twilight-library/app/naszepopowice.hs

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

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

import Text.Regex.Posix
import Text.Printf


extractRecords = extractLinksWithText "//a[contains(@href,'.pdf') and contains(text(),'popowice')]"

toShadowItem :: (String, String) -> ShadowItem
toShadowItem (url, articleTitle) =
  (defaultShadowItem url title) {
    originalDate = Just date,
    itype = "periodical",
    format = Just "pdf"
    }
  where title = "Nasze Popowice " ++ getNr url
        date = getDate url

replaceSeparatorWithSpace = map (\c -> if c=='_' then ' '; else if c=='-' then ' '; else c)
replaceUnderscoreWithDash = map (\c -> if c=='_' then '-'; else c)

getNr url =
  case url =~~ "nr[-_][0-9][0-9]" :: Maybe String of
    Just nr -> replaceSeparatorWithSpace nr
    otherwise -> error $ "unexpected url: " ++ url

getDate url =
  case url =~~ "20[0-9][0-9]_[0-1][0-9]" :: Maybe String of
    Just date -> replaceUnderscoreWithDash date
    otherwise -> error $ "unexpected url: " ++ url


main = do
    let start = "https://smpopowice.pl/index.php/numery-archiwalne"
    let shadowLibrary = ShadowLibrary {logoUrl=Nothing,
                                       lname="Nasze Popowice",
                                       abbrev="NaszPop",
                                       lLevel=0,
                                       webpage=start}
    extractItemsStartingFromUrl shadowLibrary start (extractRecords >>> arr toShadowItem)