This repository contains some example uses of GitHub Actions for automating scientific data workflows. It is inspired heavily by a workshop given at US-RSE 2024 by @valentina-s, @gbrencher, and @scottyhq. Their book and repository contain additional examples that are more python focused and maybe more advanced, while the examples in this repository are more R focused and will be used in a shorter format workshop at the University of Arizona.
This is a template repository, so you can make a copy of it by clicking the green “use this template” button.
Once you’ve successfully created a repository in your own GitHub
username or organization, you can navigate to the “Actions” tab where
you’ll find workflows numbered roughly in order of complexity. The .yaml
files defining the workflows have them all set to be triggered on
workflow_dispatch so you can run them by selecting a workflow and
clicking the “Run workflow” button.
The .yaml files defining the workflows are in .github/workflows. R
code run by some of these workflows is in the R/ folder. Reports
rendered by some of the actions include the source for this document,
README.qmd , and docs/report.qmd.
The example data in data/heliconia_sample.csv is modified from Bruna
et al. (2023). Data validation was run on this dataset using GitHub
Actions in this repo: https://github.com/BrunaLab/HeliconiaSurveys.
Bruna, Emilio M., María Uriarte, Maria Rosa Darrigo, Paulo Rubim, Cristiane F. Jurinitz, Eric R. Scott, Osmaildo Ferreira da Silva, and John W. Kress. 2023. “Demography of the Understory Herb Heliconia Acuminata (Heliconiaceae) in an Experimentally Fragmented Tropical Landscape.” Ecology 104 (12). https://doi.org/10.1002/ecy.4174.
One example workflow involves re-rendering this README to include some
updated histograms of the example dataset in data/heliconia_sample.csv
