I think it is indeed getting unwieldy. I proposed the following change to trying to make each script self-contained while maintain certain sense of modularity:
.
├── make_step2_ntuples.py # Put common functions in 'utils.py`, then rename it with `rdx.py`
├── rdx # Remove this folder entirely
│ ├── data_no_mu_bdt.sh
│ ├── data.sh
│ ├── mc.sh
│ ├── rdx-run1.yml
│ ├── rdx-run2.yml
│ ├── trigger_emulation_fs_vs_to.sh
│ └── trigger_emulation.sh
├── rdx_cutflows.py
├── rdx.py # Largely replace this with 'make_step2_ntuples.py'
└── utils.py
After said change, rdx.py should have no unknown dependency other than utils.py, which should be relatively clear.
I don't like the idea to have multiple ways to do the same thing, so let's see if we can agree on a sensible middle point. Still, this will take some work, so before I start actually working on it, I'd like to hear opinions on this from @manuelfs
Currently, our workflow folder looks like this:
I think it is indeed getting unwieldy. I proposed the following change to trying to make each script self-contained while maintain certain sense of modularity:
After said change,
rdx.pyshould have no unknown dependency other thanutils.py, which should be relatively clear.I don't like the idea to have multiple ways to do the same thing, so let's see if we can agree on a sensible middle point. Still, this will take some work, so before I start actually working on it, I'd like to hear opinions on this from @manuelfs