Python: Dynamically Update Array Cells from Parent Program

Question

I previously asked a question about updating a single value in a parallel process from a parent program. Great answers to this can be found at Python: Update Local Variable in a Parallel Process from Parent Program. Now I would like to extend this question to arrays. Say I have the following simple program that iterates through the first cell of an defined array (similar to the link provided):

import time

def loop(i):
    while 1:
        print i[0]
        i[0] += 1
        time.sleep(1)

if __name__ == "__main__":
    from multiprocessing import Process, Array

    i = Array("i", [1,2,3,4])
    p = Process(target=loop, args=(i,))
    p.start()
    time.sleep(2)
    # update i in shared memory???

From this program, how can I update "i" so the "loop" function continues to run and reads the new array? For example, if I want to set "i" to [50,51,52,53]. Is there an equivalent attribute like "value" that can do this? I did much searching around and could not find any solutions. Thank you very much in advance.

unutbu · Accepted Answer · 2013-10-09 21:54:00Z

0

To update i in-place use

i[:] = [50, 51, 52,53]

Note that it is important that you modify i in-place because if you were to assign i to a new value, such as by using

i = [50, 51, 52, 53]

Then i would simply point to a new list, without modifying the shared array.

The Python idiom i[:] = [...] is also used for modifying Python lists in-place, by the way.

import time

def loop(i):
    while 1:
        print list(i)
        # i[0] += 1
        i[:] = [50,51,52,53]
        time.sleep(1)

if __name__ == "__main__":
    from multiprocessing import Process, Array

    i = Array("i", [1,2,3,4])
    p = Process(target=loop, args=(i,))
    p.start()

answered Oct 9, 2013 at 21:54

unutbu

886k197 gold badges1.9k silver badges1.7k bronze badges

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

2 Comments

rmsrms1987 Over a year ago

Thanks you for the response. Is there a way to adjust the size of the list without resulting in an error? Say I want to change the list to i[:] = [50, 51].

unutbu Over a year ago

No, unlike Python lists, the multiprocessing.Array allocates a fixed amount of space. You can neither expand nor shrink the Array. You could artificially "shrink" it by using a multiprocessing.Value to record the last index in the array, or by placing a bogus or sentinel value after the last valid value.

Collectives™ on Stack Overflow

Python: Dynamically Update Array Cells from Parent Program

1 Answer 1

2 Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

2 Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related