Does Python List Slicing Create Copies or References?

Partial List Slicing Without Copying in Python

Python's list slicing operation generates references to the original list's elements, rather than creating copies. This behavior stems from the fact that slicing preserves the object identities of the elements within the list.

Demonstration

To illustrate this, consider the following list:

<code class="python">L = [1000 + 1, 1000 + 1, 1000 + 1]</code>

Even though these integers have the same value, they are distinct objects:

<code class="python">map(id, L)
[140502922988976, 140502922988952, 140502922988928]</code>

Slicing the list simply copies the references:

<code class="python">b = L[1:3]
map(id, b)
[140502922988952, 140502922988928]</code>

Memory Overhead of Slicing

While slicing doesn't create copies of objects, it does involve copying references. These references add to the overall memory overhead associated with lists:

<code class="python">for i in range(len(L)):
    x = L[:i]
    print('len: {}'.format(len(x)))
    print('size: {}'.format(sys.getsizeof(x)))</code>

This overhead can accumulate, especially when working with memory-intensive objects like large lists or dictionaries.

Alternative: Views

Python lacks an easy way to create views or aliases of lists. However, NumPy arrays provide this capability. Slicing a NumPy array creates a view that shares memory with the original array. This can save significant memory but introduces potential pitfalls when modifying objects.

The above is the detailed content of Does Python List Slicing Create Copies or References?. For more information, please follow other related articles on the PHP Chinese website!