twilight-library/app/miastobierun.hs

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 "//td[@cell-header='Nazwa pliku']/a"  -- pary adres-tytuł
                 >>> second (arr $ replace "\r\n" "") -- czyścimy drugi element pary, czyli tytuł z niepotrzebnych białych znaków

-- ... a tutaj te dwójki przerabiamy do docelowej struktury ShadowItem
toShadowItem :: (String, String) -> ShadowItem
toShadowItem (url, articleTitle) =
  (defaultShadowItem url title) {
    originalDate = Just date,
    itype = "periodical",
    format = Just "pdf",
    finalUrl = url,
    description = Just desc
    }
  where title = "Miasto Bierun: " ++ articleTitle
        date = getDate articleTitle
        desc = getArticleNr articleTitle

getDate title =
  case title =~~ "(19[0-9][0-9]|20[0-9][0-9])" :: Maybe [[String]] of
    Just [[_, year]] -> year
    otherwise -> "No date for: " ++ title

getArticleNr title =
  case title =~~ "([0-9][0-9]/)" :: Maybe [[String]] of
    Just [[_, nr]] -> "Article nr: " ++ (replace "/" "" nr)
    otherwise -> "No article nr for: " ++ title


main = do
    let start = "https://www.bierun.pl/mieszkancy/archiwum"
    let shadowLibrary = ShadowLibrary {logoUrl=Nothing,
                                       lname="Miasto Bieruń",
                                       abbrev="Bieruń",
                                       lLevel=0,
                                       webpage=start}
    extractItemsStartingFromUrl shadowLibrary start (extractRecords >>> arr toShadowItem)