Mason Spark

This repository contains various implementations of mason execution engine jobs which use a spark configuration.

Requirements:

docker, sbt (scala)

Run

./build

Which builds the jar file and packages it into the mason-spark docker image. mason-spark is derived from the google spark operator image and is based on

scala 2.11.12 and spark 2.4.5

by default. More versions will be supported later but for now ensure that you are using these.

Current Job Implementations

1. MergeJob: Merges all the files in a single specified path to a single schema
   • Currently Supports:
     • Parquet
     • CSV
     • JSON
     • JSON lines
   • Does not yet support mixed formats