CODEWITHSUNDEEP
python-3.xpython-decoratorsfunctools
Alesi Rowland
Alesi Rowland

Reputation: 441

@wraps() not behaving isn't returning the original functions values

I've written a fairly basic decorator:

def test_speed(f, *args, **kwargs):
    """This decorator will print out the time taken for input function to run."""

    @wraps(f)
    def wrapper():
        """Wrapper function returned by the outer function."""
        start = time.time()
        to_return = f(*args, **kwargs)
        end = time.time()
        print(f"The function {__name__} completed in {end-start} seconds.")

        return to_return
    return wrapper

In a python script called decorators in a project called tools. I've added this project to my configuration in a second project which I'm using to practice using the multiprocessing module. I wrote some test functions to check the speed of multiprocessing some loops:

""" A script to practice using multiprocessing effectively in python."""

from decorators import *
from multiprocessing import *


def _loop(i, process_number):
    for i in range(i):
        if i % 500 == 0:
            print(f'{i} iterations of loop {process_number}.')


def serial_looping():

    _loop(10000, 'one')
    _loop(10000, 'two')
    _loop(10000, 'three')


@test_speed
def parallel_looping():
    loop_one = Process(target=_loop, args=(10000, 'one'))
    loop_two = Process(target=_loop, args=(10000, 'two'))
    loop_three = Process(target=_loop, args=(10000, 'three'))


if __name__ == '__main__':
    serial_looping()
    parallel_looping()

def serial_looping():

    _loop(10000, 'one')
    _loop(10000, 'two')
    _loop(10000, 'three')


@test_speed
def parallel_looping():
    loops = []
    loops.append(Process(target=_loop, args=(10000, 'one')))
    loops.append(Process(target=_loop, args=(10000, 'two')))
    loops.append(Process(target=_loop, args=(10000, 'three')))
    for loop in loops:
        loop.start()
        print('loop')


if __name__ == '__main__':
    serial_looping()
    parallel_looping()

My issue is that when the wrapped function is called, in place of name it is stating the wrapper's project name, decorator, like so:

The function decorators completed in 5.841255187988281e-05 seconds.

surely this should read: The function serial_looping completed in 5.841255187988281e-05 seconds.

Upvotes: 1

Views: 79

Answers (1)

Reblochon Masque
Reblochon Masque

Reputation: 36682

You should call __name__ on the function f:

print(f"The function {f.__name__} completed in {end-start} seconds.")

The reason is that __name__, by default refers to the enclosing function; I don't think @wraps overrides this behavior; it cannot know what precisely you intend to print.

Upvotes: 1

Related Questions