typings 模块中的重载装饰器似乎没有按预期运行

Overload decorator in typings module doesn't seem to behave as expected

>>> from typing import overload

>>> @overload
... def hello(s: int):
...     return "Got an integer!"

>>> def hello(s: str):
...     return "Got a string"

为什么调用 hello(1) 调用带有字符串参数的函数?理想情况下,@overload 运算符应该处理它,对吗?

不幸的是,python 不允许函数重载。每次你认为你在重载函数时,你只是在覆盖以前的函数声明。引自 docs:

The @overload decorator allows describing functions and methods that support multiple different combinations of argument types. A series of @overload-decorated definitions must be followed by exactly one non-@overload-decorated definition (for the same function/method). The @overload-decorated definitions are for the benefit of the type checker only, since they will be overwritten by the non-@overload-decorated definition, while the latter is used at runtime but should be ignored by a type checker. At runtime, calling a @overload-decorated function directly will raise NotImplementedError.

typing.overload的正确用法如下:

from typing import overload


@overload
def hello(s: int) -> str:
    ...


@overload
def hello(s: str) -> str:
    ...


def hello(s):
    if isinstance(s, int):
        return "Got an integer!"
    if isinstance(s, str):
        return "Got a string"
    raise ValueError('You must pass either int or str')


if __name__ == '__main__':
    print(hello(1))

为了显示 typing.overload 的实际好处,让我们将 def hello(s: int) 更改为 return int 而不是 str:

from typing import overload


@overload
def hello(s: int) -> int:
    ...


@overload
def hello(s: str) -> str:
    ...


def hello(s):
    if isinstance(s, int):
        return "Got an integer!"
    if isinstance(s, str):
        return "Got a string"
    raise ValueError('You must pass either int or str')


if __name__ == '__main__':
    print(hello(1))
    a = hello(1) + 1
    b = hello(1) + 'a'

请注意,实际实现仍然 returns str - python 不会在此处执行任何检查。但是,PyCharm 会发出警告:

mypy 也抱怨无效类型:

➜ mypy test.py 
test.py:25: error: Unsupported operand types for + ("int" and "str")

typing 模块的目的是允许第三方工具对您的代码执行静态检查。这里没有魔法——所有类型在运行时都会被忽略。

# tested in Python 3.8.5 32-bit
# overloads the method
# imports libraries from the base distribution 
# confuses some linters
# undermines type-hinting by documenting *kwargs or dispatch signature

from functools import singledispatch

class Car:
    def __init__(self, color: str, brand: str) -> None:
        self.color = color
        self.brand = brand


@singledispatch
def describe_car(color: str, kind: str) -> str:
    return "Little " + color + " " + kind

@describe_car.register(Car)
def _(car: Car) -> str:
        return describe_car(car.color, car.brand)


newcar = Car("red", "corvette")

print(describe_car("green", "pinto"))
print(describe_car(newcar))

pythonout>
小绿斑
小红护卫舰