twilight-library/app/instytutTechnikiGorniczej.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 "//li/a[contains(@href,'/lata')]"
                   >>> first (extractLinksWithText "//li/a[contains(@href,'start=')]"
                               >>> first (extractLinksWithText "//a[contains(@href, '/archiwum/images/19') or contains(@href, '/Trybuna')]"))


toShadowItem :: (((String, String), String), String) -> ShadowItem
toShadowItem (((url, articleTitle), _), yearlyTitle) =
  (defaultShadowItem url title) {
    originalDate = Just date,
    itype = "periodical",
    format = Just form,
    finalUrl = url
    }
  where title = "Instytut Techniki Gorniczej " ++ yearlyTitle ++ " " ++ (replace "\r\n" "" (replace "\r\n          " "" articleTitle))
        date = getDate $ replace "%20" "" url
        form = unwords $ map stringToLower [getFormat url]

getDate url =
  case url =~~ "(19[0-9][0-9]|20[0-9][0-9]).*\\..{3}" :: Maybe [[String]] of
    Just [[_, year]] -> year
    otherwise -> error $ "unexpected url: " ++ url
    
getFormat url =
  case url =~~ "([a-zA-Z]+$)" :: Maybe [[String]] of
    Just [[_, format]] -> format
    otherwise -> error $ "unexpected url: " ++ url

main = do
    let start = "http://komag.gliwice.pl/archiwum/historia-komag/"
    let shadowLibrary = ShadowLibrary {logoUrl=Nothing,
                                       lname="Instytut Techniki Gorniczej",
                                       abbrev="Komag",
                                       lLevel=0,
                                       webpage=start}
    extractItemsStartingFromUrl shadowLibrary start (extractRecords >>> arr toShadowItem)