Understanding Python __call__ and __init__ Magic Methods and Their Use in Class Decorators

In this article, the author explains the difference between Python's __init__ and __call__ magic methods, demonstrates their behavior with example code, and shows a practical use case of __call__ in implementing class decorators.

First, a simple class Person is defined with both __init__ and __call__ methods; creating an instance prints "init" once, while calling the instance repeatedly prints "call". The output illustrates that __init__ runs only during object construction, whereas __call__ enables the instance to be invoked like a function.

class Person:
    def __init__(self):
        print("init")
    def __call__(self):
        print("call")

m1 = Person()
m1()
m1()
m1()

The printed result is:

init
call
call
call

The article then lists the key distinctions: __init__ is the constructor called once per instance, and __call__ makes the object callable, allowing custom behavior on each call. Attempting to call an object without __call__ raises a TypeError:

TypeError: 'Person' object is not callable

Finally, a class decorator example is presented: a MyDecorator class defines __init__ to store a function and __call__ to execute it, printing messages before and after. Applying @MyDecorator to a function hello shows how the decorator runs its __init__ once and __call__ on each function invocation.

class MyDecorator(object):
    def __init__(self, f):
        self.f = f
        print("__init__()...")
    def __call__(self):
        print("__call__()...")
        self.f()

@MyDecorator
def hello():
    print("hello...")

hello()
hello()
hello()

The printed result is:

__init__()...
__call__()...
hello...
__call__()...
hello...
__call__()...
hello...

The author invites readers to share feedback and suggests that more detailed topics like class decorators could be covered in future posts.