Clone Tools
  • last updated a few minutes ago
Constraints: committers
Constraints: files
Constraints: dates
MODE-1568 - Fixed the fact that surefire ignored the ipv4 or ipv6 profiles (and any other system properties from the parent.pom)

MODE-1558 - Updated the release of thread pools, to call eventually shutdownNow

MODE-1448: Handle duplicate inserts

MODE-1558 -The workers of the RepositoryChangeBus should be stopped when the bus is shut down.

MODE-1555 - Updated the TCK tests to use the system property known.issues in order to ignore failing tests. In addition, updated all the TCK related test suites, so that the ModeShape engine is shut down at the end of each suite run.

    • -0
    • +103
MODE-1562 changed getResourceLength impl

removed null check

MODE-1562 changed getResourceLength impl

removed null check

MODE-1546 - Cleaned up and fixed some issues in the attempt to fix the intermittent failures: - fixed the dispatching of workspace events if a transaction fails - JcrObservationManager: cleaned up some multi-threaded code and improved performance in some cases by not retrieving nodes from the schematic db - TextExtractors - added a way to shut them down and shutdown the executor - update the creation of Executors, so that they use NamedThreadFactory

MODE-1448: Unit tests

'Release: update versions for modeshape-3.0.0.Beta1'

  1. … 34 more files in changeset.
Updated release notes for 3.0.0.Beta1

MODE-1448: Add strategy helper

MODE-1448: Count columns from 1

MODE-1448: Switch to plain JDBC

MODE-1419 Enabled full text searching

Added full-text search back in. Note that it can be explicitly disabled/enabled within a repository configuration,

and it is currently *enabled* by default. Quite a few test cases were added back in, and some of them highlighted

some issues with search/query scores (were floats, but expected to be doubles).

Also, one quirk of the AS7 subsystem startup is that the text extractors are added to a running repository,

and that means that there's a time during startup when there are no text extractors (and any added binary

values don't get extracted during storage), followed quickly by the enabling of text extractors (by the time

the property with the binary value is indexed). This manifested itself as a blocked sequencing thread within

the ZIP Sequencer integration test (using AS7).

All unit and integration tests pass with these changes, including those that were enabled by these changes.

  1. … 10 more files in changeset.
MODE-1547 - Fixed full text search queries which involve stop-words. The solution was to not add empty PhraseQuery instances to the BooleanQuery.

MODE-1550 Corrected the generated Lucene criteria for ISCHILDNODE

ModeShape converts the ISCHILDNODE criteria for JCR-SQL2 queries into Lucene

queries, and this was being done incorrectly for the root node: an

extra '/' was always being appended to the path to ensure the parent node

was not included in the results, but doing this on the root path resulted

in a path of "//", which was not correct.

After this was fixed, the test case added (and ignored) for MODE-1469 was

re-enabled and modified to run correctly. All other tests also pass.

MODE-1558 - Updated the handling & releasing of thread pools. Subsystems (e.g. query, sequencing etc) should close their executors when shutdown is called on them.

MODE-1469 - Updated AS7 subsystem with the "systemContentMode" setting which controls the behavior of system content indexing.

  1. … 3 more files in changeset.
MODE-1448: -Split metadata and payload -Split payload into chunks -Added configurable idle time

MODE-1469 - Fixed indexing of system nodes. Only /jcr:system, /jcr:system/jcr:nodeTypes and /jcr:system/mode:namespaces will be indexed. Updated the JSON configuration to control whether system indexing is: sync, async or disabled, with the latter being the default. Updated the query tests to reflect those changes.

Working on this, another issue surfaced: we weren't closing properly the executor services that we're creating during the repository startup. Therefore, some changes were added in this area as well.

  1. … 11 more files in changeset.
MODE-1546 Handled periodic failure

Attempted to fix a failure that shows up periodically, most of the time when the indexing

is being performed, but sometimes in other cases. The error is due to the inability of

a CachedNode to compute its path segment, because the parent no longer contains a child reference

to the node. This change attempts to correct that problem by identifying the case and throwing

a NodeNotFoundInParent exception. The indexer can simply ignore such nodes.

Added a comment to pom that references an online thread of where I got the pom code to solve the build error.

Added comment to pom that references the online thread where I got the pom code to solve the build issue.

Fixed javadoc errors that occurred when moving from Eclipse Helios to Eclipse Indigo. Also fixed a pom error that occurred when upgrading Eclipse.

  1. … 6 more files in changeset.
MODE-1556 Corrected the handling of != (or <>) query criteria on numeric fields

Numeric values were not being handled consistently in the indexes and in queries. For example,

some values were being treated as integer values in the indexes and queried as longs (which

doesn't work). Additionally, some of the logic of handling and combining "not" queries was


All unit and integration tests pass with these changes.

MODE-1549 JCR-SQL2 uses '<>' rather than '!=' as not-equal-to operator

ModeShape abstract syntax tree used '!=' as the "not-equal-to" operator, but did

alias '<>' to '!='. This change reverses this so that the '<>' token is considered

the primary operator, and '!=' is aliased.

MODE-1469 - Initial work on fixing the indexing of the system content

MODE-1448: Initial implementation

MODE-1448: Configure JDBC binary store