Deploying Flask Applications

# CHAPTER 19

Deploying Flask Applications #

1. Introduction #

2. Learning Objectives #

• Secure sensitive keys using python-dotenv.

• Prepare app.py for a production environment (disabling debug).

• Understand the critical role of Gunicorn (WSGI Server).

• Generate a requirements.txt file for the cloud server.

• Understand the deployment pipeline to platforms like Render or Heroku.

3. Beginner-Friendly Explanation #

4. Step 1: Environment Variables (.env) #

Install: pip install python-dotenv

Create a .env file in the root of your project:

# Do not use quotes!
FLASK_DEBUG=False
SECRET_KEY=my_ultra_secure_random_string_123!
DATABASE_URL=postgres://user:password@cloud-database-url.com/db

CRITICAL: Add .env to your .gitignore file. It must never go to GitHub.

Update app.py:

import os
from flask import Flask
from dotenv import load_dotenv

# Load the variables from the hidden file!
load_dotenv()

app = Flask(__name__)

# Pull the secrets securely using os.environ
app.config['SECRET_KEY'] = os.environ.get('SECRET_KEY')
app.config['SQLALCHEMY_DATABASE_URI'] = os.environ.get('DATABASE_URL')

5. Step 2: The Production Server (Gunicorn) #

Install: pip install gunicorn

Instead of running Python, the live cloud server will start your app using this terminal command:

# syntax: gunicorn [filename]:[flask_app_object_name]
gunicorn app:app

*(Note: Gunicorn only works on Mac/Linux servers. If you develop on Windows, you can test it using a tool called Waitress, but production servers are almost always Linux).*

6. Step 3: The Requirements File #

Run this in your terminal:

pip freeze > requirements.txt

*The cloud server will read this text file and automatically run pip install -r requirements.txt for you during the build process.*

7. Hosting Options: Where to Deploy? #

• PaaS (Render, Heroku, Railway): The modern standard for beginners. You link your GitHub repository. Every time you push code, the platform automatically downloads it, reads requirements.txt, installs dependencies, and turns on Gunicorn. It is fully automated and highly recommended.

• VPS (DigitalOcean, AWS EC2): You rent a blank Linux computer. You must manually install Python, configure an Nginx reverse-proxy to route port 80 to port 5000, and set up Gunicorn as a background system service. Powerful and cheap, but extremely difficult for beginners.

8. Best Practices #

• Cloud Databases (PostgreSQL): SQLite is fantastic for local development, but it fails in production when multiple users try to write to the database simultaneously. You should transition your production app to PostgreSQL. Services like Render or Supabase provide managed Postgres databases with a connection string you simply paste into your cloud platform's Environment Variables dashboard.

9. Common Mistakes #

• Forgetting app:app Configuration: If you used the Application Factory pattern (Chapter 12) and your code is in run.py, the gunicorn command is gunicorn run:app. The first word is the name of the Python file (without .py), and the second word is the name of the Flask object inside that file.

10. Exercises #

1. 1. Explain why the built-in Flask development server (triggered by app.run()) must never be used in a live production environment. What package replaces it?

11. Coding Challenges #

• Challenge: Audit your local Flask project. Install python-dotenv, create a .env file, and successfully move your SECRET_KEY out of your Python code and into the hidden environment file. Ensure the local server still runs.

12. MCQs with Answers #

13. Interview Questions #

• Q: Walk me through the necessary configuration changes to transition a Flask project from a local development environment (app.run(debug=True)) to a secure production environment.

• Q: Explain the role of the requirements.txt file in the deployment pipeline. How does a Platform-as-a-Service (PaaS) utilize this file during the automated build process?

14. FAQs #

15. Summary #

16. Next Chapter Recommendation #