Version 1.6.3 of the Riak Spark Connector has been released, and is now available through both GitHub and Spark Packages. I'd like to thank the team, especially Sergey Galkin, for all their hard work on this release.
This release fixes some issues, and enhances some features:
Enhancement - Data locality support for Coverage Plan Based Partitioning #230
Enhancement - 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