lesson06_as_completed.py

# SuperFastPython.com
# example of handling results in task completion order
from time import sleep
from random import random
from concurrent.futures import ProcessPoolExecutor
from concurrent.futures import as_completed

# custom function executed by a worker process
def task(number):
 # block for a fraction of a second
 sleep(random())
 # return task number
 return number

# protect the entry point
if __name__ == '__main__':
 # start the process pool
 with ProcessPoolExecutor(10) as exe:
 # submit tasks and collect futures
 futs = [exe.submit(task, i) for i in range(10)]
 # handle task results as they are completed
 for future in as_completed(futs):
 # retrieve the result
 result = future.result()
 # report the result
 print(f'> result for task {result}')