run_5.py

#!/usr/bin/env python

from multiprocessing import Process, Pool, Queue
import time
import sys

def bad_ass_function(n):

    t = time.time
    u = t() + n
    while t() < u:
        pass
    return_str = "finished {}s of badass-ery".format(n)
    return return_str

if __name__ == "__main__":
    
    queue = Queue()
    pool = Pool()
    n = 3    

    ## in analogy with "apply"
    #print(pool.map(bad_ass_function,[n,6,8,9]))

    ## ...and with apply_async
    #result = pool.map_async(bad_ass_function,[n,6,8,9])
   
    ### WEIRDNESS: if "result" is not queried with the "ready" (or other) function, processes do not run
    
    #while not result.ready():
    #    print("waiting 2 seconds")
    #    time.sleep(2)
    #print(result.get())

    ## the number of workers wis the pool is configurable, 
    ## and defaults to the number of CPU cores on the machine.
    ## if there are more processes than workers, remaining jobs
    ## will be executed on the next free worker
    #pool = Pool(4)
    #result = pool.map_async(bad_ass_function,[5 for _ in range(20)])

    #while not result.ready():
    #    print("waiting 2 seconds")
    #    time.sleep(2)
    #print("Results:")
    #[sys.stdout.write(i + '\n')for i in result.get()]


"""
 NOTES:

    "map" and "apply" (along with "filter") should remind you 
of scala, and are some of the fundamental building blocks of 
functional programming. Get used to thinking in terms of them.

"""