Slicing Lists and Tuples in Python

Internal

• Lists
• Tuples

Overview

Indexing means selecting elements from a list or a tuple using the index operator [...] and a single numeric index.

Slicing means extracting of a section of elements from of a list or a tuple, using the index operator [...] and the slice notation:

[<start-index>:<end-index>:<step>]

The elements are selected as follows: start with the start index, and keep adding the step value to obtain the indices of the selected elements. Stop when the index value such computed is equal or larger than the end index. In case of a negative step, stop if the index value such calculated is equal or smaller than the end index. The end index is the index of the first element not included in the slice.

The algorithm is incomplete, it does not account for the behavior shown in Remove the Last Element.

Start Index

The start index can be missing, and in that case is either 0 for a positive step, or len - 1 for a negative step:

l = ['a', 'b', 'c', 'd', 'e', 'f', 'g', 'h']
#     0    1    2    3    4    5    6    7

assert l[1:3] == ['b', 'c']
assert l[:3] == ['a', 'b', 'c']
assert l[:3:-1] == ['h', 'g', 'f', 'e']

End Index

The end index can be missing, and in that case is either len for a positive step, or - 1 for a negative step:

l = ['a', 'b', 'c', 'd', 'e', 'f', 'g', 'h']
#     0    1    2    3    4    5    6    7

assert l[1:3] == ['b', 'c']
assert l[1:] == ['b', 'c', 'd', 'e', 'f', 'g', 'h']
assert l[1::-1] == ['b', 'a']
assert l[:] == ['a', 'b', 'c', 'd', 'e', 'f', 'g', 'h']

Step

The steps represents the increment added to start index to obtain selected indexes. The elements are selected starting from the start index (inclusive) and then successively adding the step value to obtain the value of the selected indexes, until the index value is equal or larger than the end index for positive steps, or equal or smaller than the end index for negative steps.

If the step is omitted, it defaults to 1. [1:3], [1:3:] and [1:3:1] are equivalent and result in:

l = ['a', 'b', 'c', 'd', 'e', 'f', 'g', 'h']
#     0    1    2    3    4    5    6    7

assert l[1:3] == ['b', 'c']
assert l[1:3:] == ['b', 'c']
assert l[1:3:1] == ['b', 'c']

If the step has a value larger than 1, non-successive elements are selected:

l = ['a', 'b', 'c', 'd', 'e', 'f', 'g', 'h']
#     0    1    2    3    4    5    6    7

assert l[1:6:2] == ['b', 'd', 'f']

If the step has a negative value, the same algorithm applies: the interpreter starts with start index, keeps adding the step in a loop, which now is a negative value, so the iteration advances toward the beginning of the list, and stops when the resulted index is equal or smaller than the end index:

l = ['a', 'b', 'c', 'd', 'e', 'f', 'g', 'h']
#     0    1    2    3    4    5    6    7

assert l[6:3:-1] == ['g', 'f', 'e']

Bare Slice

[:] is called bare slice and represents the entire list.

Slices Are Copies

The indexing operator produces a copy of the underlying data. Modifying a slice does not modifies the original list:

a = [1, 2, 3]
b = a[1:]
b[0]=20
b[1]=30
assert b == [20, 30]
assert a == [1, 2, 3]

Well-Known Use Cases

Copy a List using a Slice

l2 = l[:]

Reverse a List using a Slice

assert l[::-1] == ['h', 'g', 'f', 'e', 'd', 'c', 'b', 'a']

Extract a Sub-List from a Certain Index

assert l[0:] == ['a', 'b', 'c', 'd', 'e', 'f', 'g', 'h']
assert l[1:] == ['b', 'c', 'd', 'e', 'f', 'g', 'h']
assert l[2:] == ['c', 'd', 'e', 'f', 'g', 'h']
assert l[3:] == ['d', 'e', 'f', 'g', 'h']
assert l[4:] == ['e', 'f', 'g', 'h']
assert l[5:] == ['f', 'g', 'h']
assert l[6:] == ['g', 'h']
assert l[7:] == ['h']
assert l[8:] == []

Remove the Last Element

assert l[:-1] == ['a', 'b', 'c', 'd', 'e', 'f', 'g']