Question

I always wrote my list-producing recursive functions in this format:

recursiveFunc :: [a] -> [b]
recursiveFunc (x:xs) = [change x] ++ resursiveFunc xs where
    change :: a -> b
    change x = ...

I realize any function like the above could be written for the a -> b case and then simply maped over a set [a], but please take this watered-down situation as an example.

HLint suggests replacing [change x] ++ recursiveFunc xs with change x : recursiveFunc xs.

Is this suggestion purely aesthetic, or is there some effect on how Haskell executes the function?

Answer 1

When using [change x] ++ recursiveFunc xs you create a superfluous one-element list, which is then taken apart by the ++ function. Using : that doesn’t happen.

Plus ++ is a function which then will use the : constructor. When you use : directly, there’s no need to invoke a function.

So using : is more efficient in theory (the compiler might optimize those differences away, though).