Dataset reproduction package

One of the main goals for NPS is to reproduce samples outside of CMS (to be used when validating analyses).

To meet this objective, we should produce self-contained packages with the necessary information. This includes:

• All input cards for the ME generator, if relevant
• All Pythia settings (#3)
• Information about filters (#14 (closed)), if relevant

We should come up with a simple structured way of producing these packages. Maybe sorting by generator:

dataset_name/madgraph/card1.txt
dataset_name/madgraph/card2.txt
dataset_name/madgraph_version
dataset_name/pythia.cmnd
dataset_name/pythia_version
dataset_name/filters.html
dataset_name/index.html

(I used html for filters because I don't know better.)

The top-level folder with the dataset name is needed in order to combine multiple datasets in a single tarball that can be uploaded to HepData as supplementary material.

For starters we should focus on Pythia standalone and MadGraph + Pythia, without any additional filters.