Why is my list not getting reset on every loop in Python

Question

I have a function that needs to take in a list called edges. I need to preform many loops and on each loop I change properties values of edges. At the beginning of the next loop I need edges to take on its original value. To try and deal with this I have cast my list as a tuple so it wouldn't be mutable and assigned it to permEdges. I then initialize trialEdges from permEdges so that I can make changes to trialEdges. The changes to trialEdges are staying around for the next loop and I don't know why.

If someone could explain why the output isn't consistent and how I can accomplish this goal of having the same list at the start of every loop I would be forever grateful :)

def randomMinCut(edges):
    permEdges = tuple(edges)
    tries = 3

    for i in xrange(tries):
        trialEdges = list(permEdges)
        print trialEdges
        trialEdges[0][0] = 100

    return

sample input and output:

input:    
[[1, 2], [1, 3], [1, 4], [1, 7]]
output:
[[1, 2], [1, 3], [1, 4], [1, 7]]
[[100, 2], [1, 3], [1, 4], [1, 7]]
[[100, 2], [1, 3], [1, 4], [1, 7]]

The first row is what I want repeated 3 times

@PaulRooney I am confused about how this affect my problem. My understanding of scope is it causes issues when getting values from a for loop but this is all being done in the loop and trialEdges is getting "reset" overtime directly before the print statement — MattTheSnake
– MattTheSnake, Commented Sep 20, 2016 at 1:05

Hamms · Accepted Answer · 2016-09-20 01:04:48Z

4

permEdges may be a tuple, but its contents are not; remember that both permEdges and trialEdges are storing pointers to four two-element lists. When you set trialEdges[0][0] = 100, you're modifying but not replacing the first element of trialEdges, which is also the first element of permEdges.

There are several ways to achieve what you want, but copy.deepcopy is probably the best:

import copy

def randomMinCut(edges):
    tries = 3

    for i in xrange(tries):
        trialEdges = copy.deepcopy(edges)
        print trialEdges
        trialEdges[0][0] = 100

    return

answered Sep 20, 2016 at 1:04

Hamms

5,11724 silver badges30 bronze badges

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

4 Comments

MattTheSnake Over a year ago

Thanks for explaining the issue of using tuple like this. I assumed that anything in a tuple couldn't be changed but I now see what you mean and it makes perfect sense. I really don't want to use deep copy because there will be another loop containing the above for loop that goes through about 1 000 000 iterations. Do you think it would be reasonable to loop over my tuple and make all the sublists tuples too? or is there a better way to do this at the start once?

Hamms Over a year ago

note that deepcopy is going to be just as expensive as manually copying all contents to tuples. Without knowing more about what transformations you're trying to apply (or rather, what transformations you're trying to not apply), I'm not sure what advice to offer.

Hamms Over a year ago

One approach would be to selectively recreate the lists, although this is significantly less elegant, and might apply just to your example and not to the actual problem you're trying to solve; trialEdges = [[100, edges[0][1]]] + edges[1:]

MattTheSnake Over a year ago

Thanks for the suggestion. I ended up using a tuple of tuples. It allowed me to make a copy by casting it as a list of tuples and then swapping out the tuples when I needed too. Ultimately though I picked a very poor data structure that made the code run extremely slow (far to slow for practical purposes) so I am switching to a Union Find data structure.

Collectives™ on Stack Overflow

Why is my list not getting reset on every loop in Python

1 Answer 1

4 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

4 Comments

Your Answer

Sign up or log in

Post as a guest

Related