Boilerplate for Flask APIs

A skeleton Flask API only application to quickstart development.

Features

• Uses Flask Blueprints for module separation
• ORM-ready with mongoengine and MongoDB.
• Fully dockerized and deployment ready
• CORS protected
• Integrates with Marshmallow
• Sentry Integration
• Documentation generated with Flask-APISpec
• Celery worker queue support.

Deployment

Dockerfile is ready for deployment. It is recommended to reverse proxy through nginx.

Development

Setup

Note: This project requires Python 3.7+ installed. For Mac users, ensure you are using the correct version of Python because the OS preinstalls Python 2.7 and default pip and python commands execute in v2.7 rather than v3.x.

1. Clone this repo or create a new one with this as a template.

2. Create a virtual environment for the project and activate it. Run pip3 install virtualenv if virtualenv is not installed on Python3.7+

   $ cd flask-api-starter  # or your repo name
   $ virtualenv venv --python=python
   $ source venv/bin/activate
   

3. Install the required dependencies, and setup automatic code quality checking with black.

   (flask-starter-venv) $ pip install -r requirements.txt
   (flask-starter-venv) $ pip install -r requirements-dev.txt
   (flask-starter-venv) $ pre-commit install
   

4. Edit config.py.bak with the proper credentials and move it to config.py.

5. Run Migrations

   (flask-starter-venv) $ flask db upgrade
   

Run Locally

Remember to fill any necessary fields in config.py.

1. Make sure you are in your virtualenv that you setup

   $ source flask-starter-venv/bin/activate
   

2. Start server

   (flask-starter-venv) $ flask run
   

3. Start Celery worker

   (flask-starter-venv) $ celery worker -A worker.celery --loglevel=info
   

Setting up Blueprints

Flask blueprints are like modules. To create a new one, you can copy the example blueprint, and modify the __init__.py to change the prefix url. Ensure that it is also included in app.py.