I was inspired by the work done during Oceanhackweek2020 (OHW20) to investigate the github API. This started as an idea to count the github commits made by participants during OHW20 and the weeks on either side.

I could not find a good tutorial on scraping commits from a list of Github usernames (though this out of date one by kb22 set me on the right track) so I've written a fairly rough one here.

Check it out on binder from your browser .

You won't be able to run much of the scraping notebook on binder though, as it requires a GitHub API token

All the instructions for carrying out this scraping are found in the notebook ohw_github_scrape.ipynb.

If you don't fancy the scraping yourself, but are interested in the results, you will find a csv of anonymised data in this repo along with some very basic analysis in ohw_analysis.ipynb

If you found this work interesting, please feel free to fork it and consider submitting a PR. Or contact me on twitter

• Scrape github for OHW20 participant public information
• Total the number of commits during hack week and the weeks either side
• Investigate deeper including file types used and wording of commit messages
• Plot and analyse this info in a blog post
• Provide a how-to, with anonymised data
• Write a tutorial for anyone else curious about scraping Github data