IoC container
Project description
💉 sinj
sinj
(Simple Inject) is yet another IoC framework for python. If you try to avoid global variables and singletons you might end up with a complex graph of dependencies in your application. With sinj
everything is just flat. If you are coming from Java or C# you might be more familiar with IoC frameworks where dependencies are resolved by interface. In python we do not have interfaces and strict types so here we resolve dependencies by constructor argument names or "inject labels".
Basic usage and examples
import sinj
c = sinj.Container() # create dependencies container
c.register(SomeClass, "some_class") # add labeled class into container
c.resolve("some_class") # resolve instance by given label
c.inject(some_instance, "some_instance") # add resolved instance to an index
The simplest example:
import sinj
class A:
def a(self):
return "a"
class B:
def __init__(self, a):
self._a = a
def b(self):
return self._a() + " b"
ioc_container = sinj.Container()
ioc_container.register(A, "a")
ioc_container.register(B, "b")
b = ioc_container.resolve("b")
print(b.b()) # -> "a b"
The same example with annotated classes:
import sinj
class A:
inject = "a" # this label will be used to resolve instance of A
def a(self):
return "a"
class B:
inject = "b" # this label will be used to resolve instance of B
def __init__(self, a): # instance of A is injected here
self._a = a
def b(self):
return self._a() + " b"
ioc_container = sinj.Container()
ioc_container.register(A) # no need to specify label here
ioc_container.register(B) # but you can overwrite it if you want
b = ioc_container.resolve("b")
print(b.b()) # -> "a b"
More examples will be available in ./examples
Errors
sinj.DependencyNotFoundError
- thrown onresolve
when dependency is not found for a given label (and it is not optional).sinj.CircularDependencyError
- thrown onresolve
when circular dependency is detected.sinj.DependencyConflictError
- thrown onregister
orinject
when the container already has something by the label.sinj.DependencyNotMappedError
- thrown onregister
orinject
when the class is not annotated and the label is not provided in register method.
Validating types
You have an option to pass a validator into sinj.Container
to check injected types. sinj
has a simple built in validator which is able to log warnings or raise errors. You can create more complicated validators if you want, it is just a class with validate(self, instance, param: inspect.Parameter):
method:
class TypeValidator:
def __init__(self, log_warning=True, raise_error=False):
self._log_warning = log_warning
self._raise_error = raise_error
def validate(self, instance, param: inspect.Parameter):
err_msg = self._get_err_msg(instance, param)
if err_msg is None:
return
if self._log_warning:
logger.warning(err_msg)
if self._raise_error:
raise TypeValidatorError(err_msg)
def _get_err_msg(self, instance, param: inspect.Parameter):
if not hasattr(instance, "__class__"):
return f"could not determine class of instance for {param.name}"
if param.annotation == param.empty:
return None
if not inspect.isclass(param.annotation):
return f"param annotated with a non-class for {param.name}"
if not issubclass(instance.__class__, param.annotation):
return f"instance is not derived from annotated type for {param.name}"
return None
ioc_container = sinj.Container(validator=TypeValidator(True, True))
Install
From pypi.org
pip install sinj
From repository
pip install git+https://gitlab.com/mrsk/sinj
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.