covid19_analysis

Michael Pereira [email protected]

This project collects, compiles, analyzes and viusualizes COVID-19 data -- primarily for the provinc of Manitoba.

The project can be run locally in RStudio or a server-based instance of RStudio can be deployed using Docker.

Docker install

Change user name and RStudio login credentials in the Dockerfile and run_docker.sh files before proceeding

Build the image

docker build --rm --force-rm -t covid19_analysis .

Run the image

sh ./run_docker.sh

This sets data directory and defines credentials for the RStudio user defined in the Dockerfile. Adjust the docker run options to suit.

Login credentials for the default user (wfpnews) is set as a docker run option in ./run_docker.sh. Change them before deploying and starting the container.

Uploading graphics to production

Third-party credentials (AWS S3/Google API keys) are not included in build or run commands.

** Credentials must be added to a .env file, from within the RStudio UI terminal, after every reboot including rebuild-reboot.

Manually add a .env after the image is running.

AWS credentials are required for uploads to the S3 bucket used for production graphics.

GOOGLE_API_KEY=qwertyuhgfvcghjuikjhnklo-kmkl
AWS_ACCESS_KEY_ID=qwertyuhgfvcghjuik-jhnklokmkl
AWS_SECRET_ACCESS_KEY=qwertyuhgfvcghjuikjhnklokmkl-qwertyuhgfvcghjuikjhnklokmkl

---