[dtjon]

Wow this is great. I've been working around the jvm to integrate sklearn and some spark jobs that produce Parquet. This is a huge relief