twilight-library/app/chelmek.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 "//div[@class='span4']//h2[@itemprop='name']/a[contains(@href,'o-nas')]"  -- pary adres-tytuł
                 >>> first (extractLinksWithText "//a[contains(@href,'.pdf')]") -- pobieramy stronę z adresu URL i wyciągamy linki z tej strony pasujące do wyrażenia XPathowego
                 -- ostatecznie wyjdą trójki ((adres URL, tytuł artykułu), tytuł rocznika)


toShadowItem :: ((String, String), String) -> ShadowItem
toShadowItem ((url, articleTitle), yearlyTitle) =
  (defaultShadowItem url title) {
    originalDate = Just date,
    itype = "periodical",
   format = Just "pdf",
    finalUrl = url
    }  
  where title = replace "\"" "'" articleTitle
        date = getDate yearlyTitle

getDate yearlyTitle =
  case yearlyTitle =~~ "(19[0-9][0-9]|20[0-9][0-9])" :: Maybe [[String]] of
    Just [[_, year]] -> year
    otherwise -> "unexpected yearlyTitle" ++ yearlyTitle 


main = do
    let start = "http://moksir.chelmek.pl/o-nas/echo-chelmka"
    let shadowLibrary = ShadowLibrary {logoUrl=Nothing,
                                       lname="Chelmek",
                                       abbrev="Chelmek",
                                       lLevel=0,
                                       webpage=start}
    extractItemsStartingFromUrl shadowLibrary start (extractRecords >>> arr toShadowItem)