Building a list from another list is so common that Python has dedicated syntax for it. First, the long way you already know:

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

for n in numbers:
    doubled.append(n * 2)

print(doubled)

Four lines to say "double everything." A list comprehension says the same thing in one:

numbers = [1, 2, 3, 4]
doubled = [n * 2 for n in numbers]
print(doubled)

The shape

[ expression  for item in sequence ]

Read it right to left: "for each item in sequence, compute expression, and collect the results into a new list."

squares = [n ** 2 for n in range(1, 6)]
print(squares)

words = ["apple", "banana"]
upper = [w.upper() for w in words]
print(upper)

Adding a filter

Put an if at the end to keep only some items.

[ expression  for item in sequence  if condition ]

numbers = range(1, 11)
evens = [n for n in numbers if n % 2 == 0]
print(evens)

That reads as "give me n, for each n in numbers, if n is even." The equivalent loop needs four lines and an append.

You can transform and filter at once:

numbers = range(1, 11)
big_squares = [n ** 2 for n in numbers if n > 7]
print(big_squares)

When not to use one

Comprehensions are for building a list. If you are not building a list — say you just want to print each item — use a normal for loop. This works but is bad style:

[print(n) for n in range(3)]

It builds a throwaway list of None values as a side effect. Just write the loop.

The other limit is readability. If the comprehension needs multiple conditions and a nested loop, it stops being clearer than the loop it replaced. One transformation plus at most one filter is the sweet spot.

Try it yourself

From the numbers 1 to 10, build a list of the squares of the even numbers only.

Checkpoint

• [expr for item in seq] builds a new list in one line.
• Add if condition at the end to filter.
• Comprehensions replace the build-an-empty-list-then-append pattern.
• Use a plain loop when you aren't building a list, or when the logic gets complex.