Python 3.6 async function over loop in the simplest way possible

Question

All I need to do is write a simple async function that returns something. Something akin to

async def diss(lis):
  x = []
  for i in lis:
    x.append(i + 1) #the operation is arbitrary, but the input and output of a list is the desired result
  return x

lis = [1, 2, 3, 4]
res = await diss(lis)

However, this gives me a syntactical error at await diss(lis)

All I have found online were tutorials where something is printed inside the async function but when trying to return something, I always receive a coroutine or future object.

My goal is to essentially have the loop run asynchronously so to improve performance and then have something returned

Are you trying to await the coroutine outside of an event loop? You need to get or create an event loop and run the coroutine in it. The simplest way would be with asyncio.run — Patrick Haugh
– Patrick Haugh, Commented Sep 10, 2019 at 19:21
as per [stackoverflow.com/questions/52796630/… I found that python 3.6 asyncio does not support run — adam
– adam, Commented Sep 10, 2019 at 19:36

Mikhail Gerasimov · Accepted Answer · 2019-09-11 08:24:31Z

2

Here you go:

import asyncio


async def diss(lis):
  x = []
  for i in lis:
    x.append(i + 1)
  return x


async def main():
  lis = [1, 2, 3, 4]
  res = await diss(lis)
  return res


loop = asyncio.get_event_loop()
res = loop.run_until_complete(main())
print(res)

Make sure you understand why and when to use asyncio in the first place. For example, there's no point to make code above asynchronous.

Usually you only may need asyncio when you have multiple I/O operations which you can parallelize with asyncio.gather().

answered Sep 11, 2019 at 8:24

Mikhail Gerasimov

40.1k20 gold badges126 silver badges172 bronze badges

Sign up to request clarification or add additional context in comments.

3 Comments

adam Over a year ago

Thank you. I guess I am too new to asynchronous operations to have known. This was extremely helpful. The actual operation I need to perform involves calling an external api which makes calculations on a server. I guess I will need to use process pool executor in order to do what I need to. I just need to come up with a way to loop over to put each operation in its own process. Thank you, again

mic Over a year ago

so does it mean we should make this code synchronous rather than asynchronous?

Mikhail Gerasimov Over a year ago

@mic in this case: yes. There's no law that says this code should be synchronous, but making it asynchronous doesn't make sense. You just won't get any benefit out of it. Usually you want to write asynchronous code only when you deal with some I/O or as a way to parallelize some computations (when you can parallelize them). Kindly check the answer from the link, there're few examples there.

adam · Accepted Answer · 2019-09-10 19:27:42Z

0

async def diss(lis):
  x = []
  for i in lis:
    x.append(i + 1)
  return x

async def dat():
  res = await diss(lis)
  return res

loop = asyncio.get_event_loop()

done, _ = loop.run_until_complete(asyncio.wait([dat()]))

for fut in done:
    a = fut.result()

Is what I did to resolve the issue where a is the expected result. Is there anything wrong with this?

answered Sep 10, 2019 at 19:27

adam

471 silver badge5 bronze badges

1 Comment

user4815162342 Over a year ago

It can be written shorter: result = loop.run_until_complete(dat()). You only need asyncio.wait() when awaiting several things in parallel.

Collectives™ on Stack Overflow

Python 3.6 async function over loop in the simplest way possible

2 Answers 2

3 Comments

1 Comment

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

3 Comments

1 Comment

Your Answer

Sign up or log in

Post as a guest

Linked

Related