Autograding GitHub Action

This GitHub action autogrades projects based on a configurable set of metrics and gives feedback on the pull request. Currently, you can select from the following metrics:

Test statistics (e.g., number of failed tests)

Code coverage (e.g., line coverage percentage)

PIT mutation coverage (e.g., missed mutations' percentage)

Static analysis (e.g., number of warnings)

How to use?

Here is an example configuration you could use in your workflow.

name: 'Autograding Pull Pequest'

on:
  pull_request:
    branches: [ master ]

jobs:
  build:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v2
      - name: Build
        run: mvn -V -ntp clean verify -Dmaven.test.failure.ignore=true --file pom.xml
      - name: Autograding
        uses: uhafner/autograding-github-action@v0.1.0
        with:
          TOKEN: ${{ secrets.GITHUB_TOKEN }}
          HEAD_SHA: ${{github.event.pull_request.head.sha}}

Configuration

• TOKEN: ${{ secrets.GITHUB_TOKEN }}: mandatory GitHub access token.
• CHECKS_NAME: "Name of checks": optional name of GitHub checks (overwrites default: "Autograding result").
• HEAD_SHA: ${{github.event.pull_request.head.sha}}: optional SHA of the pull request head. If not set then GITHUB_SHA will be used.
• CONFIG: "{\"analysis\": { \"maxScore\": 100, \"errorImpact\": -5}}": optional configuration, see manual for details. If not provided a default configuration will be used.