This template is intended to be used as a starting point for a new project that uses FastAPI and has a database. It is designed to be scalable and easy to use. The current implementation makes use of SQLAlchemy, but it can be easily replaced with any other ORM.
- Python 3.11
- pipenv
- Standalone FastAPI app
- Dynamic endpoints based on Pydantic (v2) models
- Automatic field validation
- Scalable project structure
- Sample endpoints to get started
- Customizable HTTP errors
- Generated API documentation
- Clone this repo
- Install pipenv with
pip install pipenv - Run
pipenv installto install dependencies - Run
uvicorn main:app --reloadto start the app for development
To start development you can use the sample endpoints in project/api/endpoints/user.py as a starting
point. Some basic RESTful endpoints are defined here.
This repo revolves around Endpoint classes. These classes are defined in project/api/endpoints/base.py.
They are used to define endpoints for entities. The Endpoint class takes an InputModel and an OutputModel
To add an endpoint for the user entity, copy a similar endpoint and define the in- and output as needed.
Models inherit from Pydantic BaseModel; this adds automatic field validation.
To tweak or turn off validation, you can modify the InputModel and OutputModel base classes.
To tweak endpoint behaviour, change or subclass EndpointConfig and pass them to the Endpoint classes as you see fit.
To add endpoints for new entities, you can simply follow the same pattern as used for the user entity.
This means copying all project/api/*/user.py files and refactoring the copies to fit the new entity.
Run the app and go to http://localhost:8000/docs.
Documentation on FastAPI can be found here.
Documentation on Pydantic can be found here.