Skip to main content

Simple asynchronous dependency injector for python

Project description

# ainject Simple asynchronous dependency injector for python.

## Reasons * No asynchronous DI with async/await support. * Simplifying things.

## Features * Asynchronous instance factories. * Asynchronous __init__ and __new__ support. * Damn simple api.

## Requirements * Python 3.5+ * setuptools >= 30.3.0 (installation only)

## Usage All you have to do is bind some factories with some names and inject/instance them where you want. ` python >>> import ainject ` ### Bind and instance Simple bind with name: ` python >>> async def async_factory(): ... return "async_value" ... >>> def sync_factory(): ... return "sync_value" >>> ainject.bind(async_factory, name="async") >>> await ainject.instance("async") 'async_value' >>> ainject.bind(sync_factory, name="sync") >>> await ainject.instance("sync") 'sync_value' >>> ` As you can see you should always await your result, even if factory is actually synchronous.

Bind without name: ` python >>> ainject.bind(async_factory) >>> await ainject.instance(async_factory) 'async_value' >>> ainject.bind(sync_factory) >>> await ainject.instance(sync_factory) 'sync_value' >>> ` In this case name is factory itself. This is equivalent to: ` python >>> ainject.bind(async_factory, name=async_factory) >>> ` So, you can use any hashable value for name, or even omit it for auto naming.

By default binding is done in «singleton» mode. This means, that first time instance is accessed it will be cashed and for every next instance request cashed version will be used: ` python >>> def factory(): ... return [] ... >>> ainject.bind(factory) >>> a = await ainject.instance(factory) >>> b = await ainject.instance(factory) >>> a, b ([], []) >>> a is b True >>> ` For non-singleton usage pass singleton=False to bind method. In this case every instantiation will actually execute factory function: ` python >>> def factory(): ... return [] ... >>> ainject.bind(factory, singleton=False) >>> a = await ainject.instance(factory) >>> b = await ainject.instance(factory) >>> a, b ([], []) >>> a is b False >>> ` ### Inject Injecting is done via inject decorator: ` python >>> @ainject.inject(x=factory) ... def foo(x): ... print(x) ... ... >>> await foo() [] >>> ` Keep in mind that «name» should be defined before decorator, or just use strings for names. Also, remember, that everything you wrap with inject decorator became awaitable: ` python >>> @ainject.inject(x="async") ... class A: ... def __init__(self, x): ... self.x = x ... ... ... >>> a = await A() >>> ` Even class instantiation. Side-effect of this «magic» is that you can use async __init__ and async __new__: ` python >>> @ainject.inject() ... class A: ... async def __new__(self, x): ... ... ... return super().__new__(self) ... async def __init__(self, x): ... self.x = x ... ... ... >>> a = await A(3) ` As you can see you can even inject nothing.

## Advanced usage Most of time you only need above scenarios, but if you need low-level access to injector, or use more than one injector you should instantiate Injector class and use its bind, inject and instance methods. Default injector is global for ainject module and can be accessed as ainject._injector. Bindings are stored as dictionary with name-factory pairs in Injector._bindings. Instances (singletons) stored as name-instance pairs in Injector._instances.

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

ainject-0.0.3.tar.gz (3.1 kB view hashes)

Uploaded source

Built Distribution

ainject-0.0.3-py3-none-any.whl (3.4 kB view hashes)

Uploaded py3

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