List comprehension without saving it vs single line for loop

Question

which is a pythonic and style guide compliant way of operation in case of the following three scenarios.

sq, div2 = [], []

# 1 Classic for loop
for i in range(10):
    sq.append(i**2)
    div2.append(i/2)

# 2 ListComp
[(sq.append(i**2), div2.append(i/2)) for i in range(10)]

# 3 single line for loop
for i in range(10):(sq.append(i**2), div2.append(i/2))

# this is also there but will have multiple loops
sq, div2 = [[i**2 for i in range(10)], [i/2 for i in range(10)]]

or any better way whereby a single for loop generates multiple lists and assign vars.

Answer 1

The first is absolutely the best choice of the four you list.

Don't use a list comprehension unless you actually want the list it produces, and don't cram a for loop on to one line just to make it a one-liner. Having two loops seems a definite drawback to the fourth; you are repeating yourself a bit more than necessary, although you should profile to see if the double iteration makes a significant difference in the runtime. I suspect you would need to be generating very long lists before the difference between one loop and two really matters, though.

---

However, there is one option you have overlooked: producing a sequence of tuples, then "unzipping" it into a tuple of tuples .

sq, div2 = zip(*((i**2, i/2) for i in range(10)))

I think I would still prefer to see the first one, though. It's clearer, without requiring the reader to recognize the unzip idiom.