Automated Python Grading System

A containerized grading system for Python(for now) assignments that provides automated testing and scoring. The system supports weighted tests and maintains a history of student submissions.

For Server Administrators

System Requirements

• Python 3.12+
• Docker Desktop (for macOS/Windows) or Docker Engine (for Linux)
• uv (for dependency management)

Initial Setup

1. Clone the repository:

git clone <repository-url>
cd grader

2. Create and activate a virtual environment:

python -m venv .venv
source .venv/bin/activate  # On Windows: .venv\Scripts\activate

3. Install uv and dependencies:

pip install uv
cd service
uv sync

4. Start the grading service:

uvicorn api.main:app --host 0.0.0.0 --port 8000

Configuration

• Ensure Docker is installed and running
• The service stores submissions in service/_submissions/<course>/<student>/<submission_id>/
• Logs are written to service/grader.log

For Teachers

Creating a New Course

1. Create a container and upload it to dockerhub.
   • It has to contain a /app/src/ and /app/output/ directory. These will get mounted to the server.
   • The students code is mounted to /app/src/ and the score needs to be written to /app/output/score.txt

# TODO: Example here

2. Edit the config.yaml in your course directory:

username: example_username
password: example_password

course_name: example_course
container_name: example_user/example_repo:latest # TODO: Check if it is possible to make this a full URL

start_date: 02.12.2024
end_date: 30.12.2024

3. Run the upload script:

python submit_prof.py

For Students

Submitting Assignments

1. Edit your config.yaml:

student_name: "s-username"
course_name: "Programmieren"
server_url: "http://localhost:8000"

2. Create your solution in src/main.py:

# src/main.py
def add(a: int, b: int) -> int:
    return a + b

def multiply(a: int, b: int) -> int:
    return a * b

3. Submit your solution:

python submit_stud.py