Starlite
Starlite is a light, opinionated and flexible ASGI API framework built on top of pydantic and Starlette.
Check out the Starlite documentation.
Installation
Using your package manager of choice:
pip install starlite
OR
poetry add starlite
OR
pipenv install starlite
Minimal Example
Define your data model using pydantic or any library based on it (see for example ormar, beanie, SQLModel etc.):
from pydantic import BaseModel, UUID4
class User(BaseModel):
first_name: str
last_name: str
id: UUID4
You can alternatively use a dataclass, either the standard library one or the one from pydantic:
from uuid import UUID
# from pydantic.dataclasses import dataclass
from dataclasses import dataclass
@dataclass
class User:
first_name: str
last_name: str
id: UUID
Define a Controller for your data model:
from pydantic import UUID4
from starlite.controller import Controller
from starlite.handlers import get, post, put, patch, delete
from starlite.types import Partial
from my_app.models import User
class UserController(Controller):
path = "/users"
@post()
async def create(self, data: User) -> User:
...
@get()
async def get_users(self) -> list[User]:
...
@patch(path="/{user_id:uuid}")
async def partial_update_user(self, user_id: UUID4, data: Partial[User]) -> User:
...
@put(path="/{user_id:uuid}")
async def update_user(self, user_id: UUID4, data: list[User]) -> list[User]:
...
@get(path="/{user_id:uuid}")
async def get_user_by_id(self, user_id: UUID4) -> User:
...
@delete(path="/{user_id:uuid}")
async def delete_user_by_id(self, user_id: UUID4) -> User:
...
Import your controller into your application's entry-point and pass it to Starlite when instantiating your app:
from starlite import Starlite
from my_app.controllers.user import UserController
app = Starlite(route_handlers=[UserController])
To run you application, use an ASGI server such as uvicorn:
uvicorn my_app.main:app --reload
Project and Roadmap
This project builds on top the Starlette ASGI toolkit and pydantic modelling to create a higher-order opinionated framework. The idea to use these two libraries as a basis is of course not new - it was first done in FastAPI, which in this regard (and some others) was a source of inspiration for this framework. Nonetheless, Starlite is not FastAPI - it has a different design, different project goals and a completely different codebase.
- The goal of this project is to become a community driven project. That is, not to have a single "owner" but rather a core team of maintainers that leads the project, as well as community contributors.
- Starlite draws inspiration from NestJS - a contemporary TypeScript framework - which places opinions and patterns at its core. As such, the design of the API breaks from the Starlette design and instead offers an opinionated alternative.
- Finally, Python OOP is extremely powerful and versatile. While still allowing for function based endpoints, Starlite seeks to build on this by placing class based Controllers at its core.
Features and roadmap:
- sync and async API endpoints
- fast json serialization using orjson
- class based controllers
- decorators based configuration
- rigorous typing and type inference
- layered dependency injection
- automatic OpenAPI schema generation
- support for pydantic models and pydantic dataclasses
- support for vanilla python dataclasses
- extended testing support
- built-in Redoc based OpenAPI UI
- route guards
- detailed documentation
- schemathesis integration
Contributing
Starlite is open to contributions big and small. You can always join our discord server to discuss contributions and project maintenance. For guidelines on how to contribute, please see the contribution guide.
8 Dec 7, 2021
4 Aug 25, 2022
980 Nov 28, 2022
6k Jan 3, 2023
2 Dec 2, 2021
2.7k Jan 4, 2023
491 Dec 8, 2022
29 Nov 20, 2022
30 Mar 5, 2022
705 Jan 4, 2023
10 Jan 25, 2022
4 Nov 16, 2021
6.6k Jan 7, 2023
525 Jan 1, 2023
1 Nov 8, 2021