In a loop in Python, I assign a new instance of a class to the same variable, but it keeps pointing to the old instance? [duplicate]

Question

I created the following class that represents a chest where toys (numbers) can be stored:

class Chest:

  toys = []

  def __init__(self):
    return

  def add(self, num):
    self.toys.append(num)
    return

The main code that uses this class is as follows:

room_of_chests = []

for i in range(3):

  print "Chest", i
  temp = Chest()

  print "Number of toys in the chest:", len(temp.toys)

  for j in range(5):
    temp.add(j)

  print "Number of toys in the chest:", len(temp.toys)
  print ""

  room_of_chests.append(temp)

So, for each iteration of i, I create a new Chest and make the variable temp point to it (correct?). So, in theory, in each iteration, temp would start with an empty chest and end with a chest with 5 toys (correct?).

Therefore, the output I am expecting is:

Chest 0
Number of toys in the chest: 0
Number of toys in the chest: 5

Chest 1
Number of toys in the chest: 0
Number of toys in the chest: 5

Chest 2
Number of toys in the chest: 0
Number of toys in the chest: 5

However, what I am actually getting is:

Chest 0
Number of toys in the chest: 0
Number of toys in the chest: 5

Chest 1
Number of toys in the chest: 5
Number of toys in the chest: 10

Chest 2
Number of toys in the chest: 10
Number of toys in the chest: 15

What am I doing wrong? Can someone give a quick explanation of how instantiation works in this case? And the rules of variables pointing to objects in Python? Thanks in advance.

Answer 1

The problem is that you have a class attribute and not an instance variable. Change your class to make it an instance variable by creating it in the __init__ function as a member of self.

Also, there's no need to use return in __init__.

class Chest:

  def __init__(self):
    self.toys = []


  def add(self, num):
    self.toys.append(num)

This is a common mistake if you're coming from a language like Java or C++.