I have a AWS Redshift wrapper class that automates similar types of loads from S3 for me, and I have recently adapted it to work for Spark jobs, which don't require a manifest, and instead need a slightly different COPY statement. Other than this one method, all other code is transferable and reu...