Permalink
Switch branches/tags
Commits on Sep 27, 2012
  1. Added the traversal matcher, which uses the new traversal framework a…

    systay committed Sep 15, 2012
    …bilities to do pattern matching
Commits on Sep 23, 2012
  1. Made it generic

    tinwelint authored and systay committed Sep 17, 2012
  2. * InitialStateFactory needs to be reversed for proper support in

    tinwelint authored and systay committed Sep 14, 2012
      bidirectional traversal, so InitialBranchState extends
      InitialStateFactory and adds #reverse()
    * TraversalBranch#state() added to support proper filtering in collision
      detection
Commits on Sep 21, 2012
  1. Merge pull request #869 from tinwelint/rel-load-problem

    johan-neo committed Sep 21, 2012
    Fixes a race condition where more than one thread loads the relationship...
  2. Fixes a race condition where more than one thread loads the relations…

    tinwelint committed Sep 21, 2012
    …hip chain of the same
    
    NodeImpl at the the same time. Problem being each threads view whether or not to load more
    of reevaluate its iterators for updates from other threads is updated in the wrong order
    compared to the view of the relationship cache for that node.
  3. Merge pull request #865 from nawroth/separatefiles

    johan-neo committed Sep 21, 2012
    Write generated docs to separate files to make translations easier.
Commits on Sep 20, 2012
  1. Merge pull request #867 from systay/synchronized-parser

    jexp committed Sep 20, 2012
    Fixes problem when the Cypher parser is run concurrently
Commits on Sep 17, 2012
Commits on Sep 14, 2012
  1. Merge pull request #851 from jakewins/rm-deprecated-1.9

    peterneubauer committed Sep 14, 2012
    Removed deprecated things
Commits on Sep 13, 2012
  1. typo

    peterneubauer committed Sep 13, 2012
  2. Merge pull request #852 from nawroth/autoindexingconfig

    peterneubauer committed Sep 13, 2012
    Get the auto index docs in sync.
Commits on Sep 12, 2012
Commits on Sep 10, 2012
  1. Removing deprecated methods, leaving graph mathching and kernel alone…

    jakewins committed Sep 10, 2012
    … for now (kernel because I want to wait until after HA merge, graph matching b/c cypher still depends on it).
  2. Merged Cypher refactoring into master

    systay committed Sep 10, 2012
    Conflicts:
    	cypher/CHANGES.txt
    	cypher/src/main/scala/org/neo4j/cypher/internal/commands/ComparablePredicate.scala
    	cypher/src/main/scala/org/neo4j/cypher/internal/commands/Expression.scala
    	cypher/src/main/scala/org/neo4j/cypher/internal/executionplan/builders/FilterBuilder.scala
    	cypher/src/main/scala/org/neo4j/cypher/internal/parser/v1_8/Expressions.scala
    	cypher/src/test/scala/org/neo4j/cypher/ErrorMessagesTest.scala
    	cypher/src/test/scala/org/neo4j/cypher/internal/executionplan/builders/SortBuilderTest.scala
    	cypher/src/test/scala/org/neo4j/cypher/internal/pipes/aggregation/CollectFunctionTest.scala
    	server/src/functionaltest/java/org/neo4j/server/rest/BatchOperationFunctionalTest.java
    	server/src/main/java/org/neo4j/server/rest/repr/OutputFormat.java
    	server/src/main/java/org/neo4j/server/rest/web/RestfulGraphDatabase.java
  3. Ignoring test for tonight

    jakewins committed Sep 10, 2012
  4. Modified TestStartupTimeout to not trigger interrupt while database i…

    jakewins committed Sep 10, 2012
    …s booting, because it seems that might put windows file channels in a totally borked state.
Commits on Sep 9, 2012
  1. Imports

    tinwelint committed Sep 9, 2012
Commits on Sep 8, 2012
Commits on Sep 7, 2012
  1. efactored the server startup healtchecks, combining it with the PreS…

    jakewins committed Sep 7, 2012
    …tartupStoreUpgrader code. Checks and tasks performed before proper server
    
    start are now uniformly called PreFlightTasks. These can take arbitrary time to perform (eg. upgrades, recovery).
    
    After the preflight checks, the main server startup uses a timer thread to keep track of startup time. If startup takes longer than the configured timeout,
    the startup process is interrupted.
    
    This is all done within the main server process, which means we can remove the code in startup scripts that start a bootstrap JVM separately, as well
    as the code that keeps track of startup timeouts.