Using Lazy ByteStrings and Data.Map to quickly count unique byte occurrences in input of arbitrary length

bytecounter.hs

-- A solution to Rosalind Problem: DNA by SavinaRoja
-- http://rosalind.info/problems/dna/
-- Uses Data.ByteString to remove unnecessary overhead of string or Text
-- representation, and Data.ByteString.Lazy in particular for safety in
-- handling sequence files of arbitrary length.
--
-- The countMap function expresses the solution to the general problem
-- of efficiently counting all unique bytes encountered in a bytestring.
-- This program should perform linearly with respect to sequence length.

-- Test files may be generated by modifying the following command:
-- dd if=/dev/urandom of=sequence bs=1M count=<N>

module Main where

import Data.List
import Data.Word
import qualified Data.ByteString.Lazy as BL
import qualified Data.Map.Strict as M
import System.Environment
import System.IO

countMap :: Num a => BL.ByteString -> M.Map Word8 a
countMap bs = M.fromListWith (+) [(x, 1) | x <- BL.unpack bs]

main :: IO ()
main = do
  args <- getArgs
  contents <- BL.readFile (head args)
  let counts = countMap contents
  let lookupHelper k = show (M.findWithDefault 0 k counts)
  -- This prints out the results in problem answer format: #A, #C, #G, #T
  -- In ASCII: A = 65, C = 67, G = 71, T = 84
  putStr $ intercalate " " $ map lookupHelper [65, 67, 71, 84]