Deployment

The application can be deployed to any platform that supports Python WSGI applications. This guide covers deployment to Render and general production considerations.

Render

The repository includes a render.yaml blueprint for one-click deployment.

  1. Fork or clone the repository to your GitHub account.

  2. Create a new Web Service on Render and connect it to your repository.

  3. Set the following environment variables in the Render dashboard:

    Variable

    Value

    FLASK_SECRET_KEY

    A random string (e.g., output of python -c "import secrets; print(secrets.token_hex(32))")

    URAND_CONFIG_FILE

    config-demo.yaml (or your own config file)

    URAND_STUDY_NAME

    Must match a top-level key in your config YAML

    DEMO_MODE

    true to bypass OAuth (demo only)

    GOOGLE_OAUTH_CLIENT_ID

    Your OAuth client ID (production)

    GOOGLE_OAUTH_CLIENT_SECRET

    Your OAuth client secret (production)

  4. Render will automatically build and deploy. The start command is:

    gunicorn urand_gui.app:app

Demo Mode

Setting DEMO_MODE=true bypasses Google OAuth and automatically logs in all visitors as a pre-seeded demo user. This is intended for public demonstrations only — do not use demo mode in production.

To seed the demo database with sample participants:

flask init-demo

This creates 25 participants with randomized factor levels and treatment assignments, spread across several months for a realistic enrollment timeline.

Production Considerations

  • Use a persistent database. The default SQLite file is suitable for single-server deployments. For multi-server setups, use PostgreSQL.

  • Set a strong FLASK_SECRET_KEY.

  • Restrict OAuth users. Only registered users (added via flask add_user) can access the web interface.

  • Use HTTPS. Platforms like Render provide this automatically.

  • Back up your database regularly — it contains all randomization history.