What is dependency injection?

Dependency injection is a principle that helps to decrease coupling and increase cohesion. Your code becomes more flexible, clear and it is easier to test it.

How to implement dependency injection?

Objects do not create each other anymore. They provide a way to inject the needed dependencies instead.

Before:

import os


class ApiClient:

    def __init__(self):
        self.api_key = os.getenv('API_KEY')
        self.timeout = os.getenv('TIMEOUT')


class Service:

    def __init__(self):
        self.api_client = ApiClient()


if __name__ == '__main__':
    service = Service()

After:

import os


class ApiClient:

    def __init__(self, api_key: str, timeout: int):
        self.api_key = api_key
        self.timeout = timeout


class Service:

    def __init__(self, api_client: ApiClient):
        self.api_client = api_client


if __name__ == '__main__':
    service = Service(ApiClient(os.getenv('API_KEY'), os.getenv('TIMEOUT')))

Flexibility comes with a price: now you need to assemble your objects like this Service(ApiClient(os.getenv('API_KEY'), os.getenv('TIMEOUT'))). The assembly code might get duplicated and it’ll become harder to change the application structure.

What does Dependency Injector do?

Dependency Injector helps you assemble the objects.

It provides you the container and the providers that help you describe objects assembly. When you need an object you get it from the container. The rest of the assembly work is done by the framework:

from dependency_injector import containers, providers


class ApiClient:

    def __init__(self, api_key: str, timeout: int):
        self.api_key = api_key
        self.timeout = timeout


class Service:

    def __init__(self, api_client: ApiClient):
        self.api_client = api_client


class Container(containers.DeclarativeContainer):

    config = providers.Configuration()

    api_client = providers.Singleton(
        ApiClient,
        api_key=config.api_key,
        timeout=config.timeout.as_int(),
    )

    service = providers.Factory(
        Service,
        api_client=api_client,
    )


if __name__ == '__main__':
    container = Container()
    container.config.api_key.from_env('API_KEY')
    container.config.timeout.from_env('TIMEOUT')

    service = container.service()

Retrieving of the Service instance now is done like this container.service().

Also Dependency Injector provides a bonus in overriding any of the providers with the .override() method:

from unittest import mock


with container.api_client.override(mock.Mock()):
    service = container.service()
    assert isinstance(service.api_client, mock.Mock)

It helps in a testing. Also you can use it for configuring project for the different environments: replace an API client with a stub on the dev or stage.

More examples

Installation

The package is available on the PyPi:

pip install dependency-injector

Documentation

The documentation is available on the Read The Docs

Tutorials

Choose one of the following:

• Flask web application tutorial

• Aiohttp REST API tutorial

• Asyncio monitoring daemon tutorial

• CLI application tutorial

Concept

Dependency Injector stands on two principles:

• Explicit is better than implicit (PEP20).

• Do no magic to your code.

How does it different from the other frameworks?

• No autowiring. The framework does NOT do any autowiring / autoresolving of the dependencies. You need to specify everything explicitly. Because “Explicit is better than implicit” (PEP20).

• Does not pollute your code. Your application does NOT know and does NOT depend on the framework. No @inject decorators, annotations, patching or any other magic tricks.

Dependency Injector makes a simple contract with you:

• You tell the framework how to assemble your objects

• The framework does it for you

The power of the Dependency Injector is in its simplicity and straightforwardness. It is a simple tool for the powerful concept.

Frequently asked questions

What is the dependency injection?

• dependency injection is a principle that decreases coupling and increases cohesion

Why should I do the dependency injection?

• your code becomes more flexible, testable and clear

• you have no problems when you need to understand how it works or change it 😎

How do I start doing the dependency injection?

• you start writing the code following the dependency injection principle

• you register all of your application components and their dependencies in the container

• when you need a component, you get it from the container

Why do I need a framework for this?

• you need the framework for this to not create it by your own

• this framework gives you the container and the providers

• the container is like a dictionary with the batteries 🔋

• the providers manage the lifetime of your components, you will need factories, singletons, smart config object etc

What price do I pay and what do I get?

• you need to explicitly specify the dependencies in the container

• it will be extra work in the beginning

• it will payoff when project grows or in two weeks 😊 (when you forget what project was about)

What features does the framework have?

• building objects graph

• smart configuration object

• providers: factory, singleton, thread locals registers, etc

• positional and keyword context injections

• overriding of the objects in any part of the graph

What features the framework does NOT have?

• autowiring / autoresolving of the dependencies

• the annotations and @inject-like decorators

Have a question?

• Open a Github Issue

Found a bug?

• Open a Github Issue

Want to help?

• ⭐️ Star the Dependency Injector on the Github

• 🆕 Start a new project with the Dependency Injector

• 💬 Tell your friend about the Dependency Injector

Want to contribute?

• 🔀 Fork the project

• ⬅️ Open a pull request to the develop branch