Shawn Pearce [Fri, 7 Dec 2012 20:19:13 +0000 (12:19 -0800)]
Disable response compression on /git-receive-pack URLs
Compressing the response with gzip causes the stream to delay
flushing until gzip has seen the entire response message, or buffers
fill up and the compressed data has to be sent. This hides the
resolving progress monitor from the client, as well as any other
progress messages the server might be trying to send.
Disable compression in receive, matching what /git-upload-pack has.
Jonathan Nieder [Tue, 4 Dec 2012 00:08:04 +0000 (16:08 -0800)]
archive: Add tar support
Unlike ZIP files, tar files do not treat symlinks as ordinary files
with a different mode, so tar support involves a little more code than
would be ideal.
Jonathan Nieder [Mon, 3 Dec 2012 18:41:25 +0000 (10:41 -0800)]
archive: Record executable and symlink bits
Setting the mode for a zip entry is now as simple as
"entry.setUnixMode(mode)", so do that.
The test checks using the system's "zipinfo" command (from InfoZIP)
that the mode has been recorded correctly on systems that happen to
have a "zipinfo" command, using org.junit.Assume to distinguish them.
Colby Ranger [Mon, 3 Dec 2012 20:23:14 +0000 (12:23 -0800)]
Fix DeltaWindow.clear() to release loaded buffer bytes.
It is possible for the buffer to be set but not the index. It
ocurrs when an exception occurs during creating an index, but
after the buffer is loaded. Furthermore, the cleared DeltaWindowEntry
should have been ent and not res.
Jonathan Nieder [Mon, 3 Dec 2012 17:55:02 +0000 (09:55 -0800)]
[RFC] archive: Switch to commons-compress as ZIP archiver
The Apache Commons Compress library provides a similar interface to
java.util.zip with some features not found in java.util.zip, including
support for inclusion of metadata (file mode and symlink targets) and
support for multiple file formats (zip, .tar.xz, etc).
Use it, in preparation for making use of these features. No
functional change intended yet.
A previous version of this patch used plexus-archiver. That is a
heavier-weight dependency and offers a less convenient interface.
Thanks to James Moger and Chris Aniszczyk for advice.
Robin Stocker [Sat, 1 Dec 2012 22:51:09 +0000 (23:51 +0100)]
Enable reuse of FileTreeIterator and FileEntry for non subclasses
For EGit change I2c41d86b8b74c2a334433de1bbfed5b36af872bf,
ContainerTreeIterator also needs to create entries for File objects in
case of filtered resources. Instead of reimplementing FileEntry there,
make the constructor public so that it can be reused.
Also allow to pass a WorkingTreeIterator instead of a FileTreeIterator
in FileTreeIterator's constructor, which is enough and allows to pass
other subclasses.
Bug: 358901
Change-Id: Ie0f9c9434ef7d73a8d73d4fe46db4147ded1d267 Signed-off-by: Chris Aniszczyk <zx@twitter.com>
Jonathan Nieder [Thu, 29 Nov 2012 01:48:25 +0000 (17:48 -0800)]
add: Fix bad metaVar reference in <filepattern> help
Without this patch, "jgit add --help" throws two exceptions
(NoSuchFieldException: metavar_filepattern; MissingResourceException:
Can't find resource for bundle java.util.PropertyResourceBundle, key
metavar_filepattern) instead of printing a usage message, due to a
small typo and a missing declaration.
Change-Id: I2e1c9c3cc8cddfa12ca4bfb9afc4df0492a9fce1 Signed-off-by: Robin Rosenberg <robin.rosenberg@dewire.com> Signed-off-by: Chris Aniszczyk <zx@twitter.com>
Jonathan Nieder [Fri, 16 Nov 2012 02:48:12 +0000 (18:48 -0800)]
Add "jgit archive" tool that writes a tree as a ZIP file
C Git's "git archive" command represents a tree object using a
standard archival format like tar, zip, or tgz, ready for consumption
by other, git-unaware users or tools.
Add a bare-bones analagous "jgit archive" command to show what is
possible, supporting only ZIP format for now. It uses java.util.zip
which is not aware of the InfoZIP extensions for representing symlinks
and file permissions, so symlinks, executable files, and submodule
entries are represented as plain text files.
Making this functionality available from the library, improving
handling of special entries, and support for other output formats are
left for later patches. Ultimately the intent is to offer a
TreeArchiveStream class for use by web frontends like Gitiles to offer
"download as zip/tgz/txz" links and use by, for example, code search
tools to get easy access to the content of git tree objects.
Test with "jgit archive my-favorite-tree >out.zip".
Matthias Sohn [Sat, 3 Nov 2012 00:32:52 +0000 (01:32 +0100)]
Import non-java.* JRE packages
Otherwise loading javax.net.ssl.TrustManager fails if
osgi.compatibility.bootdelegation=false which became the Equinox default
since bug 344850 was fixed.
Bug: 392056
Change-Id: I464871723649095942dbf77da93890ac8ec39075 Signed-off-by: Matthias Sohn <matthias.sohn@sap.com> Signed-off-by: Chris Aniszczyk <zx@twitter.com>
Markus Duft [Fri, 13 Jul 2012 06:31:10 +0000 (08:31 +0200)]
Adapt Status and CleanCommand to support cleaning directories
This adds the possibility to:
* retrieve untracked directories from the status
* instruct the CleanCommand to clean those directories.
* retrieve ignored paths from the status
* instruct the CleanCommand to leave those ignored paths alone
Bug: 338717
Change-Id: Ibed0459005a5e306c010b9932f5b5fd107fb5448 Signed-off-by: Chris Aniszczyk <zx@twitter.com>
Robin Stocker [Sat, 22 Sep 2012 23:57:20 +0000 (01:57 +0200)]
CheckoutCommand: Support checking out ours and theirs
The checkoutPaths body is split into two implementations, depending on
whether we are checking out the index or a branch. This improves
readability, as in the index case we now also need to have access to
DirCacheIterator.
Bug: 390147
Change-Id: I99fd599b25b2ace9bdd84535a56565286a3cb7f1 Signed-off-by: Chris Aniszczyk <zx@twitter.com>
Robin Rosenberg [Wed, 14 Nov 2012 23:53:53 +0000 (00:53 +0100)]
Make an exception for the formatting rules (turn off) for some files
Our rule to enforce javodocs for public members gives us a problem
because there are some patterns where javadoc make little sense so we
make the comments as small as possible, which our formatting rules do
not like, so disable it for those source files.
Change-Id: I6e3edb1e650ed45428b89cf41e6151b6536bca8a Signed-off-by: Chris Aniszczyk <zx@twitter.com>
Robin Rosenberg [Tue, 13 Nov 2012 01:31:54 +0000 (02:31 +0100)]
Harmonize the JDT settings within JGit
Note the the settings are slightly less restrictive for test bundles.
-Also cleanup a couple of malformed javadocs
-Update compiler warnings/errors to include default values from Juno
-We now flag diagnosed null dereference as error. We didn't do that
earlier because of some false positives.
Change-Id: I58386d63164e65d3d8d1998da3390d99bdc7381a Signed-off-by: Chris Aniszczyk <zx@twitter.com>
Matthias Sohn [Fri, 16 Nov 2012 01:04:58 +0000 (02:04 +0100)]
Add missing @since tags to mark API added in 2.2
Change-Id: I458167739210214fa54c4b3d62fac5abc82f96f7 Signed-off-by: Matthias Sohn <matthias.sohn@sap.com> Signed-off-by: Chris Aniszczyk <zx@twitter.com>
Jonathan Nieder [Wed, 14 Nov 2012 20:14:20 +0000 (12:14 -0800)]
diff, log -p: do not use outs before it is initialized
Since commit caa362f20df1 (Check for write errors in standard out and
exit with error, 2012-09-14), running "jgit diff" results in a
NullPointerException:
| $ jgit diff
| java.lang.NullPointerException
| at java.io.BufferedOutputStream.flushBuffer(BufferedOutputStream.java:82)
| at java.io.BufferedOutputStream.write(BufferedOutputStream.java:126)
| at org.eclipse.jgit.diff.DiffFormatter.format(DiffFormatter.java:688)
| at org.eclipse.jgit.diff.DiffFormatter.format(DiffFormatter.java:630)
| at org.eclipse.jgit.diff.DiffFormatter.format(DiffFormatter.java:616)
| at org.eclipse.jgit.diff.DiffFormatter.format(DiffFormatter.java:600)
| at org.eclipse.jgit.pgm.Diff.run(Diff.java:211)
| at org.eclipse.jgit.pgm.TextBuiltin.execute(TextBuiltin.java:166)
| at org.eclipse.jgit.pgm.Main.execute(Main.java:200)
| at org.eclipse.jgit.pgm.Main.run(Main.java:120)
| at org.eclipse.jgit.pgm.Main.main(Main.java:94)
That patch replaced most uses of System.out with a wrapper, with
changes like the following:
class Diff extends TextBuiltin {
private final DiffFormatter diffFmt = new DiffFormatter( //
- new BufferedOutputStream(System.out));
+ new BufferedOutputStream(outs));
outs is not set for TextBuiltin objects until init() has been run.
Moving the initialization to after the super.init() call gets
"jgit diff" and "jgit log -p" working well again.
Robin Stocker [Sat, 20 Oct 2012 20:00:13 +0000 (22:00 +0200)]
CommitCommand: Ensure unmerged paths are added correctly with setOnly
With bug 391855, PathEdit will be changed to apply an edit for each
stage. With that, CommitCommand would no longer work correctly when
committing an unmerged path.
This changes it to use a DirCacheBuilder which allows us to correctly
replace the entries for the three stages with one, which is not possible
with PathEdit.
Robin Stocker [Sun, 28 Oct 2012 14:57:43 +0000 (15:57 +0100)]
ResetCommand: Use DirCacheBuilder in resetIndex
With bug 391855, DirCacheEditor's PathEdit will be applied for each
stage. For an unmerged path, this would result in 3 equal entries for
the same path.
By using a DirCacheBuilder, the code is simpler and does not have the
above problem with unmerged paths.
Bug: 391860
Change-Id: I785deeaeb8474f8c7a7fbc9ef00d3131fac87e41 Signed-off-by: Chris Aniszczyk <zx@twitter.com>
Robin Rosenberg [Wed, 10 Oct 2012 06:02:30 +0000 (08:02 +0200)]
Suppress resource warnings with Java 7
For streams that should not be closed, i.e. don't own an underlying
stream, and in-memory streams that do not need to be closed we just
suppress the warning. This mostly apply to test cases. GC is enough.
For streams with external resources (i.e. files) we add the necessary
call to close().
Shawn O. Pearce [Wed, 17 Oct 2012 14:55:15 +0000 (07:55 -0700)]
Simplify push error message when ref already exists
If a client attempts to create a branch that already exists on the
remote side, tell them "already exists" rather than repeat lots of
information about the reference. Previously the error looked like:
Robin Rosenberg [Mon, 8 Oct 2012 04:27:30 +0000 (06:27 +0200)]
Add a hint on the purpose of JGitInternalException to the constructor
In code review we often see uses of JGitInternalException where a high
level GitAPIException would be more appropriate. Hopefully the word
low-level in the constructor comment will lead to fewer such cases.
Robin Rosenberg [Sat, 6 Oct 2012 11:08:16 +0000 (13:08 +0200)]
Fix for Iff768422c, use offset 0 when going back to work tree iterator
In Iff768422c the offset used for the content id was fixed to use the
offset that applied to the dircache iterator. Unfortunately the index
for the dircache content id offset stuck for entries that were not in
the index. Few caller probably cared about that, unless it actually
caused an ArrayIndexOutOfBoundsException.
Robin Rosenberg [Tue, 2 Oct 2012 10:45:29 +0000 (12:45 +0200)]
Fix unstable test
Add more time to a timestamp to safely go beyond the granularity of
the timestamp resolution of the file system. The lowest resolution we
know of is FAT with two second resolution. Then add some to make sure we
are above the limit.
Markus Keller [Wed, 26 Sep 2012 22:17:53 +0000 (00:17 +0200)]
Fix compile error (method not available in 1.5)
Change-Id: I07aca821010daca75a66506b9ca738bc8c262abb Signed-off-by: Markus Keller <markus_keller@ch.ibm.com> Signed-off-by: Matthias Sohn <matthias.sohn@sap.com>
07f9 seems to require a more recent args4j, but I cannot locate a CQ
that verifies we can use a version more recent then 2.0.12. 2.0.16
has been released, but the Hudson CI instance at Eclipse won't
build it.
Since the commit fixes an issue identified in March but wasn't
actually submitted to the tree until September, we can continue to
ignore whatever the problem is/was until someone can attempt a more
correctly working solution.
projects / jgit.git / log