Matthias Sohn [Thu, 3 Dec 2020 01:04:36 +0000 (02:04 +0100)]
[spotbugs] Fix potential NPE in WorkingTreeIterator#isModified
File#list can return null. Fix the potential NPE by using Files#list
which is also faster since it retrieves directory entries lazily while
File#list retrieves them eagerly.
Change-Id: Idf4bda398861c647587e357326b8bc8b587a2584 Signed-off-by: Matthias Sohn <matthias.sohn@sap.com>
Matthias Sohn [Thu, 3 Dec 2020 00:44:58 +0000 (01:44 +0100)]
[spotbugs] Fix potential NPE in FileRepository#convertToReftable
File#listFiles can return null. Use Files#list which does not return
null and should be faster since it's returning directory entries lazily
while File#listFiles fetches them eagerly.
Change-Id: I3bfe2a52278244fc469143692c06b05d9af0d0d4 Signed-off-by: Matthias Sohn <matthias.sohn@sap.com>
If TransportException is thrown in the finally block of
downloadPackedObject() ensure that the exception handled in the previous
catch block isn't suppressed.
Change-Id: I23982a2b215e38f681cc1719788985e60232699a Signed-off-by: Matthias Sohn <matthias.sohn@sap.com>
Martin Fick [Thu, 26 Apr 2018 16:53:57 +0000 (10:53 -0600)]
Split out loose object handling from ObjectDirectory
The ObjectDirectory class manages the interactions for the entire object
database, this includes loose objects, packfiles, alternates, and
shallow commits. To help reduce the complexity of this class, abstract
some of the loose object specific details into a class which understands
just this, leaving the ObjectDirectory to focus more on the interactions
between the different mechanisms.
Change-Id: I39f3a74d6308f042a2a2baa57769f4acde5ba5e0 Signed-off-by: Martin Fick <mfick@codeaurora.org> Signed-off-by: Matthias Sohn <matthias.sohn@sap.com>
Martin Fick [Wed, 25 Apr 2018 17:59:21 +0000 (11:59 -0600)]
Split out packfile handling from ObjectDirectory
The ObjectDirectory class manages the interactions for the entire object
database, this includes loose objects, packfiles, alternates, and
shallow commits. To help reduce the complexity of this class, abstract
some of the packfile specific details into a class which understands
just this, leaving the ObjectDirectory to focus more on the interactions
between the different mechanisms.
Change-Id: I5cc87b964434b0afa860b3fe23867a77b3c3a4f2 Signed-off-by: Martin Fick <mfick@codeaurora.org> Signed-off-by: Matthias Sohn <matthias.sohn@sap.com>
Thomas Wolf [Tue, 8 Dec 2020 14:45:35 +0000 (15:45 +0100)]
TagCommand: propagate NO_CHANGE information
Some clients may wish to allow NO_CHANGE lightweight tag updates
without setting the force flag. (For instance EGit does so.)
Command-line git does not allow this.
Propagate the RefUpdate result via the RefAlreadyExistsException.
That way a client has the possibility to catch it and check the
failure reason without having to parse the exception message, and
take appropriate action, like ignoring the exception on NO_CHANGE.
Change-Id: I60e7a15a3c309db4106cab87847a19b6d24866f6 Signed-off-by: Thomas Wolf <thomas.wolf@paranor.ch>
Thomas Wolf [Sat, 5 Dec 2020 21:01:25 +0000 (22:01 +0100)]
TagCommand: support signing annotated tags
Add the two config constants from C git that can switch on signing
of annotated tags. Add them to the GpgConfig, and implement actually
signing a tag in TagCommand.
The interactions between command line options for "git tag" and config
options is a bit murky in C git. There are two config settings for it:
* tag.gpgSign is the main option, if set to true, it kicks in if
neither -s nor -u are given on the command line.
* tag.forceSignAnnotated signs only tags created via "git tag -m",
but only if command-line option "-a" is not present. It applies
even if tag.gpgSign is set explicitly to false.
Giving -s or -u on the command line also forces an annotated tag
since lightweight tags cannot be signed.
Bug: 386908
Change-Id: Ic8a1a44b5f12f47d5cdf3aae2456c1f6ca9ef057 Signed-off-by: Thomas Wolf <thomas.wolf@paranor.ch>
Matthias Sohn [Fri, 2 Oct 2020 19:54:53 +0000 (21:54 +0200)]
Don't install 3rd party dependency bundles via features
Instead provide them only in the p2 repository. This way they are
available when installing from the jgit p2 repository but we are not
enforcing the version we bring but can also use the version available in
Eclipse if it matches our requirements.
Bug: 514326
Bug: 566475
Change-Id: I3e8d0bad12cfb0c1003ade3e6f13e9af35626f14 Signed-off-by: Matthias Sohn <matthias.sohn@sap.com>
Matthias Sohn [Wed, 2 Dec 2020 14:49:35 +0000 (15:49 +0100)]
Merge branch 'master' into stable-5.10
* master:
Prepare 5.3.10-SNAPSHOT builds
JGit v5.3.9.202012012026-r
Prepare 5.1.16-SNAPSHOT builds
JGit v5.1.15.202012011955-r
Fix PackInvalidException when fetch and repack run concurrently
Upgrade maven-pmd-plugin to 3.14.0
Update Orbit to R20201130205003 for 2020-12
Use new protocol version constants
PacketLineInTest: test for END and DELIM being distinguishable
Add constants for parsing git wire protocol version
Ignore missing javadoc tags in test bundles
Bazel: Allow to build and run the tests with JDK 15
[releng] japicmp: update last release version
Add support for reading symrefs from pack capabilities
Change-Id: I5afbbb912f502991f0cf9c2501b024f5f00144ba Signed-off-by: Matthias Sohn <matthias.sohn@sap.com>
Petr Hrebejk [Mon, 30 Nov 2020 19:19:53 +0000 (20:19 +0100)]
Fix PackInvalidException when fetch and repack run concurrently
We are running several servers with jGit. We need to run repack from
time to time to keep the repos performant. I.e. after push we test how
many small packs are in the repo and when a threshold is reached we run
the repack.
After upgrading jGit version we've found that if someone does the clone
at the time repack is running the clone sometimes (not always) fails
because the repack removes .pack file used by the clone. Server
exception and client error attached.
I've tracked down the cause and it seems to be introduced between jGit
5.2 (which we upgraded from) and 5.3 and being caused by this commit:
Move throw of PackInvalidException outside the catch -
https://github.com/eclipse/jgit/commit/afef866a44cd65fef292c174cad445b3fb526400
The problem is that when the throw was inside of the try block the last
catch block catched the exception and called openFailed(false) method.
It is true that it called it with invalidate = false, which is wrong.
The real problem though is that with the throw outside of the try block
the openFail is not called at all and the fields activeWindows and
activeCopyRawData are not set to 0. Which affects the later called tests
like: if (++activeCopyRawData == 1 && activeWindows == 0).
The fix for this is relatively simple keeping the throw outside of the
try block and still having the invalid field set to true. I did
exhaustive testing of the change running concurrent clones and pushes
indefinitely and with the patch applied it never fails while without the
patch it takes relatively short to get the error.
David Ostrovsky [Sun, 29 Nov 2020 12:06:00 +0000 (13:06 +0100)]
Add constants for parsing git wire protocol version
This would allow other JGit users to access and reuse the constants.
Change-Id: I1608802f45586af5f8582afa592e26679e9cebe3 Signed-off-by: David Ostrovsky <david@ostrovsky.org> Signed-off-by: Matthias Sohn <matthias.sohn@sap.com>
Matthias Sohn [Thu, 19 Nov 2020 00:55:35 +0000 (01:55 +0100)]
Ignore missing javadoc tags in test bundles
It seems Eclipse 4.18 reports them as error whereas earlier versions
ignored this maybe since we don't require javadoc comments for all the
test bundles.
Change-Id: I3f4d42ce681ea5c2b4b302991d2641290ac8561d Signed-off-by: Matthias Sohn <matthias.sohn@sap.com>
David Ostrovsky [Mon, 23 Nov 2020 19:04:07 +0000 (20:04 +0100)]
Bazel: Allow to build and run the tests with JDK 15
To avoid JDK specific bugs in future, like Bug: 568950, and given that
upcoming Bazel release 4.0.0 added support to JDK 15 java toolchain,
add definition for remote JDK 15 to WORKSPACE file and add build and
test instructions.
To build and execute the tests with JDK 15 on Linux run:
Lee Worrall [Sat, 10 Oct 2020 03:31:19 +0000 (11:31 +0800)]
Add support for reading symrefs from pack capabilities
A SymbolicRef is added to the advertised refs for any symref in
capabilities whose target is an advertised ref; this may replace an
existing entry, such as HEAD.
When cloning, if any advertised HEAD is symbolic then use the target
rather than looking for an advertised ref with a matching objectId.
Add --symref option to LsRemote command.
Bug: 514052
Change-Id: Idfb48e6f6e8dcfe57a6896883fe6d84d533aa9d0 Signed-off-by: Lee Worrall <worrall.la@gmail.com>
Matthias Sohn [Wed, 25 Nov 2020 16:00:09 +0000 (17:00 +0100)]
Ensure that GC#deleteOrphans respects pack lock
If pack or index files are guarded by a pack lock (.keep file)
deleteOrphans() should not touch the respective files protected by the
lock file. Otherwise it may interfere with PackInserter concurrently
inserting a new pack file and its index.
The problem was caused by the following race.
All mentioned files are located in "objects/pack/".
File endings relevant in "pack" dir:
.pack
.keep
.idx
.bitmap
When ReceivePack receives a pack file it executes the following steps:
ReceivePack.service():
receivePackAndCheckConnectivity():
receivePack():
receive the pack
parse the pack, returns packLock (.keep file)
PackInserter.flush():
write tmpPck file: "insert_<random>.pack"
write tmpIdx file: "insert_<random>.idx"
real pack name: "pack-<SHA1>.pack"
real index name: "pack-<SHA1>.idx"
atomic rename tmpPack to realPack
atomic rename tmpIdx to tmpIdx
execute commands
unlock pack by removing .keep file
trigger auto gc if enabled
When PackInserter.flush() renames the temporary pack to the final
"pack-xxx.pack" file the temporary pack index file "insert_xxx.idx"
has no matching .pack file with the same base name for a short interval.
If deleteOrphans() ran during that interval it deduced the pack index
file was orphaned. Subsequently the missing pack index caused
MissingObjectExceptions since objects contained in the pack couldn't be
looked up anymore.
Bug: https://bugs.chromium.org/p/gerrit/issues/detail?id=13544
Change-Id: I559c81e4b1d7c487f92a751bd78b987d32c98719 Signed-off-by: Matthias Sohn <matthias.sohn@sap.com>
Matthias Sohn [Wed, 25 Nov 2020 16:39:25 +0000 (17:39 +0100)]
Merge branch 'master' into stable-5.10
* master:
Update Orbit to S20201118210000 and add target for 4.18
PacketLineIn: ensure that END != DELIM
PacketLineIn: ensure that END != DELIM
Allow to resolve a conflict by checking out a file
Update Orbit to I20201111205634
Document that setLastModified sets time of symlink target
Fix bug in PerformanceLogContext
Fix IOException occurring during gc
Change-Id: I2980552381d1ae61b9b2d81d7289de37d6bf4cae Signed-off-by: Matthias Sohn <matthias.sohn@sap.com>
Thomas Wolf [Thu, 19 Nov 2020 12:23:05 +0000 (13:23 +0100)]
PacketLineIn: ensure that END != DELIM
Just allocate the two string objects directly. The previously used
new StringBuilder(0).toString() returns the same object for both END
and DELIM when run on Java 15, which breaks the wire protocol since
then END and DELIM cannot be distinguished anymore.
Bug: 568950
Change-Id: I9d54d7bf484948c24b51a094256bd9d38b085f35 Signed-off-by: Thomas Wolf <thomas.wolf@paranor.ch>
(cherry picked from commit 7da0f0a8f37e35e9c3108588f1e6f7a7381d8f77)
Thomas Wolf [Thu, 19 Nov 2020 12:23:05 +0000 (13:23 +0100)]
PacketLineIn: ensure that END != DELIM
Just allocate the two string objects directly. The previously used
new StringBuilder(0).toString() returns the same object for both END
and DELIM when run on Java 15, which breaks the wire protocol since
then END and DELIM cannot be distinguished anymore.
Bug: 568950
Change-Id: I9d54d7bf484948c24b51a094256bd9d38b085f35 Signed-off-by: Thomas Wolf <thomas.wolf@paranor.ch>
(cherry picked from commit 7da0f0a8f37e35e9c3108588f1e6f7a7381d8f77)
Thomas Wolf [Thu, 19 Nov 2020 12:23:05 +0000 (13:23 +0100)]
PacketLineIn: ensure that END != DELIM
Just allocate the two string objects directly. The previously used
new StringBuilder(0).toString() returns the same object for both END
and DELIM when run on Java 15, which breaks the wire protocol since
then END and DELIM cannot be distinguished anymore.
Bug: 568950
Change-Id: I9d54d7bf484948c24b51a094256bd9d38b085f35 Signed-off-by: Thomas Wolf <thomas.wolf@paranor.ch>
Thomas Wolf [Sun, 15 Nov 2020 17:45:12 +0000 (18:45 +0100)]
Allow to resolve a conflict by checking out a file
DirCacheEditor unconditionally applied a PathEdit to all stages in the
index. This gives wrong results if one wants to check out a file from
some commit to resolve a conflict: JGit would update the working tree
file multiple times (once per stage), and set all stages to point to
the checked-out blob.
C git replaces the stages by the entry for the checked-out file.
To support this, add a DirCacheEntry.setStage() method so that
CheckoutCommand can force the stage to zero. In DirCacheEditor, keep
only the zero stage if the PathEdit re-set the stage.
Bug: 568038
Change-Id: Ic7c635bb5aaa06ffaaeed50bc5e45702c56fc6d1 Signed-off-by: Thomas Wolf <thomas.wolf@paranor.ch>
projects / jgit.git / log