VMChecker Next API

The VMChecker API is built on Django. The project is made up of two parts, the API that exposes the URLs for Moodle, and a second process, that submits assignments to Gitlab, polls for the result or retrieves the output/assignment archive. It uses the Postgres database as a task queue.

Requirements

1. python3 >= 3.10
2. pipenv (pip3 install pipenv --user. See docs)
3. docker (See official docker docs)
4. Create a test assignment on Gitlab.com by following the TA handbook. You will only require the private git repository. You can directly fork the following template assignment.

Setting up the development environment

1. After cloning the repository install the python packages using pipenv install.
2. Clone the Vmchecker Moodle plugin from: VMChecker Next and change the compose-dev moodle service's path to point to it
3. Start the environment using ./bin/dev.sh up --watch

First time configuration:

Note: The first time you run dev.sh the moodle service might fail. Comment out the vmchecker volume bind and do dev.sh down && dev.sh up --watch.

If this is the first time starting up the development environment you'll have to set up a Moodle assignment and configure the Moodle plugin.

To configure the Moodle block plugin:

1. Go to Site Administration > Plugins > Blocks > VMChecker block
2. Change vmck backend to http://backend:8000/api/v1/

To set up the Moodle assignment create a course and some users and add them to the course. Then you can follow the TA handbook on how to prepare the assignment.

Useful notes:

• Moodle is avaialbe at http://localhost/
  • User: admin, Password: Parola-123
  • If you are also developing the Moodle plugin check out the Moodle debugging notes
• VMChecker service port 8000

Debugging

The backend service exposes port 5678 for debugging. A VS Code configuration example:

{
            "name": "Python Debugger: Remote Attach",
            "type": "debugpy",
            "request": "attach",
            "connect": {
                "host": "localhost",
                "port": 5678
            },
            "pathMappings": [
                {
                    "localRoot": "${workspaceFolder}",
                    "remoteRoot": "/opt/api"
                }
            ]
        }

Running tests

pipenv run pytest test/