This repository is a collaboration between RMI and Catalyst Cooperative to connect FERC Form 1 plant records, EIA plant records and depreciation study records at the most granular level.

Installation

To install the software in this repository, clone it to your computer using git. If you're authenticating using SSH:

git clone [email protected]:catalyst-cooperative/rmi-ferc1-eia.git

Or if you're authenticating via HTTPS:

git clone https://github.com/catalyst-cooperative/rmi-ferc1-eia.git

Then in the top level directory of the repository, create a conda environment based on the environment.yml file that is stored in the repo:

conda env create --file environment.yml

Note that the software in this repository depends on the dev branch of the main PUDL repository, and the setup.py in this repository indicates that it should be installed directly from GitHub. This can be a bit slow, as pip (which in this case is running inside of a conda environment) clones the entire history of the repository containing the package being installed. How long it takes will depend on the speed of your network connection. It might take ~5 minutes.

The environment.yml file also specifies that the Python package defined within this repository should be installed such that it is editable. This will allow you to change the modules that are part of the repository and have the installed software reflect your changes.

If you want to make changes to the PUDL software as well, you can clone the PUDL repository into another directory (outside of this repository), and direct conda to install the package from there. A commented out example of how to do this is included in environment.yml. NOTE: if you want to install PUDL in editable mode from a locally cloned repo, you'll need to comment out the dependency in setup.py as it may otherwise conflict with the local installation (pip can't resolve the precedence of different git based versions).

After any changes to the environment specification, you'll need to recreate the conda environment. The most reliable way to do that is to remove the old environment and create it from scratch. If you're in the top level rmi-ferc1-eia directory and have the pudl-rmi environment activated, that process would look like this:

conda deactivate
conda env remove --name pudl-rmi
conda env create --file environment.yml
conda activate pudl-rmi

In order to use this repository, you will need a recent copy of the PUDL database. You You can either create one for yourself by running the ETL pipeline, or you can follow the instructions in the PUDL examples repository to download the pre-processed data alongside a Docker container.

To work with the pre-processed data outside of the Docker container, you will need to tell the PUDL software where to find that data on your computer. When you extract the pre-processed data archive, it will include a directory named pudl_data -- you need to put the path to that directory in a file called .pudl.yml in your home directory. The contents will need to look like the following (but with real paths...):

pudl_in: /path/to/your/downloaded/pudl_data
pudl_out: /the/same/path/to/pudl_data

NOTE: If you get to a point where you need or want to run the PUDL ETL for yourself, you will need to reset these paths to another location so that you don't accidentally overwrite the pre-processed data.

• If you're unfamiliar with file paths, directories, and the command line in general, we recommend checking out The Basics of the Unix Shell from Research Software Engineering in Python.
• If you'd like more background on reproducible software environments, including software package managers and the role played by containerization systems like Docker, check out the chapter on Reproducible Research from The Turing Way.

Tests

This repo finally has some tests! wahoo! Unfortunately, there are memory issues getting in the way of letting us run all of the tests via github actions (PUDL issue #1457).

Regenerate All Outputs & Validate

The full CI tests can be run via pytest or tox. This will take a while because it regenerates all of the outputs and then runs relatively quick tests on those outputs.

pytest test/integration/rmi_out_test.py

OR

tox

Validate Existing Outputs

If you have recently processed output cached in the output directory (pudl_rmi.OUTPUTS_DIR) and just want to test the consistency of the outputs, there is a quick test to run. This test checks whether the processing of the data has or has not introduced errors. There are known errors being stored in the input directory (pudl_rmi.INPUTS_DIR). We expect most of these error exist because of missing connections between datasets.

Only run these tests if you know your cached outputs are up to date and consistent with each other.

pytest test/integration/rmi_out_test.py::test_consistency_of_data_stages

Process Overview

Below is a visual overview of the main processes in this repo:

Each of the outputs shown above have a dedicated module:

• EIA Master Unit List: make_plant_parts_eia.py
• EIA & Depreciation Connected: connect_deprish_to_eia.py
• EIA & FERC Connected: connect_ferc1_to_eia.py
• Connected Depreciation & FERC: connect_ferc1_to_eia.py