Understanding Decorators in Python

In simple words, Decorators in Python Programming Language are used to add additional functionalities to functions or classes. Decorators allow programmers to wrap another function in order to extend the functionality of the wrapped function, without permanently modifying it.

Function Based

Considering the below function which basically just prints ‘Hello World’.

def print_greeting():
   print('Hello John')

The above function’s behavior can be extended from just printing a one-line sentence to doing several things. So let’s create our first decorator.

def extend_our_function(func):
    def wrapper():
        print('function start')
        func()
        print('function end')
    return wrapper

The above decorator function is a regular function that takes another function as an argument, in our case - func, while in it we create another function which we use to extend the functionality of our normal function, in our case print_greeting, and then we return the newly created function - wrapper. To make use of the just created decorator we can simply do this

# supply the function we want to extend its functionality as the argument to our
# decorator
print_greeting = extend_our_function(print_greeting)
# invoke the returned function we store in the variable print_greeting
print_greeting()

But instead of the above, python has a syntax sugar as seen below, we just add the decorator at the top of the function we want to extend its functionality using @ along with the decorator function name.

@extend_our_function
def print_greeting():
   print('Hello John')

If our print_greeting function instead takes an argument,

def print_greeting(name):
    print(f'Hello {name}')

then, we will modify the wrapper function in our decorator to take the name parameter, P.S we can also use - args, *kwargs

def extend_our_function(func):
    def wrapper(name):
        print(f'function start')
        func(name)
        print('function end')
    return wrapper

A more complicated example will be if the decorator itself needs some kind of argument and also the function that is to be extended also takes some arguments.

def expect_result_less_than(max_value):
    def inner_fuction(func):
        def wrapper(*args, **kwargs):
            result = func(*args, **kwargs)
            if result > max_value:
                raise Exception(f'The value should be less than {max_value}')
            return result
        return wrapper
    return inner_fuction

@expect_result_less_than(8)
def add(a, b):
    return a + b

print(add(2,5)) # this passes
print(add(2,12)) # this fails

Question: print(add(2,5)) pass while print(add(2,12)) fails, WHY?

Learn More

Function Based#

Function Based