Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

Awesome Lists | Featured Topics | Projects

https://github.com/volfpeter/fasted

FastAPI dependencies and utilities.
https://github.com/volfpeter/fasted

fastapi python utility

Last synced: 15 days ago
JSON representation

FastAPI dependencies and utilities.

Awesome Lists containing this project

README

        

![Tests](https://github.com/volfpeter/fasted/actions/workflows/tests.yml/badge.svg)
![Linters](https://github.com/volfpeter/fasted/actions/workflows/linters.yml/badge.svg)
![Documentation](https://github.com/volfpeter/fasted/actions/workflows/build-docs.yml/badge.svg)
![PyPI package](https://img.shields.io/pypi/v/fasted?color=%2334D058&label=PyPI%20Package)

**Source code**: [https://github.com/volfpeter/fasted](https://github.com/volfpeter/fasted)

**Documentation and examples**: [https://volfpeter.github.io/fasted](https://volfpeter.github.io/fasted/)

# FastED

FastAPI dependencies and utilities.

## Installation

The package is available on PyPI and can be installed with:

```console
$ pip install fasted
```

## Features

### `selfdependent`

Decorator that let's you use your business objects' instance methods as FastAPI dependencies without writing any additional code.

Supports:

- **Sync and async** instance **methods**.
- **Sync and async generator** methods.
- **Inheritence** and **`super()`** calls is decorated methods.
- An **optional factory** (FastAPI dependency) for creating the `self` instance. If not set, the class' `__init__()` method will serve as the dependency for creating the `self` instance.
- **Decorated** instance **methods will behave as expected** if called directly.

Example use:

```python
from typing import Annotated

from fastapi import Depends, FastAPI
from fasted import selfdependent

def double() -> "Multiplier":
# Dependency that returns a Multiplier with base = 2.
return Multiplier(2)

class Multiplier:
def __init__(self, base: float) -> None:
self.base = base

@selfdependent()
def multiply(self, mul: float) -> float:
# `__init__()` will be used as the dependency to create `self`, so the route
# where this method is used will have a `base` and a `mul` query parameter.
return self.base * mul

@selfdependent(factory=double)
async def double(self, mul: float) -> float:
# `double()` will be used as the dependency to create `self`, so the route
# where this method is used will only have a `mul` query parameter.
return self.base * mul

app = FastAPI()

@app.get("/multiply")
def multiply_route(value: Annotated[float, Depends(Multiplier.multiply)]) -> float:
# FastAPI will create the `Multiplier` instance based on `Multiplier.__init__()` and
# automatically feed this `instance` as `self` to `Multiplier.multiply()` to calculate
# the value of the dependency.
return value

@app.get("/double")
def double_route(value: Annotated[float, Depends(Multiplier.double)]) -> float:
# FastAPI will create the `Multiplier` instance using the `double()` factory (dependency)
# and automatically feed this instance as `self` to `Multiplier.multiply()` to
# calculate the value of the dependency.
return value
```

### `Dependency`

Generic type for FastAPI dependencies.

Example use:

```python
from typing import Annotated, Generator

from fastapi import FastAPI, APIRouter
from fasted import Dependency
# from x import Session

def make_api(make_session: Dependency[Session]) -> APIRouter:
DependsSession = Annotated[Session, Depends(make_session)]

api = APIRouter()

@api.get("/")
def get(session: DependsSession) -> int:
return 4

return api

def make_db_session() -> Generator[Session, None, None]:
with Session(database) as session:
yield session

app = FastAPI()
app.include_router(make_api(make_db_session), prefix="/random-number")
```

## Dependencies

Being a FastAPI utility library, the only dependency is (and will remain) `fastapi`.

## Development

Use `ruff` for linting and formatting, `mypy` for static code analysis, and `pytest` for testing.

The documentation is built with `mkdocs-material` and `mkdocstrings`.

## Contributing

All contributions are welcome.

## License - MIT

The package is open-sourced under the conditions of the [MIT license](https://choosealicense.com/licenses/mit/).