Python List Comprehensions

Summary: in this tutorial, you’ll learn about Python List comprehensions that allow you to create a new list from an existing one.

Introduction to Python list comprehensions

In programming, you often need to transform elements of a list and return a new list.

For example, suppose that you have a list of five numbers like this:

numbers = [1, 2, 3, 4, 5]Code language: Python (python)

And you want to get a list of squares based on this numbers list

The straightforward way is to use a for loop:

numbers = [1, 2, 3, 4, 5]

squares = []
for number in numbers:
    squares.append(number**2)

print(squares)
Code language: Python (python)

In this example, the for loop iterates over the elements of the numbers list, squares each number and adds the result to the squares list.

Note that a square number is the product of the number multiplied by itself. For example, square number 2 is 2*2 = 4, square number of 3 is 3*3 = 9, and so on.

To make the code more concise, you can use the built-in map() function with a lambda expression:

numbers = [1, 2, 3, 4, 5]

squares = list(map(lambda number: number**2, numbers))

print(squares)
Code language: Python (python)

Since the map() function returns an iterator, you need to use the list() function to convert the iterator to a list.

Both the for loop and map() function can help you create a new list based on an existing one. But the code isn’t really concise and beautiful.

To help you create a list based on the transformation of elements of an existing list, Python provides a feature called list comprehensions.

The following shows how to use list comprehension to make a list of squares from the numbers list:

numbers = [1, 2, 3, 4, 5]
squares = [number**2 for number in numbers]

print(squares)Code language: Python (python)

And here’s the list comprehension part:

squares = [number**2 for number in numbers]
Code language: Python (python)

A list comprehension consists of the following parts:

• An input list (numbers)
• A variable that represents the elements of the list (number)
• An output expression (number**2) that returns the elements of the output list from the elements of the input list.

The following shows the basic syntax of the Python list comprehension:

[output_expression for element in list]Code language: Python (python)

It’s equivalent to the following:

output_list = []
for element in list:
    output_list.append(output_expression)Code language: Python (python)

Python list comprehension with if condition

The following shows a list of the top five highest mountains on Earth:

mountains = [
    ['Makalu', 8485],
    ['Lhotse', 8516],
    ['Kanchendzonga', 8586],
    ['K2', 8611],
    ['Everest', 8848]
]Code language: Python (python)

To get a list of mountains where the height is greater than 8600 meters, you can use a for loop or the filter() function with a lambda expression like this:

mountains = [
    ['Makalu', 8485],
    ['Lhotse', 8516],
    ['Kanchendzonga', 8586],
    ['K2', 8611],
    ['Everest', 8848]
]


highest_mountains = list(filter(lambda m: m[1] > 8600, mountains))

print(highest_mountains)Code language: Python (python)

Output:

[['K2', 8611], ['Everest', 8848]]Code language: Python (python)

Like the map() function, the filter() function returns an iterator. Therefore, you need to use the list() function to convert the iterator to a list.

Python List comprehensions provide an optional predicate that allows you to specify a condition for the list elements to be included in the new list:

[output_expression for element in list if condition]Code language: Python (python)

This list comprehension allows you to replace the filter() with a lambda expression:

mountains = [
    ['Makalu', 8485],
    ['Lhotse', 8516],
    ['Kanchendzonga', 8586],
    ['K2', 8611],
    ['Everest', 8848]
]

highest_mountains = [m for m in mountains if m[1] > 8600]

print(highest_mountains)Code language: Python (python)

Output:

[['K2', 8611], ['Everest', 8848]]Code language: Python (python)

Summary

• Python list comprehensions allow you to create a new list from an existing one.
• Use list comprehensions instead of map() or filter() to make your code more concise and readable.