python-dependency-injector

Dependency injection framework for Python

4,432

331

4,432

190

View on GitHub

Top Related Projects

pinject

1,348

A pythonic dependency injection library.

injector

1,420

Python dependency injection framework, inspired by Guice

Quick Overview

Python Dependency Injector is a powerful and lightweight dependency injection framework for Python. It provides a set of tools for implementing the Dependency Injection (DI) pattern in Python applications, helping developers create loosely coupled and easily testable code.

Pros

Improves code modularity and maintainability
Facilitates easier unit testing through dependency isolation
Supports various injection methods (constructor, method, attribute)
Integrates well with popular Python frameworks like Flask and FastAPI

Cons

Learning curve for developers new to dependency injection concepts
Can add complexity to smaller projects where DI might be overkill
Potential performance overhead in very large applications with many injections

Code Examples

Basic container and dependency injection:

from dependency_injector import containers, providers

class Container(containers.DeclarativeContainer):
    config = providers.Configuration()
    db = providers.Singleton(Database, db_url=config.db.url)
    user_service = providers.Factory(UserService, db=db)

container = Container()
container.config.from_dict({"db": {"url": "sqlite:///database.db"}})

user_service = container.user_service()

Wiring dependencies in a Flask application:

from flask import Flask
from dependency_injector.wiring import inject, Provide

app = Flask(__name__)

@app.route("/users")
@inject
def get_users(user_service: UserService = Provide[Container.user_service]):
    return user_service.get_all_users()

container.wire(modules=[__name__])

Overriding dependencies for testing:

from unittest.mock import Mock

def test_get_users():
    container = Container()
    container.user_service.override(providers.Factory(Mock))
    
    with container.user_service.override(providers.Factory(Mock)) as mock_service:
        mock_service.return_value.get_all_users.return_value = ["user1", "user2"]
        result = get_users()
        assert result == ["user1", "user2"]

Getting Started

Install the library:

pip install dependency-injector

Create a container with dependencies:

from dependency_injector import containers, providers

class Container(containers.DeclarativeContainer):
    config = providers.Configuration()
    service = providers.Factory(SomeService, param=config.param)

container = Container()
container.config.from_dict({"param": "value"})

Use the container in your application:

service = container.service()
result = service.do_something()

Competitor Comparisons

pinject

1,348

A pythonic dependency injection library.

Pros of Pinject

Simpler API with fewer concepts to learn
Automatic dependency injection without explicit configuration
Built-in support for circular dependencies

Cons of Pinject

Less actively maintained (last update in 2018)
Limited runtime configuration options
Lacks some advanced features like contextual binding

Code Comparison

Pinject:

import pinject

class SomeClass:
    @pinject.inject()
    def __init__(self, foo, bar):
        self.foo = foo
        self.bar = bar

obj_graph = pinject.new_object_graph()
some_instance = obj_graph.provide(SomeClass)

Python Dependency Injector:

from dependency_injector import containers, providers

class Container(containers.DeclarativeContainer):
    foo = providers.Factory(Foo)
    bar = providers.Factory(Bar)
    some_class = providers.Factory(SomeClass, foo=foo, bar=bar)

container = Container()
some_instance = container.some_class()

Python Dependency Injector offers more explicit control over dependency configuration, while Pinject aims for simplicity with automatic injection. Pinject's development has stalled, whereas Python Dependency Injector is actively maintained and provides more advanced features for complex applications.

injector

1,420

Python dependency injection framework, inspired by Guice

Pros of injector

Simpler API with fewer concepts to learn
More Pythonic approach to dependency injection
Better integration with type hints and static type checking

Cons of injector

Less flexible configuration options
Fewer advanced features for complex scenarios
Limited support for runtime configuration changes

Code Comparison

injector:

from injector import inject, Injector

class Database:
    pass

class UserRepository:
    @inject
    def __init__(self, database: Database):
        self.database = database

injector = Injector()
user_repository = injector.get(UserRepository)

python-dependency-injector:

from dependency_injector import containers, providers

class Container(containers.DeclarativeContainer):
    database = providers.Singleton(Database)
    user_repository = providers.Factory(UserRepository, database=database)

container = Container()
user_repository = container.user_repository()

Both libraries provide dependency injection functionality, but injector offers a more straightforward approach with less boilerplate code. python-dependency-injector, on the other hand, provides more explicit configuration and greater flexibility for complex scenarios. The choice between the two depends on the specific needs of your project and personal preference for API style.

Convert designs to code with AI

Introducing Visual Copilot: A new AI model to turn Figma designs to high quality code using your components.

Try Visual Copilot

README

.. figure:: https://raw.githubusercontent.com/wiki/ets-labs/python-dependency-injector/img/logo.svg :target: https://github.com/ets-labs/python-dependency-injector

.. image:: https://img.shields.io/pypi/v/dependency_injector.svg :target: https://pypi.org/project/dependency-injector/ :alt: Latest Version

.. image:: https://img.shields.io/pypi/l/dependency_injector.svg :target: https://pypi.org/project/dependency-injector/ :alt: License

.. image:: https://img.shields.io/pypi/pyversions/dependency_injector.svg :target: https://pypi.org/project/dependency-injector/ :alt: Supported Python versions

.. image:: https://img.shields.io/pypi/implementation/dependency_injector.svg :target: https://pypi.org/project/dependency-injector/ :alt: Supported Python implementations

.. image:: https://pepy.tech/badge/dependency-injector :target: https://pepy.tech/project/dependency-injector :alt: Downloads

.. image:: https://pepy.tech/badge/dependency-injector/month :target: https://pepy.tech/project/dependency-injector :alt: Downloads

.. image:: https://pepy.tech/badge/dependency-injector/week :target: https://pepy.tech/project/dependency-injector :alt: Downloads

.. image:: https://img.shields.io/pypi/wheel/dependency-injector.svg :target: https://pypi.org/project/dependency-injector/ :alt: Wheel

.. image:: https://img.shields.io/github/actions/workflow/status/ets-labs/python-dependency-injector/tests-and-linters.yml?branch=master :target: https://github.com/ets-labs/python-dependency-injector/actions :alt: Build Status

.. image:: https://coveralls.io/repos/github/ets-labs/python-dependency-injector/badge.svg?branch=master :target: https://coveralls.io/github/ets-labs/python-dependency-injector?branch=master :alt: Coverage Status

What is `Dependency Injector`?

Dependency Injector is a dependency injection framework for Python.

It helps implement the dependency injection principle.

Key features of the Dependency Injector:

Providers. Provides Factory, Singleton, Callable, Coroutine, Object, List, Dict, Configuration, Resource, Dependency, and Selector providers that help assemble your objects. See Providers <https://python-dependency-injector.ets-labs.org/providers/index.html>_.
Overriding. Can override any provider by another provider on the fly. This helps in testing and configuring dev/stage environment to replace API clients with stubs etc. See Provider overriding <https://python-dependency-injector.ets-labs.org/providers/overriding.html>_.
Configuration. Reads configuration from yaml, ini, and json files, pydantic settings, environment variables, and dictionaries. See Configuration provider <https://python-dependency-injector.ets-labs.org/providers/configuration.html>_.
Resources. Helps with initialization and configuring of logging, event loop, thread or process pool, etc. Can be used for per-function execution scope in tandem with wiring. See Resource provider <https://python-dependency-injector.ets-labs.org/providers/resource.html>_.
Containers. Provides declarative and dynamic containers. See Containers <https://python-dependency-injector.ets-labs.org/containers/index.html>_.
Wiring. Injects dependencies into functions and methods. Helps integrate with other frameworks: Django, Flask, Aiohttp, Sanic, FastAPI, etc. See Wiring <https://python-dependency-injector.ets-labs.org/wiring.html>_.
Asynchronous. Supports asynchronous injections. See Asynchronous injections <https://python-dependency-injector.ets-labs.org/providers/async.html>_.
Typing. Provides typing stubs, mypy-friendly. See Typing and mypy <https://python-dependency-injector.ets-labs.org/providers/typing_mypy.html>_.
Performance. Fast. Written in Cython.
Maturity. Mature and production-ready. Well-tested, documented, and supported.

.. code-block:: python

from dependency_injector import containers, providers from dependency_injector.wiring import Provide, inject

class Container(containers.DeclarativeContainer):

   config = providers.Configuration()

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

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

@inject def main(service: Service = Provide[Container.service]) -> None: ...

if name == "main": container = Container() container.config.api_key.from_env("API_KEY", required=True) container.config.timeout.from_env("TIMEOUT", as_=int, default=5) container.wire(modules=[name])

   main()  # <-- dependency is injected automatically

   with container.api_client.override(mock.Mock()):
       main()  # <-- overridden dependency is injected automatically

When you call the main() function the Service dependency is assembled and injected automatically.

When you do testing, you call the container.api_client.override() method to replace the real API client with a mock. When you call main(), the mock is injected.

You can override any provider with another provider.

It also helps you in a re-configuring project for different environments: replace an API client with a stub on the dev or stage.

With the Dependency Injector, object assembling is consolidated in a container. Dependency injections are defined explicitly. This makes it easier to understand and change how an application works.

.. figure:: https://raw.githubusercontent.com/wiki/ets-labs/python-dependency-injector/img/di-readme.svg :target: https://github.com/ets-labs/python-dependency-injector

Visit the docs to know more about the Dependency injection and inversion of control in Python <https://python-dependency-injector.ets-labs.org/introduction/di_in_python.html>_.

Installation

The package is available on the PyPi_::

pip install dependency-injector

Documentation

The documentation is available here <https://python-dependency-injector.ets-labs.org/>_.

Examples

Choose one of the following:

Application example (single container) <https://python-dependency-injector.ets-labs.org/examples/application-single-container.html>_
Application example (multiple containers) <https://python-dependency-injector.ets-labs.org/examples/application-multiple-containers.html>_
Decoupled packages example (multiple containers) <https://python-dependency-injector.ets-labs.org/examples/decoupled-packages.html>_
Boto3 example <https://python-dependency-injector.ets-labs.org/examples/boto3.html>_
Django example <https://python-dependency-injector.ets-labs.org/examples/django.html>_
Flask example <https://python-dependency-injector.ets-labs.org/examples/flask.html>_
Aiohttp example <https://python-dependency-injector.ets-labs.org/examples/aiohttp.html>_
Sanic example <https://python-dependency-injector.ets-labs.org/examples/sanic.html>_
FastAPI example <https://python-dependency-injector.ets-labs.org/examples/fastapi.html>_
FastAPI + Redis example <https://python-dependency-injector.ets-labs.org/examples/fastapi-redis.html>_
FastAPI + SQLAlchemy example <https://python-dependency-injector.ets-labs.org/examples/fastapi-sqlalchemy.html>_

Tutorials

Choose one of the following:

Flask web application tutorial <https://python-dependency-injector.ets-labs.org/tutorials/flask.html>_
Aiohttp REST API tutorial <https://python-dependency-injector.ets-labs.org/tutorials/aiohttp.html>_
Asyncio monitoring daemon tutorial <https://python-dependency-injector.ets-labs.org/tutorials/asyncio-daemon.html>_
CLI application tutorial <https://python-dependency-injector.ets-labs.org/tutorials/cli.html>_

Concept

The framework stands on the PEP20 (The Zen of Python) <https://www.python.org/dev/peps/pep-0020/>_ principle:

.. code-block:: bash

Explicit is better than implicit

You need to specify how to assemble and where to inject the dependencies explicitly.

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

Frequently asked questions

What is 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 ð

How do I start applying 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 specify where to inject it or get it from the container

What price do I pay and what do I get?

you need to explicitly specify the dependencies
it will be extra work in the beginning
it will payoff as project grows

Have a question?

Open a Github Issue <https://github.com/ets-labs/python-dependency-injector/issues>_

Found a bug?

Open a Github Issue <https://github.com/ets-labs/python-dependency-injector/issues>_

Want to help?

|star| Star the Dependency Injector on the Github <https://github.com/ets-labs/python-dependency-injector/>_
|new| Start a new project with the Dependency Injector
|tell| Tell your friend about the Dependency Injector

Want to contribute?

|fork| Fork the project
|pull| Open a pull request to the develop branch

.. _PyPi: https://pypi.org/project/dependency-injector/

Top Related Projects

pinject

1,348

A pythonic dependency injection library.

injector

1,420

Python dependency injection framework, inspired by Guice

Convert designs to code with AI

Introducing Visual Copilot: A new AI model to turn Figma designs to high quality code using your components.

Try Visual Copilot

python-dependency-injector

Top Related Projects

pinject

injector

Quick Overview

Pros

Cons

Code Examples

Getting Started

Competitor Comparisons

pinject

Pros of Pinject

Cons of Pinject

Code Comparison

injector

Pros of injector

Cons of injector

Code Comparison

Convert designs to code with AI

README

What is Dependency Injector?

Installation

Documentation

Examples

Tutorials

Concept

Frequently asked questions

Top Related Projects

pinject

injector

Convert designs to code with AI

What is `Dependency Injector`?