Skip to main content

Dependency injection for python

Project description

drawing By: CenturyBoys

Witch-doctor

A simple dependency injection for python

Register

Witch Doctor provides a method to register interfaces, implementation, injection type, instance args and container name.

  • The interface and implementation inheritance will be checked and will raise a TypeError if was some issue.
  • The injection type will be checked and will raise a TypeError if was some issue. There are two types the singleton and factory types, the singleton will return the same instance for all injection and the factory will return a new instance for each injection.
  • If no values was giving will not pass the args to the class constructor.
  • The container name will segregate the injections by scopes.
class WitchDoctor:
    @classmethod
    def register(  # pylint: disable=R0913
        cls,
        interface: Type[ABC],
        class_ref: Any,
        injection_type: InjectionType,
        args: List[any] = None,
        container: str = DEFAULT,
    ):
        """
        WitchDoctor.register will check inherit of the interface and class_ref.
        Will raise a TypeError on validation error\n
        :param interface: Interface that inherits from ABC
        :param class_ref: A implementation of the interface
        :param injection_type: The injection type that must be used for this register. Allowed Factory or Singleton
        :param args: List of args tha will be used to instantiate the class object
        :param container: Container name where the reference will be saved.
        """
        pass

Container

You can register your injections using containers. The method contianer will provide a container register with the same signature as teh register without the container param. To use the created container you need to load it using load_container. The base work load will set all registers in the DEFAULT group

from abc import ABC, abstractmethod

from witch_doctor import WitchDoctor, InjectionType

class IStubFromABCClass(ABC):
    @abstractmethod
    def sum(self, a: int, b: int):
        pass
    
class StubFromABCClass(IStubFromABCClass):
    def sum(self, a: int, b: int):
        return a + b

container = WitchDoctor.container("prod")
WitchDoctor.register(IStubFromABCClass, StubFromABCClass, InjectionType.SINGLETON)   
WitchDoctor.load_container("prod")

Injection

Witch Doctor can be used as decorator. The function signature will ber check and if some values was not provide Witch Doctor will search on the registered interfaces to inject the dependencies.

class WitchDoctor:
    @classmethod
    def injection(cls, function: Callable):
        """
        WitchDoctor.injection is a function decorator that will match the
        function params signature and inject the  dependencies.
        Will raise AttributeError is some args was pass throw\n

        :type function: Callable
        """
        pass

Usage example

from abc import ABC, abstractmethod

from witch_doctor import WitchDoctor, InjectionType


# Abstract class
class IStubFromABCClass(ABC):
    @abstractmethod
    def sum(self, a: int, b: int):
        pass

    
# Implementation
class StubFromABCClass(IStubFromABCClass):
    def __init__(self, a: int):
        self.a = a

    def sum(self, a: int, b: int):
        return a + b + self.a

# Usage
@WitchDoctor.injection
def func_t(a: int, b: int, c: IStubFromABCClass):
    return c.sum(a, b)

# Containers
container = WitchDoctor.container()
container(IStubFromABCClass, StubFromABCClass, InjectionType.FACTORY, args=[10])

container = WitchDoctor.container("prod")
container(IStubFromABCClass, StubFromABCClass, InjectionType.SINGLETON, args=[20])

# Loading and using
WitchDoctor.load_container()

result_a1 = func_t(a=1, b=2)
result_a2 = func_t(a=2, b=2)

assert result_a1 == 13
assert result_a2 == 14

WitchDoctor.load_container("prod")

result_a1 = func_t(a=1, b=2)
result_a2 = func_t(a=2, b=2)

assert result_a1 == 23
assert result_a2 == 24

Resolve

Witch Doctor can be used by the method resolve. The class signature will ber check and search on the registered interfaces to inject the dependencies.

class WitchDoctor:
    @classmethod
    def resolve(cls, interface: T) -> Type[T]:
        """
        WitchDoctor.resolve will return an instance of the registered class_ref interface.
        Will raise a TypeError if interface is not registered\n
        :param interface: A implementation of the interface
        """
        pass

Usage example

from abc import ABC, abstractmethod

from witch_doctor import WitchDoctor, InjectionType


# Abstract class
class IStubFromABCClass(ABC):
    @abstractmethod
    def sum(self, a: int, b: int):
        pass

    
# Implementation
class StubFromABCClass(IStubFromABCClass):
    def __init__(self, a: int):
        self.a = a

    def sum(self, a: int, b: int):
        return a + b + self.a

# Usage
@WitchDoctor.injection
def func_t(a: int, b: int, c: IStubFromABCClass):
    return c.sum(a, b)

# Containers
container = WitchDoctor.container()
container(IStubFromABCClass, StubFromABCClass, InjectionType.FACTORY, args=[10])

# Loading and using
WitchDoctor.load_container()

result_a1 = WitchDoctor.resolve(IStubFromABCClass)
result_a2 = WitchDoctor.resolve(IStubFromABCClass)

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

witch_doctor-1.2.0.tar.gz (8.3 kB view hashes)

Uploaded Source

Built Distribution

witch_doctor-1.2.0-py3-none-any.whl (8.5 kB view hashes)

Uploaded Python 3

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