Clone
 

vkorukanti <venki.korukanti@gmail.com> in drill

DRILL-4483: Fix text plan regression in query profiles

DRILL-4446: Support mandatory work assignment to endpoint requirements of operators

  1. … 9 more files in changeset.
DRILL-4434: Deprecate GroupScan.enforceWidth API

DRILL-4353: Add HttpSessionListener to release resources of expired/invalidated sessions

DRILL-4353: Add HttpSessionListener to release resources of expired/invalidated sessions

Closes #359

DRILL-4354: Remove sessions in anonymous (auth disabled) WebUI access

DRILL-4328: Fix backward compatibility regression caused by DRILL-4198

DRILL-4256: Create HiveConf per HiveStoragePlugin and reuse it wherever needed.

Creating new instances of HiveConf() are very costly, we should avoid creating new ones as much as possible.

Also get rid of hiveConfigOverride and use HiveConf in HiveStoregPlugin wherever we need the HiveConf.

DRILL-4198: Enhance StoragePlugin interface to expose logical space rules for planning purpose

Also move Hive partition pruning rules to logical storage plugin rulesets.

this closes #300

DRILL-4194: Improve performance of the HiveScan metadata fetch operation

+ Use the stats (numRows) stored in Hive metastore whenever available to

calculate the costs for planning purpose

+ Delay the costly operation of loading of InputSplits until needed. When

InputSplits are loaded, cache them at query level to speedup subsequent

access.

this closes #301

Roll forward to 1.5.0-SNAPSHOT

    • -1
    • +1
    /contrib/data/tpch-sample-data/pom.xml
  1. … 12 more files in changeset.
[maven-release-plugin] prepare release drill-1.4.0

    • -3
    • +2
    /contrib/data/tpch-sample-data/pom.xml
  1. … 12 more files in changeset.
Add Venki's key to KEYS

DRILL-3980: Upgrade hadoop dependency version to 2.7.x in mapr profile

DRILL-3739: Fix issues in reading Hive tables with StorageHandler configuration (eg. Hive-HBase tables)

DRILL-3893: Change Hive metadata cache invalidation policy to "1 min after last write".

DRILL-3938: Support reading from Hive tables that have schema altered after the creation

Also:

+ Remove "redoRecord" logic which is not needed after "automatic reallocation" (DRILL-1960) changes.

+ Remove HiveTestRecordReader. This is incomplete in implementation and not used anywhere. It is currently just

a burden to maintain with changes in its superclass HiveRecordReader

DRILL-3884: Fix lower parallelization issues with Hive's native scan.

This closes #185

DRILL-3209: Add support for reading Hive parquet tables using Drill native parquet reader

DRILL-3201: Support security for access through Web UI

  1. … 9 more files in changeset.
DRILL-3746: Get Hive partition values from MetaStore instead of from parsing the partition location path

1) Added a partition with custom location to test Hive table. Existing partition tests now work after the fix.

2) Enabled a test which was disabled previously due to a bug in interpreter code which is fixed recently.

This closes #151

DRILL-3622: When user authentication is enabled, enforce admin privileges to update SYSTEM options

+ define what user is considered an admin

+ remove a stray file in test module (exec/java-exec/src/test/java/org/apache/drill/exec/server/rest/RootResource.java)

DRILL-3467: Restrict visibily of workspaces in 'show schemas' based on access permissions of querying user.

Also:

Fixed failing tests in TestJdbcMetadata as the result depends on whether /tmp/drill_test is available on

test system or not. Change it to use the default connection test parameters (which creates a unique tmp directory)

so that the test result doesn't depend on individual machine.

DRILL-3725: Add HTTPS support for Drill web interface

DRILL-4169: Upgrade Hive storage plugin to work with Hive 1.2.1

+ HadoopShims.setTokenStr is moved to Utils.setTokenStr. There is no change

in functionality.

+ Disable binary partitions columns in Hive test suites. Binary

partition column feature is regressed in Hive 1.2.1 (HIVE-12680). This

should affect only the Hive execution which is used to generate the test

data. If Drill is talking to Hive v1.0.0 (which has binary partition

columns working), Drill should be able to get the data from Hive

without any issues (tested).

+ Move to tinyint_part from boolean_part as there is an issue with boolean

type partition columns too (HIVE-6590).

+ Update StorageHandler based test as there is an issue with test data

generation in Hive 1.2.1. Need a separate test with custom test StorageHandler.

this closes #302

DRILL-3579: Fix issues in reading Hive tables with partition value __HIVE_DEFAULT_PARTITION__

Also:

1) Currently the code that interprets partition values in string format to appropriate type

is duplicated in HiveRecordReader and HivePartitionDescriptor. Refactor the code into

common place HiveUtilities.

2) Add tests to test deserialization of partitions of all supported types.

DRILL-3413: When SASL is enabled use DIGEST mechanism in creating HiveMetaStoreClient for proxy users.

DRILL-3398: Fix memory leak in WebServer

DRILL-3328: Add unittest for convert_from function on Hive BINARY type

DRILL-3203: Add support for impersonation in Hive storage plugin

    • -0
    • +10
    /contrib/storage-hive/core/src/test/resources/student.txt
    • -0
    • +10
    /contrib/storage-hive/core/src/test/resources/voter.txt
  1. … 7 more files in changeset.