Find words in the English language that can be made with the characters given on the command line.

ScrabbleSolver.hs

-- This version runs in ~0.7 seconds due to String actually being a linked list 
-- of Char. Not super efficient.
main = do
  chars <- head <$> getArgs
  englishWords <- words <$> readFile "en.txt"
  mapM_ putStrLn $ sortBy (comparing length) $ findValidWords chars englishWords

findValidWords :: String -> [String] -> [String]
findValidWords chars englishWords = filter (`canBeMadeWith` chars) englishWords

canBeMadeWith :: String -> String -> Bool
str `canBeMadeWith` cs = all hasValidCount str
  where hasValidCount c = (occurrences c str) <= (occurrences c cs)

occurrences :: Char -> String -> Int
occurrences c = length . elemIndices c


-- For this version you need to import ByteStream, and we import the functions
-- from it with the prefix L so they don't interfere. A ByteStream is a packed 
-- array of Chars and is much more efficient. This version runs in ~0.15s and 
-- the only change is having to call L.pack to convert the chars to a ByteStream 
-- and to use the ByteStream version of the functions.
import qualified Data.ByteString.Char8 as L

main = do
  chars <- L.pack <$> head <$> getArgs
  englishWords <- L.words <$> L.readFile "en.txt"
  mapM_ L.putStrLn $ sortBy (comparing L.length) $ findValidWords chars englishWords

findValidWords :: L.ByteString -> [L.ByteString] -> [L.ByteString]
findValidWords chars englishWords = filter (`canBeMadeWith` chars) englishWords

canBeMadeWith :: L.ByteString -> L.ByteString -> Bool
str `canBeMadeWith` cs = L.all charHasValidCount str
  where charHasValidCount c = (occurrences c str) <= (occurrences c cs)

occurrences :: Char -> L.ByteString -> Int
occurrences c = length . L.elemIndices c