twilight-library/app/ptd.hs

{-# LANGUAGE UTF #-}
{-# 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

extractNestedLinksWithText xpathCondition = (downloadDocumentWithEncoding "UTF-8" &&& this)
                                      >>> first (getXPathTrees xpathCondition
                                                 >>> ((getXPathTrees "//a" >>> getAttrValue "href")
                                                      &&& (listA (deep isText >>> getText)
                                                           >>> arr (intercalate " "))))
                                      >>> arr rotateSecTh
                                      >>> first expandURIFixed

extractRecords = extractLinksWithText "//div[@class='entry-content']/p/a[contains(@href, 'id')]"
                >>> first (extractNestedLinksWithText "//div[@class='entry-content']/p[strong[a]] | //div[@class='entry-content']/p[a]")

toShadowItem :: ((String, String), String) -> ShadowItem
toShadowItem ((url, articleTitle), magazineTitle) =
  (defaultShadowItem url title) {
    originalDate = Just date,
    itype = "periodical",
    format = Just "pdf",
    finalUrl = url
    }
  where title = magazineTitle ++ " - " ++ (replace " \8211 pdf" "" articleTitle)
        date = getYear url

getYear :: String -> String
getYear url =
  case url =~~ "/rocznik[0-9]{2}/" :: Maybe [[String]] of
    Just [[_, year]] -> year
    otherwise ->  case url =~~ "/(19[0-9][0-9]|20[0-9][0-9])/" :: Maybe [[String]] of
                  Just [[_, year]] ->  "19" ++ (replace "rocznik" "" year)
                  otherwise -> ""


main = do
    let baseUrl = "https://www.ptd.pl/?page_id=7"
    let shadowLibrary = ShadowLibrary {logoUrl=Nothing,
                                    lname="Polskie Towarzystwo Dendrologiczne",
                                    abbrev="ptd",
                                    lLevel=0,
                                    webpage=baseUrl}
    extractItemsStartingFromUrl shadowLibrary baseUrl (extractRecords >>> arr toShadowItem)