Teiid

Clone Tools
  • last updated a few minutes ago
Constraints
Constraints: committers
 
Constraints: files
Constraints: dates
TEIID-1720: Admin operations except for the data source stuff were added

    • -16
    • +15
    /admin/src/main/java/org/teiid/adminapi/Admin.java
  1. … 20 more files in changeset.
TEIID-1750 fixed update leaking with transactional temp tables. switching the buffermanager caching logic to focus solely on memory management. minimized the memory footprint of the buffermanager and managed batches. switching the disk cache logic to a block approach - which will also allow for the introduction of off-heap caching. adding a batched mode for ordered temp table inserts.

    • -0
    • +39
    /engine/src/main/java/org/teiid/common/buffer/Cache.java
    • -0
    • +101
    /engine/src/main/java/org/teiid/common/buffer/CacheEntry.java
  1. … 38 more files in changeset.
Updated pom.xml to reflect release version 7.1.1.CP3-r3467

    • -1
    • +1
    /connectors/connector-salesforce/pom.xml
  1. … 24 more files in changeset.
TEIID-1706 fix to join planning with nested tables

One-off patch branch for SOA-3362 based on 7.1.1.CP3 branch

TEIID-1720: Changes for dynamic vdb deployment

    • -0
    • +31
    /build/kits/jboss-as7/docs/teiid/datasources/file.xml
TEIID-1750 TEIID-1753 converting compaction logic to in place and adding tail tracking

TEIID-1755 preventing an npe if not in a clustering profile

TEIID-1720: adding a sample connection factory

TEIID-1753 addressing compaction issues

TEIID-1657: making view name available to the mattable request such that it can be proactively loaded at other nodes in the cluster

TEIID-1657: Implemented distributed refreshMatView call based on the JBoss Cache Listener mechanism. When user issues a refreshMatView call on one node, then all the nodes gets notified about the change in the original node after load of the mat view table is finished. The proactive load on the passive nodes are done in a separate thread. Also fixed a memory leak with the TupleBufferCacheLoader in loading batches.

TEIID-1493: When a node rejoins the cluster after the initial cache has been populated, during the join time the state has been set to not transfer. This need be set to transfer, also JBoss cache only transfers the state on "active" regions.

TEIID-1493 Rolling back the previous change.

TEIID-1493: If the same VDB vesion is added and removed recently, the code was not flushing the cache contents; however in the case of startup, there will be no-entries in the recently removed map, the cache entries will be dumped for all the VDBs, thus invalidate the distributed cache entries. Here TEIID-1256 being reverted as this was wrong assumption from before of the behavior required. Removing the clearing of the cache entries entirely. Actually I am little confused about the intent of the map to keep track of the recently removed vdbs, as this doing opposite to what is documented in terms of the active connections.

TEIID-1652, TEIID-1653 : Handling the error conditions and doing rollback in case of an error

TEIID-1653: correcting the regex to account for sql with new line characters

TEIID-1493 changing the listener logic to prevent unnecessary cache clearing

TEIID-1656 adding support for pushdown of aggs over unions with grouping expressions

TEIID-1657 changes to ensure that matview refresh is handled properly with distributed caching enabled.

TEIID-1654 ensuring that missing cache entries are not restored

SOA-3103: removing the sandbox translator

changed pom.xml files for CP3 branch naming

changed pom.xml files for CP3 branch naming

    • -1
    • +1
    /connectors/connector-salesforce/pom.xml
  1. … 22 more files in changeset.
SOA-3103 supportsOuterJoins=false was removed, teradata does support this.

SOA-3103: removing the sandbox translator