Skip to main content

Python dependency injection library

Project description

pythondi

license pypi pyversions Downloads


pythondi is a lightweight dependency injection library for python

Support both sync and async functions

Installation

pip3 install pythondi

Usage

First, you have to binding classes to provider.

from pythondi import Provider


provider = Provider()
provider.bind(Repo, SQLRepo)
provider.bind(Usecase, CreateUsecase)

After binding, you need to configure it to container

from pythondi import configure, configure_after_clear


# Inject with configure
configure(provider=provider)

# Or if you want to fresh inject, use `configure_after_clear`
configure_after_clear(provider=provider)

Import inject

from pythondi import inject

Add type annotations that you want to inject dependencies

class Usecase:
    def __init__(self, repo: Repo):
        self.repo = repo

Add decorator

class Usecase:
    @inject()
    def __init__(self, repo: Repo):
        self.repo = repo

Initialize class with no arguments

usecase = Usecase()

Or, you can also inject manually through decorator arguments

class Usecase:
    @inject(repo=SQLRepo)
    def __init__(self, repo):
        self.repo = repo

In this case, do not have to configure providers and type annotation.

Lazy initializing

from pythondi import Provider


provider = Provider()
provider.bind(Repo, SQLRepo, lazy=True)

You can use lazy initializing through lazy option. (default False)

For singleton, use lazy=False.

class Usecase:
    @inject(repo=SQLRepo)
    def __init__(self, repo):
        self.repo = repo

By default, manual injection is lazy. If you want a singleton, instantiate it like repo=SQLRepo().

For test

In case of test codes, you probably want to use mock objects.

In that case, you must use keyword arguments.

class MockRepo:
    pass


@inject()
def test(repo: Repo):
    return repo

Yes:

test(repo=MockRepo())

No:

test(MockRepo())
test(MockRepo)

Note

At the moment of inject, class is automatically initialized.

So you don't have to initialize your class inside of code.

Yes:

@inject()
def __init__(self, repo: Repo):
    self.repo = repo

No:

@inject()
def __init__(self, repo: Repo):
    self.repo = repo()

General example

import abc

from pythondi import Provider, configure, configure_after_clear, inject


class Repo:
    """Interface class"""
    __metaclass__ = abc.ABCMeta

    @abc.abstractmethod
    def get(self):
        pass


class SQLRepo(Repo):
    """Impl class"""
    def get(self):
        print('SQLRepo')


class Usecase:
    @inject()
    def __init__(self, repo: Repo):
        self.repo = repo


if __name__ == '__main__':
    # Init provider
    provider = Provider()

    # Bind `Impl` class to `Interface` class
    provider.bind(Repo, SQLRepo)

    # Inject with configure
    configure(provider=provider)

    # Or if you want to fresh injection, use `configure_after_clear`
    configure_after_clear(provider=provider)

    # Init class without arguments
    u = Usecase()

FastAPI example

from fastapi import FastAPI, APIRouter

from pythondi import Provider, configure, inject
import abc

router = APIRouter()


class Repo:
    """Interface class"""
    __metaclass__ = abc.ABCMeta

    @abc.abstractmethod
    def get(self):
        pass


class SQLRepo(Repo):
    """Impl class"""
    def __init__(self):
        pass

    def get(self):
        print('SQLRepo')


@router.route('/')
def home():
    usecase = Usecase()
    usecase.repo.get()
    return {'hello': 'world'}


class Usecase:
    @inject()
    def __init__(self, repo: Repo):
        self.repo = repo


def create_app():
    provider = Provider()
    provider.bind(Repo, SQLRepo)
    configure(provider=provider)
    app = FastAPI()
    app.include_router(router)
    return app


if __name__ == '__main__':
    app = create_app()
    app.run(debug=True)

Flask example

from flask import Flask, Blueprint, jsonify

from pythondi import Provider, configure, inject
import abc

bp = Blueprint('home', __name__)


class Repo:
    """Interface class"""
    __metaclass__ = abc.ABCMeta

    @abc.abstractmethod
    def get(self):
        pass


class SQLRepo(Repo):
    """Impl class"""
    def __init__(self):
        pass

    def get(self):
        print('SQLRepo')


@bp.route('/')
def home():
    usecase = Usecase()
    usecase.repo.get()
    return jsonify({'hello': 'world'})


class Usecase:
    @inject()
    def __init__(self, repo: Repo):
        self.repo = repo


def create_app():
    provider = Provider()
    provider.bind(Repo, SQLRepo)
    configure(provider=provider)
    app = Flask(__name__)
    app.register_blueprint(bp)
    return app


if __name__ == '__main__':
    app = create_app()
    app.run(debug=True)

Sanic example

import abc

from sanic import Sanic, Blueprint
from sanic.response import json

from pythondi import Provider, configure, inject


class Repo:
    """Interface class"""
    __metaclass__ = abc.ABCMeta

    @abc.abstractmethod
    def get(self):
        pass


class SQLRepo(Repo):
    """Impl class"""
    def __init__(self):
        pass

    def get(self):
        print('SQLRepo')


bp = Blueprint('home', url_prefix='/')


@bp.route('/')
async def home(request):
    usecase = Usecase()
    usecase.repo.get()
    return json({'hello': 'world'})


class Usecase:
    @inject()
    def __init__(self, repo: Repo):
        self.repo = repo


def create_app():
    provider = Provider()
    provider.bind(Repo, SQLRepo)
    configure(provider=provider)
    app = Sanic(__name__)
    app.blueprint(bp)
    return app


if __name__ == '__main__':
    app = create_app()
    app.run(debug=True)

Django example

"""
In case of django, just put the initializing code inside of django startup

You can use project folder's __init__.py or urls.py
"""

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

pythondi-1.2.4.tar.gz (7.5 kB view details)

Uploaded Source

File details

Details for the file pythondi-1.2.4.tar.gz.

File metadata

  • Download URL: pythondi-1.2.4.tar.gz
  • Upload date:
  • Size: 7.5 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.1.1 pkginfo/1.5.0.1 requests/2.23.0 setuptools/46.0.0 requests-toolbelt/0.9.1 tqdm/4.45.0 CPython/3.8.2

File hashes

Hashes for pythondi-1.2.4.tar.gz
Algorithm Hash digest
SHA256 9d5b0c13b172b655a170ee4f5fce63e05e93e24138623b11f3b38f3bec6b2c56
MD5 74511a5d6b9b688e1dcc51135125af13
BLAKE2b-256 ff529f64db9f6cc59726b031c86768f2c47dc4c38fe69895b3e2143bd8e0b3d5

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page