Actually this is a more complex issue in Spark -> https://issues.apache.org/jira/browse/SPARK-24287
I'm trying to see of we are able to resolve this for this particular case - RSparkling, however usage of maven artifact ai.h2o:sparkling-water-package_2.11 will remain broken until this is fixed
Found the core issue. The issue is that instead of jar, javadoc is downloaded.
It is caused transitively by sparkling-water-core_2.11 -> h2o-algos -> com.github.fommil.netlib-core where you actually see http://central.maven.org/maven2/com/github/fommil/netlib/core/1.1.2/core-1.1.2.pom the dependency on javadoc enforced via the classifier.
Trying to see if I can exclude this on the sparkling_water_package package level
This will be fixed on H2O side by
Will keep this jira open just as a wrapper on Sparkling Water side even though no code change is required in Sparkling Water