{-# LANGUAGE Arrows, NoMonomorphismRestriction #-}
import ShadowLibrary.Core

import Text.XML.HXT.Core
import Text.XML.HXT.XPath
-- import Text.XML.HXT.Curl
import Data.List
import Data.List.Utils (replace)

import Text.Regex.Posix
import Text.Printf

extractRecords = extractLinksWithText "//a[contains(@href,'.pdf')]"
                 >>> second (arr $ replace "\r\n            " " ")

toShadowItem :: ((String, String)) -> ShadowItem
toShadowItem ((url, periodicalTitle)) =
  (defaultShadowItem url title) {
    originalDate = Just date,
    itype = "periodical",
    format = Just "pdf",
    finalUrl = url
    }
  where title = "Sulimczyk - pismo harcerskie" ++ " " ++ (replace "\r\n" "" (replace "\r\n          " "" periodicalTitle))
        date = getDate $ show $ periodicalTitle
        
getDate title = 
    case title =~~ "(([0-9][0-9]{0,1}) ([I|V|X]{1,3}) (19[0-9][0-9]|20[0-9][0-9]))" :: Maybe String of
        Just date -> date
        otherwise ->
          case title =~~ "(0[1-9]|[12][0-9]|3[01]).(0[1-9]|1[0-2]).((1[789]|20)[0-9][0-9])" :: Maybe String of
            Just date -> date
            otherwise ->
              case title =~~ "(19[0-9][0-9]|20[0-9][0-9])" :: Maybe String of
                Just year -> year
                otherwise -> error $ "unexpected title: " ++ title

main = do
    let start = "http://www.16wdh.pl/s28,Sulimczyk"
    let shadowLibrary = ShadowLibrary {logoUrl=Nothing,
                                       lname="Sulimczyk - pismo harcerskie",
                                       abbrev="SulPis",
                                       lLevel=0,
                                       webpage=start}
    extractItemsStartingFromUrl shadowLibrary start (extractRecords >>> arr toShadowItem)