This release is about three things:
- adding support for more recent major Spark versions, in particular Spark 2.3 and Spark 2.4; since the latter is the first Spark version available for Scala 2.12, this release of kontextfrei is also the first one cross-compiled against Scala 2.12, at least for that specific Spark version
- adding a few
RDD
operators that were still missing:repartitionAndSortWithinPartitions
,partitionBy
,filterByRange
andsortByKey
; there is also anempty
method now inDistributionOps
, which invokesemptyRDD
in theRDD
implementation - finally, and most importantly, improving the developer experience by supporting the
setName
method and by setting the proper call site; both of these lead to a better experience in the Spark UI when using kontextfrei
Issues
The following issues have been closed as part of this release:
#55, #56, #57, #59, #61, #62, #64, #66
Contributors
This is the first release including external contributions. Thanks a lot to Simon J. Scott for his invaluable contributions regarding the Spark UI, the creation of empty RDD
s and the partitionBy
operator.