factors :: Int -> [Int]
factors x = [z | z <- [1..(x `div` 2)], x `mod` z == 0]

is_perfect :: Int -> Bool
is_perfect x = sum(factors x) == x

do_perfect :: [Int] -> [Int]
do_perfect x = [z |z <- x, is_perfect z ]

main = print $ do_perfect [1..9000]
-- compile with ghc -O2, it's more than ten times faster for me
-- althought using 1..(x `div 2) instead of 1..x-1 is a bit of cheating ;-))

On Thu, Oct 2, 2008 at 6:45 AM, Matthew Williams <Matthew_Williams@xyratex.com> wrote:

Hi Guys,

I'm new to Haskell and I was wondering if you can help me:

One of the first program's I tend to write when I'm looking at a new language is a program to generate a list of perfect numbers:

--My First Perfect Number Generator
factors :: Integer -> [Integer]
factors x = [z | z <- [1..x-1], x `mod` z == 0]

is_perfect :: Integer -> Bool
is_perfect x = if sum(factors x) == x then True else False

do_perfect :: [Integer] -> [Integer]
do_perfect x = [z |z <- x, is_perfect z ]

Then to run it:

> do_perfect [1..9000]

I'm using GHC to run it. My problem / question is this: It's running quite a lot slower than equivalent programs in erlang and python. I suspect it's down to the way I've written it. Any thoughts (or comments in general)

Many thanks

Matt