Python Decorators

Functions as Objects

Recall that Python functions can be treated as variables. For example:

# inner function
def func_factory():
    # outer function
    def a_func():
        print('This is a_func here!')
    
    return a_func

func = func_factory()
print(type(func))
func()

This outputs <class 'function'> and This is a_func here.

Python Decorators

Decorators are a design pattern where a function is wrapped within another function. This allows us to extend its behavior without changing the original behavior.

def my_decorator(func):
    def wrapper():
        print("Something before the function runs...")
        func()
        print("Something after the function runs...")

    return wrapper

Python provides syntactic sugar to let us wrap functions in this decorator:

@my_decorator
def say_hello():
    print("Hello!")

say_hello()

where @my_decorator is shorthand for say_hello = my_decorator(say_hello).

When we run this code, we print:

Something before the function runs...
Hello!
Something after the function runs...

Real World Example

The following retry() function calls a (presumably flaky) function a few times, allowing you to specify the max number of retries and the delay between retries:

import time

def retry(max_retries=3, delay=1):
    def decorator(func):
        def wrapper(*args, **kwargs):
            for attempt in range(1, max_retries + 1):
                try:
                    return func(*args, **kwargs)
                except Exception as e:
                    print(f"Attempt {attempt} failed: {e}")
                    if attempt == max_retries:
                        raise
                    time.sleep(delay)
        return wrapper
    return decorator

Why Decorators?

These exist in other languages in some form, but are especially popular in Python because:

• Python functions can be treated as objects/variables.
• The @decorator syntax is pretty and concise.
• Python is interpreted and dyanmically typed, so decorators are lightweight.

Pythonic web frameworks (Flask, Django, FastAPI) and the overall Python ecosystem heavily rely on decorators. With that said, let's proceed to Flask!