Python 使用数据类将基础 class object 转换为子类

Question

I have an Employee base class and a Manager subclass. The subclass accepts an Employee obj and converts it to Manager.

The below code works fine, however I am trying to determine how I can create the same using a dataclass.

class Employee:
    def __init__(self, id: int, name: str):
        self.id = id
        self.name = name

class Manager(Employee):
    def __init__(self, employee: Employee, region: str):
        super().__init__(**vars(employee))
        self.region = region

e1 = Employee(1, 'Bob')
e2 = Employee(2, 'Sally')

# Sally gets promoted to manager of southwest region
e2 = Manager(e2, 'southwest')

Here is what I have so far, I'm not sure how to pass all the required parameters.

@dataclass
class Employee:
    id: int
    name: str

@dataclass
class Manager(Employee):
    region: str
    def __new__(cls, employee: Employee, region: str):
        obj = object.__new__(cls)
        super().__init__(obj, **vars(employee))
        return obj

e2 = Employee(2, 'Sally')

>>> Manager(e2, region='southwest')
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
TypeError: __init__() missing 1 required positional argument: 'name'

Answer 1

I'm not sure what all of the __new__ metaclass hackery is meant to be accomplishing, but you very seldom need to override __new__ in Python, and this is no exception.

@dataclass
class Employee:
    id: int
    name: str

@dataclass
class Manager(Employee):
    region: str

e2 = Employee(2, 'Sally')
m2 = Manager(2, 'Sally', 'southwest')

Dataclasses are just that; classes meant to store data. Inheritance on dataclasses works like you'd expect. Just inherit and it all works out.

If you'd still rather take an Employee as argument, you can use __init__ rather than __new__

@dataclass
class Employee:
    id: int
    name: str

@dataclass
class Manager(Employee):
    region: str
    def __init__(self, employee: Employee, region: str):
        super().__init__(**vars(employee))
        self.region = region