/covid-19-jhu-json-converter

Converts the CSV files published by Johns Hopkins regarding Covid into a cleaner JSON format

Primary LanguageRubyMIT LicenseMIT

covid-19-jhu-json-converter

Converts the CSV files published by Johns Hopkins regarding Covid-19 into a cleaner JSON format.

Background

Many people around the world are currently working a lot with statistical models regarding the COVID-19 spread. One of the most reliable and most quoted sources for infection numbers is the Johns Hopkins University in Baltimore. They update their numbers several times a day and publish them in a CSV format.

What the converter does

Since for many computational tasks JSON is a more appropriate format than CSV, we wrote a converter that takes the Hopkins data from the above repo and transforms them into a single JSON array.

Each element in the array represents a geographical area from the Hopkins data (equal to one CSV row). It also combines all three figures: infections, deaths and recoveries into a single object for each area.

How to run

The converter is a simple ruby script that we packed into a Docker container.

You can run it directly from Docker like so: docker run docker.pkg.github.com/cweyer/covid-19-jhu-json-converter/covid19-csse-exporter:latest

You can also fork this repo and run the ruby script directly. The JSON is written to STDOUT. ruby jhu2json.rb.

Output methods

This script supports output to STDOUT, to a file and directly into a database via ActiveRecord, so all ActiveRecord databases are supported. You can select your desired output with the --output command line option. When using --output file you may use --filename to specifiy a file name, otherwise the file is named with the current date. When using --database you must also set --credentials containing Rails style database credentials (i.e. sqlite3:///covid.db or postgresql://user:password@host:port/database?pool=10).