import Data.List.Split
import Data.Word
import Text.HTML.TagSoup
import qualified Data.ByteString.Lazy as B

bsFromString :: [String] -> B.ByteString
bsFromString = B.pack . map (read . ("0x"++))

soupToChunks :: String -> [String]
soupToChunks = filter len2 . wordsBy (`elem` " \n") . fromTagText . getPreText

len2 (_:_:[]) = True
len2 _ = False

getPreText = (!! 1) . last . partitions (isTagOpenName "pre") . parseTags

main = do
  f <- readFile "Cache entry information.xhtml"
  B.writeFile "test" . bsFromString . soupToChunks $ f