Mutable and immutable objects

Mutable and immutable objects#

import numpy as np

There is a distinction in Python between container objects that are mutable and those that are immutable.

Container objects are objects that contain a collection of other objects.

One example of container object is a list:

# A list.  Notice the square brackets.
a = [1, 2, 3]

A list is mutable. This means that we can change the values inside the list without having to make a new value. For example:

a[1] = 99
a

[1, 99, 3]

Notice that a is the same value that we created in the first cell, but the contents of that value have changed — the second element is now 99. The list is mutable.

The tuple is another type of container object. In many ways it is similar to a list, but, by contrast, tuples are immutable by design:

# A tuple.  Notice the round brackets (parentheses).
t = (1, 2, 3)
t

(1, 2, 3)

Trying to set any value contained in the tuple raises an error:

t[1] = 99

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
Cell In[5], line 1
----> 1 t[1] = 99

TypeError: 'tuple' object does not support item assignment

Examples of mutable and immutable containers#

  • Immutable:

    • Tuple (tuple)

    • String (str)

  • Mutable:

    • List (list)

    • Dictionary (dict)

    • Array (np.array)

Strings are containers for characters — a string represents a sequence of characters. They are immutable.

# Setting character in string raises error
s = 'My string'
s[1] = 'i'

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
Cell In[6], line 3
      1 # Setting character in string raises error
      2 s = 'My string'
----> 3 s[1] = 'i'

TypeError: 'str' object does not support item assignment

However, dictionaries are mutable:

wages = dict(Liverpool=160_868_000,
             Everton=80_707_000,
             Leeds=20_800_000)
wages

{'Liverpool': 160868000, 'Everton': 80707000, 'Leeds': 20800000}

We can change the keys and values contained in the dictionary:

wages['Liverpool'] = 164_000_000
# The original wages value has changed.
wages

{'Liverpool': 164000000, 'Everton': 80707000, 'Leeds': 20800000}

Arrays are mutable:

arr = np.array([1, 2, 3])
arr[0] = 99
arr

array([99,  2,  3])

And this matters because?#

Actually, we can often forget about whether an object is mutable or not, but it does sometimes matter.

Contents