Improve UseCase Dev Experience

[acostapazo]

Now we create use case with constructor

class MyUseCase(UseCase):

    def __init__(self, repository: Repository, app_service: AppService) -> None:
         self.repository = repository
         self.app_service = app_service

    def execute(user_id: UUID) -> AnyResult:
        user = self.repository.retrive(user_id).unwrap_or_return()
        return self.app_service.execute(user)

It'd be fantastic to have something like:

class MyUseCase(UseCase):
    repostory: Repository
    app_service: AppService

    def execute(user_id: UUID) -> AnyResult:
         user = self.repository.retrive(user_id).unwrap_or_return()
        return self.app_service.execute(user)

Now it's possible with dataclasses

@dataclass
class MyUseCase(UseCase):
    repostory: Repository
    app_service: AppService

    def execute(user_id: UUID) -> AnyResult:
         user = self.repository.retrive(user_id).unwrap_or_return()
        return self.app_service.execute(user)

[acostapazo]

I'd be very nice to automatize dataclass within current metaclass.

I tried with

 if bases and "UseCase" == bases[0].__name__:
     return dataclass(instance)

in the MetaClass

class MetaUseCase(type, ABC):
    def __new__(
        mcs, name: str, bases: Tuple[Any], namespace: Dict[str, Any]
    ) -> "MetaUseCase":
        new_class_dict = {}
        if "execute" not in namespace:
            raise NotImplementedError(
                "Petisco UseCase or AsyncUseCase must implement an execute method"
            )

        for attributeName, attribute in namespace.items():
            if isinstance(attribute, FunctionType) and attribute.__name__ == "execute":
                if iscoroutinefunction(attribute):
                    attribute = async_use_case_wrapper(attribute, name)
                else:
                    attribute = use_case_wrapper(attribute, name)
            new_class_dict[attributeName] = attribute

        instance = super().__new__(mcs, name, bases, new_class_dict)
         if bases and "UseCase" == bases[0].__name__:
             return dataclass(instance)

        return instance

Unfortunately its not working!

Something similar does pydantic.BaseModel to automatize the __init__ creation from annotations