FastInject: easy, lazy dependency injection for Python

FastInject provides lazy dependency injection for Python that makes your code decoupled, testable, uncomplicated and more readable. Decorate your services with the @injectable decorator and decorate your function with @inject. Done! Your function will now be injected with instances of the required service.

pip install fastinject

Table of Contents

• Main Features
• Usage Example
• How to
• Installation
• Dependencies
• License
• Documentation
• Development
• Contributing to FastInject

---

Main Features

• 🦥 Lazily declare and register your dependencies
• 🐇 Eagerly validate your dependencies
• 🤸 Flexible
• 🎩 Tailor-made for your app
• 👨‍🎨 Easy to use with decorators

---

Usage Example

Below details a minimal use case

Step 1: Declare service to be injectable

We have a service that we want to inject, so we mark it injectable with a decorator:

import time, datetime
from fastinject import injectable

@injectable()           # <-- Just add this decorator to declare the TimeStamp service to be injectable
class TimeStamp:
    ts: float

    def __init__(self) -> None:
        self.ts = time.time()

    @property
    def datetime_str(self) -> str:
        return datetime.datetime.fromtimestamp(self.ts).strftime("%Y-%m-%d %H:%M:%S")

Step 2: Use the service in a function that is injected in

from fastinject import inject

@inject()               # <-- This decorator will inject required services in this function
def function_with_injection(ts: TimeStamp):
    print(f"In the injected function, the current time is {ts.datetime_str}.")

if __name__ == "__main__":
    function_with_injection()

Step 3: Optinonally validate that all injected services can be injected:

from fastinject import get_default_registry

if __name__ == "__main__":
    registry = get_default_registry()

    # Validate
    registry.validate()
    function_with_injection()

---

How to

Injecting services

• Register and inject a single service.
• Register and inject a single service as a singleton.

Inject services that depend on one another

• ServiceConfig: Register multiple dependencies for injection.
• ServiceConfig: Register nested dependencies for injection.

Use the service registy imperatively to get and set dependencies on the fly

• Declare service to be injectable imperatively.
• Declare service to be injectable and declare function to inject imperatively.

Use multiple registries?

• Use multiple registries

Register similar services?

• Register multiple services of the same type?

Validation

• Validated registered services?

---

Installation

pip install fastinject

The source code is currently hosted on GitHub at: https://github.com/mike-huls/fastinject

Binary installers for the latest released version are available at the Python Package Index (PyPI).

---

Dependencies

FastInject is built on injector, aiming to provide additional features and ease-of-use.

---

License

MIT

---

Development

Find the changelog and list of upcoming features here.
Contributions are always welcome; feel free to submit bug reports, bug fixes, feature requests, documentation improvements or enhancements!

---

Go to Top