Copy a list

[Xx r3negade]

Consider the following code:

listA = [['a', 'b', 'c', 'd'], ['e', 'f', 'g', 'h'], ['i', 'j', 'k', 'l']]
changeCoordinates = ((2, 3), (1, 3))
 
for coordinateList in changeCoordinates:
    listB = listA[:]
    x = coordinateList[0]
    y = coordinateList[1]
 
    listB[x][y] = 'foo'
    print "list A: %s" % listA
    print "list B: %s" % listB
 

listA should never change. But it does. It works the way I expected on a one-dimensional list, but on a two-dimensional list, listA changes. How do I avoid this, and (out of curiosity) why the hell doesn't python just assign the value rather than the reference anyway?

[boxfish]

Hi,
listB is actually a separate copy of listA, but each of the sub-lists in listB is just a shallow copy of the corresponding sub-list in listA, if that makes any sense. To make a completely independent copy of listA, you will either have to use a loop like this:

for sub_list in listA:
    listB.append(sublist[:])

or use the deepcopy() function of the copy module:

import copy
listB = copy.deepcopy(listA)

Hope this helps.

[Xx r3negade]

OK thanks. And about the second part of my question...is there any logical reason for python to do this? Isn't it inconsistent for the assignment operator to assign a value to some data types and a reference to others? Just wondering why they did this.

[boxfish]

Not sure. Seems like the only types that get copied as a reference are lists and object types. I think it would make more sense if tuples, strings, and dictionaries got copied as references too, because they can get really huge also. I wonder if there's a way to do shallow copying of those types in Python.