twilight-library/app/pbsociety.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 "//p"
                     >>> second (arr $ replace "\r\n              " "")
--                   >>> first (arr ((++"tr") . init))
                     >>> first (extractLinksWithText "//span//a[contains(@href,'.pdf')]")

toShadowItem :: ((String, String), String) -> ShadowItem
toShadowItem ((url, articleTitle), yearlyTitle) =
  (defaultShadowItem url title) {
    originalDate = Just date,
    itype = "periodical",
    format = Just "pdf",
    finalUrl = url
    }
  where title = "Okragly stol " ++ yearlyTitle ++ " " ++ (replace "\r\n" "" (replace "\r\n          " "" articleTitle))
        date = getDate url
getDate url =
  case url =~~ "/(19[0-9][0-9]|20[0-9][0-9])/(0[1-9]|1[0-2])/" :: Maybe [[String]] of
    Just [[_, year, month]] -> year ++ "-" ++ month
    otherwise -> error $ "unexpected url: " ++ url
	
main = do
    let start = "http://okragly-stol.pl/stenogramy/"
    let shadowLibrary = ShadowLibrary {logoUrl=Nothing,
                                       lname="Okragly stol",
                                       abbrev="OS",
                                       lLevel=0,
                                       webpage=start}
    extractItemsStartingFromUrl shadowLibrary start (extractRecords >>> arr toShadowItem)