Gw____tt COVID-19 Dashboard

This repo scrapes data from PDF reports (now also located here) put together by the Gw____tt School District, packages it with a shiny web app built by a collaborator, then deploys it to https://shinyapps.io. We also automatically upload up-to-date consolidated datasets to the project's Open Science Framework page --- if you like, hop over there to grab a copy of the raw data.

Check out live the Covid Dashboard app here.

Play-By-Play of the Build Process

We use Travis CI to build. Travis automatically launches jobs in response to fresh commits to certain branches on this repo. You can peek at this repo's current build jobs here.

1. A pdf gets added to the input/ folder on the raw-data branch.
2. Travis launches a build on the raw-data branch to extract tables from all pdfs in input/, generating one csv file per dataset in the pdf. (All content from datasets split over multiple pages end up in the same csv.)
3. A copy of the data-joiner branch is cloned down and the csv files are loaded into its input/ folder.
4. The loaded-up data-joiner branchis force-pushed to the parsed-data branch.
5. Travis launches a job on the parsed-data branch to join csv files corresponding to different report dates.
6. A copy of the app-builder branch is cloned down and the joined csv files are loaded up into its root directory.
7. The loaded-up parsed-data branch branch is force-pushed to the built-app.
8. Travis launches a job on the parsed-data branch to build our shiny app and deploy it.

Acknowledgement

The idea of scraping data out of pdfs for this project was inspired by @jaredmoore's covid_tracker project.