should ban System.exit via the policy file. Unfortunately System.exit is
used by the runner itself too (otherwise it wouldn't be able to exit the
JVM in the presence of background threads) so the policy would need to be
selective and either ban certain packages or revoke the grant for the
runner.
git-svn-id: https://svn.apache.org/repos/asf/lucene/dev/trunk@1380200 13f79535-47bb-0310-9956-ffa450edef68
SOLR-3772: On cluster startup, we should wait until we see all registered replicas before running the leader process or if they all do not come up, N amount of time.
SOLR-3752: When a leader goes down, have the Overseer clear the leader state in cluster.json
SOLR-3751: Add defensive checks for SolrCloud updates and requests that ensure the local state matches what we can tell the request expected.
git-svn-id: https://svn.apache.org/repos/asf/lucene/dev/trunk@1379002 13f79535-47bb-0310-9956-ffa450edef68
test this cleaner, we don't need a nested test, just want to ensure MDW throws exception on close
........
SOLR-3730: Rollback is not implemented quite right and can cause corner case fails in SolrCloud tests.
........
SOLR-3690: Fixed binary release packages to include dependencies needed for the solr-test-framework
........
SOLR-2981: Fixed StatsComponent to no longer return duplicated information when requesting multiple stats.facet fields
........
SOLR-3690: need to copy the license + sha1 files to be included in solr artifacts
........
LUCENE-4308: don't delete a test's files if the test fails
........
LUCENE-4308: clean up imports, move below license
........
add more missing javadocs
........
javadocs
........
add hostContext to solr.xml by default
........
SOLR-2857: restore /update/json and /update/csv
........
improve javadocs
........
use assertEquals so we will see whats going on when this fails
........
added javadocs to SortedBytesMergeUtils public inner classes
........
added javadocs to SortedBytesMergeUtils public inner classes
........
fix javadocs
........
javadocs
........
Better verbosity, also fix bug in while loop with char==0.
I have one explanation for this: Maybe Apaches SVN server returns something (like error message) that's not UTF8?
........
javadocs
........
Fix bug in test by using a local file instead of HTTP URL. Also fix unclosed file in FileStream.getContentType(), remove dead code.
........
Don't use IOUtils in SolrJ! Sorry, hope maven build did not start in this time frame *g*
........
SOLR-3737: fix Stempel factory resource loading, use singleton instance
........
javadocs
........
enforce all lucene core classes have javadocs
........
LUCENE-4310: non-BMP characters were failing to match with MappingCharFilter
........
LUCENE-4310: revert ... new test is failing
........
LUCENE-4310: non-BMP characters were failing to match with MappingCharFilter
........
Rewrite the copying loop slightly; scratch.grow has an assert offset == 0
so moved this reset up before grow.
........
SOLR-3715: fix FastOutputStream, FastWriter when buffer sizes can change
........
license
........
SOLR-3691: SimplePostTool: Mode for indexing a web page
........
SOLR-3707: Upgrade Solr to Tika 1.2
........
SOLR-3691: Fixed javadocs-lint errors
........
use hasNext not .next() == null to detect empty iterator
........
2nd fix: The same problem had size()
........
Solr extraction contrib should have compile time dependency on lucene, not just test.
........
Fix javadocs of Fields.size(), use FilteredIterator from StoredFields branch. Return -1 for Fields.size() in FieldFilterAtomicReader. If this breaks some test, we found a bug and have to change test to respect -1 (which is valid return value)
........
Fix javadocs
........
git-svn-id: https://svn.apache.org/repos/asf/lucene/dev/branches/lucene3312@1374718 13f79535-47bb-0310-9956-ffa450edef68
move class to o.a.l.index and make pkg-private, silence java7 warnings
........
SOLR-3727: improve solrcloud close/shutdown
........
LUCENE-4303 Use ResourceLoader not Class.forName
........
LUCENE-4302: Fix facet userguide to have HTML loose doctype like all other javadocs
........
LUCENE-4152: add one syllable method to enumerate subreaders
........
LUCENE-4152: use simpler sugar
........
SOLR-3670: New CountFieldValuesUpdateProcessorFactory
........
SOLR-3727: improve solrcloud close/shutdown
........
Ignore process reaper (forked subprocess on unixish systems).
........
document what IW.close is doing and how to 'actually close' if you want that
........
LUCENE-4304: remove PayloadProcessorProvider, map facet ords using a FilterAtomicReader
........
LUCENE-4304: Make CompositeReader.getSequentialSubReaders() protected
........
solrcloud: improve some logging, improve some testing, other minor tweaks
........
LUCENE-4307: rename IR.getTopReaderContext to IR.getContext
........
SOLR-3572: Make schema-browser show custom similarities
........
better failures from MDW if you have unclosed indexwriter, even if you setLockFactory
........
Improve Javadocs, fix wrong getContext() javadocs (contained outdated info, leaves() throws UOE if not top-level), remove lucene.experimental from contexts, as this is now the official API to walk reader tree.
........
git-svn-id: https://svn.apache.org/repos/asf/lucene/dev/branches/lucene3312@1373337 13f79535-47bb-0310-9956-ffa450edef68
Add support to also reload HashFunctions when Solr boots (otherwise codecs using new hash functions may fail to load).
We may need a better "automatic" reloading with one method call, that reloads all NamedSPILoaders. I will think about it and open issue. Solr could then only call NamedSPILoader.reloadAll(), so all instantiated ones get reloaded automatically. Currently its to risky to add new SPIs without Solr support (because it cannot be tested).
........
LUCENE-4298: MultiFields.getTermDocsEnum(Reader,Bits,String,BytesRef) did not work at all
........
SOLR-3715: remove sync around tlog serialization
........
fix test msg
........
add recovery strat to commented out list of log configs for tests
........
change log format - core= rather than Core:
........
add some tests mixing up payload options
........
tests: specify a seed to Random()
........
tests: use random seed of doc base
........
set the SVN variables so this task can build the unsigned release
........
Remove smoke again, this is now a new ReleaseNightly job on ASF Jenkins (without scripts, only ANT). I will later remove the artifact building from the standard nightly scripts, too. Those will now only run tests and clover.
........
Smoketester should also run with OpenJDK to be able to run on FreeBSD
........
clean up BS2's Coordinator, use ctor instead of init, make coordFactors final
........
LUCENE-4300: BooleanQuery's rewrite was unsafe if coord(1,1) != 1
........
SOLR-3725: fix package-local-src-tgz to not bring in jars/binary content
........
LUCENE-4299: add Terms hasPositions/hasOffsets, so you know what features a docs TVs have
........
ignore silly bugs in sun's regex impl
........
LUCENE-4187: Die, shell scripts, die, die, die!
........
LUCENE-4187: Allow disabling validation tasks and javadocs-linting from jenkins
........
nuke useless scripts - part #1
........
LUCENE-4187: Merge clover reports of Lucene/Solr for easier and more useful Jenkins reporting
........
LUCENE-4187: Remove remaining clover reference
........
LUCENE-4187: Add jenkins clover task
........
LUCENE-4187: Nuke remaining shell scripts
........
LUCENE-4187: exclude backwards (when it comes)
........
LUCENE-4187: simplier pattern
........
LUCENE-4187: Fix title for clover
........
LUCENE-4187: Make clover configuration for jenkins fixed and faster
........
Don't allow negatives in the positions file
........
LUCENE-1888: added option to store payloads on the term vectors
........
throw IllegalArgumentException if you try to index term vector offsets/positions when term vectors are not indexed, or if you try to index term vectors when field is not index
........
fix test to actually compare index1 vs index2, not index1 against itself
........
LUCENE-3985: Upgrade to randomizedtesting 2.0.0. Added support for thread leak detection. Added support for suite timeouts. (Dawid Weiss)
........
fix test bug
........
re-enable test
........
Allow better debugging this test
........
fix clustering and DIH problems with conflicting servlet-api dependencies by making scope be provided in solr's test-framework POM
........
revisit payloads API in DocsAndPositionsEnum
........
maven build: reduce solr test logging output, to allow Jenkins mailer to succeed
........
disable test timeouts until nightly builds have been addressed
........
Fix test failure caused in Maven, because test-framework classes are packaged in JAR.
........
git-svn-id: https://svn.apache.org/repos/asf/lucene/dev/branches/lucene3312@1372438 13f79535-47bb-0310-9956-ffa450edef68
We may need a better "automatic" reloading with one method call, that reloads all NamedSPILoaders. I will think about it and open issue. Solr could then only call NamedSPILoader.reloadAll(), so all instantiated ones get reloaded automatically. Currently its to risky to add new SPIs without Solr support (because it cannot be tested).
git-svn-id: https://svn.apache.org/repos/asf/lucene/dev/trunk@1371150 13f79535-47bb-0310-9956-ffa450edef68
SOLR-3259: default /get to json
........
minor tweaks to update script example
........
LUCENE-4268: Rename ResourceAsStreamReasourceLoader to ClasspathResourceLoader, provide FilesystemResourceLoader, bug fixing
........
LUCENE-4268: Fix test bug
........
SOLR-3648: Fix Velocity template loading in SolrCloud mode
........
fix confusing IW infoStream message
........
Fix rawtypes warning in java 7 and 8, make the SuppressWarnings more local
........
Nicer solution to generic array creation (still problematic in Java 6, but correct in Java 7 if done this way)
........
Disable test failing with Java 8
........
Allow detecting of Java 8
........
LUCENE-4109: BooleanQueries are not parsed correctly with the flexible queryparser
........
LUCENE-4269: remove BalancedSegmentMergePolicy (use TieredMergePolicy instead)
........
LUCENE-4269: deprecate BalancedSegmentMergePolicy (use TieredMergePolicy instead)
........
LUCENE-4190: restrict allowed filenames to reduce risk of deleting non-lucene file from the index directory
........
fix the monkey: connection loss and expiration cause NPE
........
upgrade checkJavaDocs.py to python3
........
LUCENE-3884: Move ElisionFilter out of .fr package
........
fix encoding in javadocs checker
........
LUCENE-2501: fix thread hazard when threads add same field with different IndexOptions at the same time
........
git-svn-id: https://svn.apache.org/repos/asf/lucene/dev/branches/lucene3312@1371142 13f79535-47bb-0310-9956-ffa450edef68
move getRecentUpdates into try block - we should continue with no updates if there is an exception getting them
move cancelRecovery call into recovery wait sync block
git-svn-id: https://svn.apache.org/repos/asf/lucene/dev/trunk@1371030 13f79535-47bb-0310-9956-ffa450edef68
Chris M. Hostetter
Yonik Seeley
Robert Muir
Steven Rowe
Stefan Matheis
Adrien Grand
David Wayne Smiley
Sami Siren
Mark Robert Miller
Dawid Weiss
Uwe Schindler
Erik Hatcher
Jan Høydahl
Greg Bowyer
Michael McCandless