twilight-library/app/robot.hs
2021-04-05 22:45:06 +02:00

38 lines
1.3 KiB
Haskell

{-# 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"
>>> first (extractLinksWithText "//a[contains(@href,'.pdf')]")
toShadowItem :: ((String, String), String) -> ShadowItem
toShadowItem ((url, articleTitle), yearlyTitle) =
(defaultShadowItem url title) {
originalDate = extractMonthAndYear title,
itype = "periodical",
format = Just "pdf",
finalUrl = url
}
where title = "Kosmos " ++ yearlyTitle ++ (replace ".pdf" "" articleTitle)
extractMonthAndYear :: String -> Maybe String
extractMonthAndYear n =
case n =~~ ("(1[6789]|20)[0-9][0-9]" :: String) of
Just year -> Just year
otherwise -> Nothing
main = do
let start = "http://kosmos.icm.edu.pl/PDF/"
let shadowLibrary = ShadowLibrary {logoUrl=Nothing,
lname="kosmos.icm.edu",
abbrev="Kos",
lLevel=0,
webpage=start}
extractItemsStartingFromUrl shadowLibrary start (extractRecords >>> arr toShadowItem)