern would need
to be
- create POST connection
- setDoOutput(true)
- write output stream
- connect()
- get & read input stream
which is fine for both. JDK connects implicitly in getOutputStream()
and treats the later explicit connect() as a no-op, and Apache works
because the entity is written when connect() is called.
Because JDK connects implicitly on getOutputStream(), the following
pattern also works with JDK:
- create POST connection
- setDoOutput(true)
- write output stream
- get & read input stream
Support this with Apache HTTP too: let getInputStream() execute
the request if it wasn't executed already.
Remove explicit connect() calls from test code, since JGit doesn't do
those either.
Change-Id: Ica038c00a7b8edcc01d5660d18e961146305b87f
Signed-off-by: Thomas Wolf <thomas.wolf@paranor.ch>
Make the command die with proper messages when the revision
cannot be resolved or the file doesn't exist in the repository.
Previously the command would throw NPEs in these cases.
Bug: 490798
Change-Id: Ia457347aa22cf6bd2c2b6e7b9d705a66b3826307
Signed-off-by: Thomas Wolf <thomas.wolf@paranor.ch>
Keep the last release and the latest snapshot only.
Change-Id: I48dcc76e1b536f402e41c69fabb5ab46096f6c0a
Signed-off-by: Matthias Sohn <matthias.sohn@sap.com>
* rename 4.12 staging to 4.12
* verified there are no version number changes between 4.12-staging and
new 4.12 final
* create 4.13-staging from former 4.12-staging
* update orbit to latest snapshot
* sort entries in S20190819183153.tpd
Change-Id: I1c3992b1dae2282a8c057e1453a6b63664d09207
Signed-off-by: Matthias Sohn <matthias.sohn@sap.com>
Bazel uses a security manager to sandbox tests, it cannot handle another
security manager.
Change-Id: Id7dd11073178e5e65337414681c1c6e74e14cba7
Signed-off-by: Matthias Sohn <matthias.sohn@sap.com>
b9d2926d missed to add this dependency used in
SeparateClassloaderTestRunner which broke the build in Eclipse.
Change-Id: I7ef79021ad41cabc9f2fa10ac6916eed2745d2c6
Signed-off-by: Matthias Sohn <matthias.sohn@sap.com>
* stable-5.4:
Prepare 5.4.3-SNAPSHOT builds
JGit v5.4.2.201908231537-r
Prepare 5.3.5-SNAPSHOT builds
JGit v5.3.4.201908231101-r
Prepare 5.1.11-SNAPSHOT builds
JGit v5.1.10.201908230655-r
Use AtomicReferences to cache user and system level configs
Fix copy-paste typo in CloneCommand#cleanup
SystemReader: Use correct constructor of FileBasedConfig
Change-Id: I241dd1314f5535147ad641576a25f49cae6fe62d
Signed-off-by: Matthias Sohn <matthias.sohn@sap.com>
It's expected that jgit should work without native git installation.
In such case Security Manager can be configured to deny access to the
files outside of git repository. JGit tries to find cygwin
installation. If Security manager restricts access to some folders
in PATH, it should be considered that those folders are absent
for jgit.
Also JGit tries to detect if symbolic links are supported by OS. If
security manager forbids creation of symlinks, it should be assumed
that symlinks aren't supported.
Bug: 550115
Change-Id: Ic4b243cada604bc1090db6cc1cfd74f0fa324b98
Signed-off-by: Nail Samatov <sanail@yandex.ru>
Matthias Sohn
David Pursehouse
Han-Wen Nienhuys
Masaya Suzuki
Thomas Wolf
Jonathan Nieder
Jonathan Tan
Alex Spradlin
Nail Samatov