Issue 1408: Inconsistence in multiply list (original) (raw)
There is no way to create a big nested list without references using the multiplication operator.
'*' is supposed to work like + ... + in this cases:
a=[0, 0] b=[a[:]]+[a[:]] b [[0, 0], [0, 0]] b[0][0]=1 b [[1, 0], [0, 0]]
Ok! Copy here, not reference. Mainly because we use [:] explicitly expressing we want a copy.
c=[a[:]]*2 c [[0, 0], [0, 0]] c[0][0]=2 c [[2, 0], [2, 0]]
Inconsistence here. It is supposed to be clear and copy, not reference in between.
Consequence: there is no clear way to create a nested list of, lets say, 60x60, using multiplications.
Even when using this, we cannot deal with the problem:
import copy d=[copy.deepcopy(a[:])]*2 d [[0, 0], [0, 0]] d[0][0]=3 d [[3, 0], [3, 0]]
Workaround:
from numpy import * a=zeros((2,2),int) a array([[0, 0], [0, 0]]) b=a.tolist() b [[0, 0], [0, 0]] b[0][0]=4 b [[4, 0], [0, 0]]
And that is the expected behaviour.
Thanks.
I'm sorry, this is no bug. List multiplication works by referencing, there is no way to implement it differently in a straightforward way.
Note that in
[a[:]] + [a[:]] the expression "a[:]" is evaluated twice, yielding two independent copies of a. In contrast, [a[:]] * 2 evaluates "a[:]" only once, before the list multiplication is done. Because of the same reason, [deepcopy(a)] * 2 doesn't work as you want.
One way to do what you have in mind is
[a[:] for i in range(2)] which evaluates the "a[:]" once for each iteration of the list comprehension loop.