Starlette API

Starlette API aims to bring a layer on top of Starlette to provide a fast and easy way for building highly performant REST APIs.

It is production-ready and provides the following:

Generic classes for API resources that provides standard CRUD methods over SQLAlchemy tables.
Schema system based on Marshmallow that allows to declare the inputs and outputs of endpoints and provides a reliable way of validate data against those schemas.
Dependency Injection that ease the process of managing parameters needed in endpoints. Starlette ASGI objects like Request, Response, Session and so on are defined as components and ready to be injected in your endpoints.
Components as the base of the plugin ecosystem, allowing you to create custom or use those already defined in your endpoints, injected as parameters.
Auto generated API schema using OpenAPI standard. It uses the schema system of your endpoints to extract all the necessary information to generate your API Schema.
Auto generated docs providing a Swagger UI or ReDoc endpoint.
Pagination automatically handled using multiple methods such as limit and offset, page numbers...

Requirements

Python 3.6+
Starlette 0.10+
Marshmallow 3.0+

Installation

$ pip install starlette-api

Example

from marshmallow import Schema, fields, validate
from starlette_api.applications import Starlette


# Data Schema
class Puppy(Schema):
    id = fields.Integer()
    name = fields.String()
    age = fields.Integer(validate=validate.Range(min=0))


# Database
puppies = [
    {"id": 1, "name": "Canna", "age": 6},
    {"id": 2, "name": "Sandy", "age": 12},
]


# Application
app = Starlette(
    components=[],      # Without custom components
    title="Foo",        # API title
    version="0.1",      # API version
    description="Bar",  # API description
    schema="/schema/",  # Path to expose OpenAPI schema
    docs="/docs/",      # Path to expose Swagger UI docs
    redoc="/redoc/",    # Path to expose ReDoc docs
)


# Views
@app.route("/", methods=["GET"])
def list_puppies(name: str = None) -> Puppy(many=True):
    """
    description:
        List the puppies collection. There is an optional query parameter that 
        specifies a name for filtering the collection based on it.
    responses:
        200:
            description: List puppies.
    """
    return [puppy for puppy in puppies if puppy["name"] == name]
    

@app.route("/", methods=["POST"])
def create_puppy(puppy: Puppy) -> Puppy:
    """
    description:
        Create a new puppy using data validated from request body and add it 
        to the collection.
    responses:
        200:
            description: Puppy created successfully.
    """
    puppies.append(puppy)
    
    return puppy

Dependencies

Following Starlette philosophy Starlette API reduce the number of hard dependencies to those that are used as the core:

starlette - Starlette API is a layer on top of it.
marshmallow - Starlette API data schemas and validation.

It does not have any more hard dependencies, but some of them are necessaries to use some features:

pyyaml - Required for API Schema and Docs auto generation.
apispec - Required for API Schema and Docs auto generation.
python-forge - Required for pagination.
sqlalchemy - Required for Generic API resources.
databases - Required for Generic API resources.

You can install all of these with pip3 install starlette-api[full].

Credits

That library started as an adaptation of APIStar to work with Starlette, but a great amount of the code has been rewritten to use Marshmallow as the schema system.

Contributing

This project is absolutely open to contributions so if you have a nice idea, create an issue to let the community discuss it.