How do list comprehensions work and when should I use one?

The short version

A list comprehension builds a list in a single expression, combining the value you want, a for loop, and an optional if filter. Use one for a clear transform or filter. When the logic grows several branches or side effects, a plain loop is more readable, and readability wins.

The basic shape

numbers = [1, 2, 3, 4]

squares = [n * n for n in numbers]
print(squares)   # [1, 4, 9, 16]

Read it left to right: the result value first (n * n), then where the values come from (for n in numbers).

Adding a filter

An if at the end keeps only the items you want.

evens = [n for n in numbers if n % 2 == 0]
print(evens)   # [2, 4]

This is equivalent to a loop with an if and an append, but shorter and, for simple cases, clearer.

The wrong way that people write first

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

Nothing is wrong with this, but for a straight transform the comprehension says the same thing in one line, and the intent (build a list of squares) is visible at a glance.

The if-else placement gotcha

A filtering if goes at the end. A choosing if-else goes at the front, because it is part of the value expression. Mixing these up is a common error.

[n if n > 2 else 0 for n in numbers]   # choose per item: [0, 0, 3, 4]
[n for n in numbers if n > 2]          # filter items: [3, 4]

When not to use one

If you find yourself nesting three loops, adding a ternary inside a ternary, or wanting to run code that does something other than produce a value, stop and write a normal loop. Comprehensions are for building a collection cleanly, not for cramming logic onto one line. A comprehension that a colleague has to decode slowly is worse than a plain loop they read instantly.

Related forms

The same syntax with braces builds sets and dictionaries, and with parentheses builds a lazy generator that produces items on demand rather than all at once.

{n * n for n in numbers}          # set
{n: n * n for n in numbers}       # dict
(n * n for n in numbers)          # generator