humphd / github-contrib-stats

Download and count GitHub Issue, Repo, and PR stats from web pages with GitHub URLs included

GitHub Contrib Stats

This is a set of scripts I use to help me extract GitHub contribution stats for my open source classes. It takes a set of URLs to web pages with links to GitHub Issues and/or Pull Requests, then downloads and extracts data about them.

Setup

First, install all deps:

$ npm install

Next, copy the env.sample file to .env and alter it to add your information. See the Octokit docs for details on how to get a GitHub token.

You need to include a list of URLs to the web pages that contain the URLs you want to process. These can be HTML, markdown, or any other text format. The page(s) will be downloaded and the URLs extracted.

# List of URLs to process, separated by spaces
URL_LIST="https://url1.com/file.html https://url2.com/file.md"

How to Run

With everything installed and the .env setup, you can now run the analysis:

npm start

This will do a number of things:

1. Use the URLs in URL_LIST to download all URLs specified
2. Extract all GitHub URLs from the text at each page
3. Build a list of repos, issues, and pull requests in data/github-data.json and make requests for of these items from the GitHub API. All results will be cached in data/, so you can interrupt it, or retry later.
4. Generate stats for all Issues in data/issue-data.json
5. Generate stats for all Pull Requests in data/pr-data.json