1.6.3

Artifacts

This release is available in the Maven Central Repository for both Scala 2.10 and Scala 2.11. Please choose the version that matches the version of Scala that your Spark installation is using.

This release is also available via Spark Packages.

Change Log

• Data locality support for Coverage Plan Based Partitioning #230
• Improve Coverage Plan Based Partitioning: smart split calculation and more accurate coverage entry distribution across the partitions #231
• Critical, Fix Python serialization for empty JSON objects #226
• Fix double filtering for DataFrames #228

1.6.2

Artifacts

This release is available in the Maven Central Repository for both Scala 2.10 and Scala 2.11. Please choose the version that matches the version of Scala that your Spark installation is using.

This release is also available via Spark Packages.

Change Log

Critical fix Python KV: if object values are JSON objects with list fields (empty or not) then exception happens #219.

1.6.1

Artifacts

This release is available in the Maven Central Repository for both Scala 2.10 and Scala 2.11. Please choose the version that matches the version of Scala that your Spark installation is using.

This release is also available via Spark Packages.

Change Log

• Make Namespace not to be mandatory for KV operations (#212)
• Make the Python tests be executed on TravisCI for both: TS and KV (#213)
• Critical, Fix improper Spark partition order (#211)
• Critical, Fix Python serialization for JSON values with more than 4 keys (#210)
• Critical, fix empty writes (#205)
• Fix value mapping for JSON array values on KV reads (#215)
• Fix Python .partitionBy2iKeys() error (#216)

1.6.0

Artifacts

This release is available in the Maven Central Repository for both Scala 2.10 and Scala 2.11. Please choose the version that matches the version of Scala that your Spark installation is using.

This release is also available via Spark Packages.

Change Log

• Support Python and KV buckets
• Support Spark Streaming context in the connector
• When a Riak node is unavailable during a Spark job, failover to another node
• Remove separate java-connector and incorporate java classes into main connector project
• Build with SBT instead of Maven
• Build artifacts for Scala 2.10 and 2.11
• Use Docker for builds and tests
• Update docs and examples
• Miscellaneous enhancements and bug fixes
  • Improve TS Partitioning #96
  • Query2iRange parallelization for KV #146
  • Allow passing quanta manually #153
  • Connector stuck on application exit #164
  • Fix TSDataQueryingIterator issue with breaking iteration on the first sub-query returned no results #178

1.5.1

This release is only for publishing in maven central

1.5.0

Artifacts

The latest version of the Spark-Riak Connector can be found in Bintray:

• Scala (uber jar) - https://dl.bintray.com/basho/data-platform/com/basho/riak/spark-riak-connector/1.5.0/spark-riak-connector-1.5.0-uber.jar
• Scala - https://dl.bintray.com/basho/data-platform/com/basho/riak/spark-riak-connector/1.5.0/spark-riak-connector-1.5.0.jar
• Java - https://dl.bintray.com/basho/data-platform/com/basho/riak/spark-riak-connector-java/1.5.0/spark-riak-connector-java-1.5.0.jar

Change log

• Reorganize and expand documentation (README)
• Add bulk size parameter for writes to Riak TS table (spark.riakts.write.bulk-size) #129
• Rename riakTSBucket() to riakTSTable() #128
• Update Spark version to 1.6.1 #109
• Add support for Spark partitioning with Riak TS tables #96
• Miscellaneous bug fixes

1.5.0-rc1

v1.5.0-rc1

bump version to 1.5.0-rc1

v1.3.0-beta1

Merge pull request #92 from basho/develop

Release 1.3.0-beta1

Release v1.1.0

Merge pull request #68 from basho/pex-without-streaming-values-for-oss

Disable FBR Streaming values for OSS

PreRelease v1.1.0-RC1

Merge pull request #68 from basho/pex-without-streaming-values-for-oss

Disable FBR Streaming values for OSS