Python面试：类的特殊方法

class Foo():
    def __init__(self):
        print("Foo-init")
    def __call__(self):
        print("Foo-call")
f = Foo()
f()  # 重载了括号运算符
"""
Foo-init
Foo-call
"""

class Decorator():
    def __init__(self, func):
        print("Decorator-init")
        self.fun = func
    def __call__(self, args1, args2):
        print("Decorator-call")
        self.fun(args1, args2)
@Decorator
def add(a, b):
    print("a+b:", a+b)
# add(1, 2)
"""
Decorator-init
Decorator-call
a+b: 3
"""

class MyTuple(tuple):
    def __init__(self, seq):
        print("MyTuple-init")
        print("self:", self)
        # takes no parameters
        super(MyTuple, self).__init__()
    def __new__(cls, seq):
        print("MyTuple-new")
        # 过滤
        g = [i for i in seq if isinstance(i ,int)]
        # 返回对象：self
        return super(MyTuple, cls).__new__(cls, g)  
t = MyTuple([1, 2, "-1"])
print(type(t), t)
"""
MyTuple-new
MyTuple-init
self: (1, 2)
<class '__main__.MyTuple'> (1, 2)
"""

# 描述符 : 允许你自定义在引用一个对象属性时应该完成的事情
# __set__:在设计属性的时候被调用
# __get__:在读取属性的时候被调用
# __delete__:在删除属性的时候被调用

class Descriptor():
    def __get__(self, instance, owner):
        print("get")
    def __set__(self, instance, value):
        print("set")
    def __delete__(self, instance):
        print("delete")
class A():
    x = Descriptor()
a = A()
a.x
a.x = 1
del a.x
"""
get
set
delete
"""

class Descriptor():
    def __get__(self, instance, owner):
        print("get")
    def __set__(self, instance, value):
        print("set")
    def __delete__(self, instance):
        print("delete")
class A():
    x = Descriptor()
a = A()
a.x
a.x = 1
del a.x
"""
get
set
delete
"""