Initializing and evaluating Matrix in Python

Question

I was wondering, if someone could explain this behavior in python while initializing matrix in python in the following way

rows,cols = 3,3
visited = [[False]*rows]*cols
visited[0][0] = True 
print visited
[[True, False, False], [True, False, False], [True, False, False]]

As apposed to

visited = []
for i in range(rows):
   visited.append([False]*cols)
visited[0][0] = True
print visited
[[True, False, False], [False, False, False], [False, False, False]]

I am not sure, why in the first case all the elements of first col gets initialized to true.

Amadan · Accepted Answer · 2016-11-11 09:08:45Z

Oswald shot the 35th president of the US, and then a certain John Fitzgerald Kennedy died. Why? Because they're the same person.

[[False]*rows]*cols will first construct an array [False, False, False] - let's call it John - then construct another array that has three Johns in it. Not three copies of John - three references to John.

A simpler demonstration of the same principle:

a = [False]
b = a
a[0] = True
b[0]
# => True

b's [False] is not just a copy of a's [True]: there is only one [True], with two names for it (a and b).

The difference can easily be seen in this visualisation (click "forward" to step through the program):

Three Johns

john = [False] * 3
three_johns = []
for i in range(3):
    three_johns.append(john)
three_johns[1][1] = True

Three random guys

three_random_guys = []
for i in range(3):
    random_guy = [False] * 3
    three_random_guys.append(random_guy)
three_random_guys[1][1] = True

Thus, the easiest way to initialise a 2D matrix properly is this:

[[False] * rows for _ in range(cols)]

Collectives™ on Stack Overflow

Initializing and evaluating Matrix in Python

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related