OpenMario is a platform for anonymously collecting and sharing cooperative education (co-op) salary information. This project helps students make informed decisions about co-op opportunities by providing transparent salary data.

• Anonymous submission of co-op salary information
• Search and filter salary data by company, position, location, and more
• User authentication via magic links
• Detailed data visualization of compensation trends
• Company and position autocomplete to ensure data consistency

• Backend: Deno with REST API
• Frontend: React with Vite
• Database: PostgreSQL for data storage
• Search Engine: Meilisearch for fast and relevant search results
• Deployment: Docker and Docker Compose for containerization
• Web Server: Caddy for reverse proxy and SSL termination

• Docker and Docker Compose (only requirement)
• Git (for cloning the repository)

1. Clone the repository:

   git clone https://github.com/yourusername/openmario.git
   cd openmario
   

2. Configure environment variables:

   • Copy .dev.env to .env if running locally
   • Update the following variables in your .env file:
     • MEILI_MASTER_KEY: Set a secure key for Meilisearch
     • EMAIL_API_KEY: API key for your email provider
     • SENDER_EMAIL: Email address for sending magic links
     • DOMAIN or APP_URL: Your application domain
     • JWT_SECRET_MAGIC_LINK: Secret key for magic link tokens
     • JWT_SECRET_CLIENT: Secret key for client authentication
     • Update image names in PRESTART_IMAGE, BACKEND_IMAGE, and REVERSE_PROXY_IMAGE

The application setup requires just two main steps:

1. Seed the database (required on first run):

   docker compose --profile seed up

   This initializes the PostgreSQL database with the necessary tables and initial data.

2. Start all services:

   docker compose --profile deploy up

   This command launches all the required services:

   • PostgreSQL database
   • Meilisearch server
   • Backend API server (Deno)
   • Frontend server with reverse proxy

The application should now be accessible at http://localhost (or the domain you configured).

To stop all services, press Ctrl+C or run:

docker compose --profile deploy down

For development with hot reloading:

docker compose --profile deploy up

This configuration enables:

• File watching and automatic reloading
• Exposed ports for direct access to services
• Development-specific settings

For production deployment:

docker compose --profile deploy -f docker-compose.yml up -d

This will:

• Start all services in detached mode with production configurations
• Use production-optimized Docker images
• Apply memory limits as defined in the docker-compose.yml file

The API follows OpenAPI 3.0 standards. The full API documentation is available in the openapi.json file.

The application uses magic link authentication:

1. Users request a login by providing their email
2. A magic link is sent to their email
3. Clicking the link validates the token and authenticates the user

• Authentication

  • POST /auth/login: Request a magic link
  • GET /auth/login/{token}: Verify magic link token

• Search Autocomplete

  • GET /autocomplete/company: Search for companies
  • GET /autocomplete/position: Search for positions within a company
  • GET /autocomplete/location: Search for locations

• Companies and Positions

  • GET /company-position: Get user's companies and positions
  • POST /company-position: Create a new company-position pair

• Submissions

  • GET /submissions: Retrieve salary submissions with filtering
  • POST /submissions: Create new salary submission(s)
  • PATCH /submissions: Update existing submission
  • GET /submissions/me: Get authenticated user's submissions

.
├── back/                  # Backend code (Deno)
├── front/                 # Frontend code (React + Vite)
├── caddy/                 # Caddy server configuration
├── docker-compose.yml     # Production Docker Compose configuration
├── docker-compose.override.yml # Development overrides
└── .env                   # Environment variables

The application is optimized for low memory usage:

• Backend: 250MB limit, 150MB reservation
• Frontend: 200MB limit, 100MB reservation
• Database: 200MB limit, 100MB reservation
• Meilisearch: 500MB limit, 150MB reservation

The application uses Meilisearch for:

• Fast full-text search
• Typo-tolerant autocomplete
• Filtering and faceting of salary data

A search token is generated for users to perform searches securely.

Contributions are welcome! Please refer to our contribution guidelines for more information.