GitClassrooms

Version: v1.0.0

This project aims to develop a teaching tool integrated with GitLab to manage digital classrooms, distribute assignments, and automate the grading of students' work. Traditional learning management systems often lack integration with version control systems like Git, making it difficult to manage and evaluate programming tasks. This project addresses that gap by creating a standalone application that interacts with GitLab's API. The solution enables instructors to create and manage classrooms, assign programming tasks, and implement both manual and automated grading systems. Through features like OAuth-based authentication, team management, and GitLab repository integration, the tool facilitates efficient course management, offering both autonomy and data sovereignty to educational institutions.

Documents

• Project report (German)
• Presentation slides (German)

Project Structure

The project has two parts:

• Golang with fiber as backend=> /
• React with Typescript and vite as Frontend => /frontend/

The frontend proxies the requests for the path /api/* to the backend server.

Production

Docker Compose is used to deploy the application.

Steps for deployment

1. Cloning the repository
   Alternatively, you can directly use the docker-compose.yaml and .env.example files.
2. Configure environment variables bash cp .env.example .env
3. Creating an application in GitLab
   1. Create a new application in GitLab with the following Redirect URI:
      <PUBLIC_URL>/api/v1/auth/gitlab/callback
   2. Set the application to confidential (Confidential: true) and select the scope "api". These are the default settings.
   3. Copy the Application ID and Secret into the .env file and add the URLs for PUBLIC_URL and GITLAB_URL.
4. SMTP configuration
   Add SMTP credentials to send invitation emails.
5. Configure the database in the .env file.
6. Starting the application
   To start the application and a PostgreSQL database using Docker Compose: bash docker compose -–env-file .env up -d

Alternatively, deployment can also take place with an already present PostgreSQL server by running only the application image

Create .env like specified above and add the database credentials of your existing database. and run the image like:

bash docker run -d --env-file .env --name git-classrooms -p 3000:3000 ghcr.io/git-classrooms/git-classrooms:latest

Development

For development, we use the git flow branching model for simplicity.

Setup

The Setup for development is documented in the following file dev_setup.md

Code Generation

The following file-types in our repo are auto-generated: - mocks - gorm-gen queries - swagger documentation - swagger-client

The first three types are generated when running go generate which is automatically executed before each build when running the app through air.

To generate the frontend swagger-client after some changes to the docs the following script needs to be executed:

```bash

Linux / Mac / WSL

./script/swagger-codegen.sh

Windows

./script/swagger-codegen.ps1 ```

Run dev image locally

To test the completely build image locally you can use the following command:

bash docker compose -f docker-compose.dev.yml up --build

And access the application via http://localhost:3000

Swagger-UI testing (recommended)

1. Run the app described in Dev-Setup
2. Add Swag comments directly above your endpoint
3. Visit the page http://localhost:5173/docs.html
4. Click on the Sign-In-Button on the top-right
5. Get your csrf-token
6. Test your endpoints
7. Refresh page after swag-comments were edited

Postman testing (not recommended)

1. Login via gitlab in the browser
2. Copy the session_id cookie from the browser to your postman environment
3. Copy the csrf_ cookie from the browser to your postman environment
4. Add the following header to your POST|PUT|PATCH|DELETE requests:
   • X-CSRF-Token: {{csrf_}}

Writing Tests

...

To run the test simply exec the following script: ```bash

Linux / Mac / WSL

./script/testbackend.sh ./script/testfrontend.sh

Windows

./script/testbackend.ps1 ./script/testfrontend.ps1 ```

Environments

We have to environments:

• Staging: https//staging.hs-flensburg.dev