GitHub pages generator

Generate and deploy minimalistic website with your repositories statistics with GitHub Actions.

Key features:

• lightweight, html and css only
• responsive, mobile friendly
• customizable and extensible

Example workflow

• To run and deploy at 12:00pm every day:

  name: Build and deploy website
  
  on:
    push:
      branches:
        - master
    schedule:
      - cron: "0 12 * * *"
  
  jobs:
    build:
      runs-on: ubuntu-latest
      steps:
      - name: Build
        uses: adamws/github-pages-generator@master
        with:
          github_token: ${{ secrets.GITHUB_TOKEN }}
          username: ${{ github.actor }}
          output_dir: ./output
      - name: Deploy
        uses: peaceiris/actions-gh-pages@v3
        with:
          github_token: ${{ secrets.GITHUB_TOKEN }}
          publish_branch: gh-pages
          publish_dir: ./output

• To run in data-only mode for creating repositories.json file at output_dir path, which can be used for integration with other tools:

  jobs:
    build:
      runs-on: ubuntu-latest
      steps:
      - name: Get data
        uses: adamws/github-pages-generator@master
        with:
          github_token: ${{ secrets.GITHUB_TOKEN }}
          username: ${{ github.actor }}
          data_only: true

• Other examples:

  • action's projects (default theme, trigger on master push)
  • my projects (customized theme, trigger on schedule)

  

Local usage

Clone this repository, install dependencies, build and serve:

git clone https://github.com/adamws/github-pages-generator.git && cd github-pages-generator
python -m venv .env
source .env/bin/activate
pip install -r requirements.txt
python github-pages-generator/build.py --username <<github username>>
cd output
python -m http.server

After that, open http://0.0.0.0:8000/ in browser. Files from output directory can be manually committed to gh-pages branch or your GitHub Pages repository, but it is recommended to set up automatic deployment with GitHub Actions which will trigger periodically, keeping your statistics up to date.

Rate limits

GitHub's REST API has rate limits which can be easily reached while experimenting with this tool. There are two mechanisms built-in to circumvent that:

• requests are cached with requests-cache
• requests can be optionally authenticated with personal access token via GITHUB_USER and PERSONAL_ACCESS_TOKEN environment variables.