A GitHub action monitors changes in this examplary document and provides different diagrams automatically. They depict the code count and number of commits for various time intervals.
The corresponding python file contains methods for extracting git log parameter. It maps the information on a python pandas table and generates standard diagrams Day.png, Week.png" etc. Additionally, you can add more specific evaluations too. For this purpose the script provides a csv file containing all gathered data.
Examples
- Lines of code / commits depicted for all day beginning with the first commit
- Lines of code / commits depicted for each week beginning with the first commit
The corresponding action description is available in .github\workflow. It is activated by changes in files with extension .md! The filtering avoids an endless loop by generating new content based on the action.
name: Generate activity diagrams
on:
push:
paths:
- '**.md'
jobs:
build:
runs-on: ubuntu-latest
steps:
- name: Check out repository
uses: actions/checkout@v2
with:
fetch-depth: 0
- name: Set up Python 3.6
uses: actions/setup-python@v1
with:
python-version: '3.6'
- name: Display Python version
run: python -c "import sys; print(sys.version)"
- name: Install dependencies
run: |
python -m pip install --upgrade pip
pip install matplotlib
pip install pandas
- name: Move python script to repository root
run: |
cp statistics/generateStatistic.py .
- name: Run python script
run: |
python generateStatistic.py
rm generateStatistic.py
- name: Commit files
run: |
git config --local user.email "[email protected]"
git config --local user.name "GitHub Action"
mv *.png statistics/
git add ./statistics/*.png
git status
git commit -m "Add new statistics" -a
- name: Push changes
uses: ad-m/[email protected]
with:
github_token: ${{ secrets.GITHUB_TOKEN }}