This version solves a three bugs:

• Fixed replacing chunks
• Fixed lots of error printing when a wrong option was parsed, this started when the code was converted to python3
• Fixed not properly checking if a dataset has bad data files

Thanks for the pull requests and the nice suggestions!