twilight-library/app/my_robot.hs

{-# 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 "//li/a[contains(@href,'.pdf')]" 


toShadowItem :: (String, String) -> ShadowItem
toShadowItem (url, articleTitle) =
  (defaultShadowItem url title) {
    originalDate = Just date,
    format = Just "pdf",
    finalUrl = url
    }
  where title =  findText url
        date = getDate url


findText url =
  case url =~~ "[0-9]{4}.*" :: Maybe String of
    Just year -> replace "%C5%82" "l" (replace "%20" " " (replace ".pdf" "" year))
    otherwise -> error $ "unexpected url: " ++ url
    

getDate url =
  case url =~~ "[0-9]{4}[-.][0-9]{2}[-.][0-9]{2}|[0-9]{4}" :: Maybe String of
    Just year -> replace "." "-" year
    otherwise -> error $ "unexpected url: " ++ url
  


main = do
    let start = "http://kolej-szyndzielnia.pl/kolej-gondolowa/"
    let shadowLibrary = ShadowLibrary {logoUrl=Nothing,
                                       lname="kolej_gondolowa",
                                       abbrev="kolej_gondolowa",
                                       lLevel=0,
                                       webpage=start}
    extractItemsStartingFromUrl shadowLibrary start (extractRecords >>> arr toShadowItem)