forked from filipg/twilight-library
55 lines
1.9 KiB
Haskell
55 lines
1.9 KiB
Haskell
|
|
{-# 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 "" -- pary adres-tytuł
|
|
>>> first (extractLinksWithText "//a[@target='_blank'][contains(@href,'.pdf')]") -- pobieramy stronę z adresu URL i wyciągamy linki z tej strony pasujące do wyrażenia XPathowego
|
|
-- ostatecznie wyjdą trójki ((adres URL, tytuł artykułu), tytuł rocznika)
|
|
|
|
|
|
toShadowItem :: ((String, String), String) -> ShadowItem
|
|
toShadowItem ((url, articleTitle), yearlyTitle) =
|
|
(defaultShadowItem url title) {
|
|
originalDate = Just date,
|
|
itype = "periodical",
|
|
format = Just "pdf",
|
|
finalUrl = url
|
|
}
|
|
where homeTitle = getTitle url
|
|
title = replace "_" " " homeTitle ++ "ROK " ++ date ++ " " ++ articleTitle
|
|
date = getDate url
|
|
|
|
|
|
getDate url =
|
|
case url =~~ "_(19[0-9][0-9]|20[0-9][0-9])" :: Maybe [[String]] of
|
|
Just [[_, year]] -> year
|
|
otherwise -> error $ "unexpected url: " ++ url
|
|
|
|
|
|
getTitle url =
|
|
case url =~~ "/([A-Za-z]+_|[A-Za-z]+_[A-Za-z]+_|[A-Za-z]+_[A-Za-z]+_[A-Za-z]+_|[A-Za-z]+_[A-Za-z]+_[A-Za-z]+_[A-Za-z]+_|[A-Za-z]+_[A-Za-z]+_[A-Za-z]+_[A-Za-z]+_[A-Za-z]+_)" :: Maybe [[String]] of
|
|
Just [[_, title]] -> title
|
|
otherwise -> error $ "unexpected url2: " ++ url
|
|
|
|
|
|
main = do
|
|
let start = "https://www.czasopisma.centralnabibliotekapttk.pl"
|
|
let shadowLibrary = ShadowLibrary {logoUrl=Nothing,
|
|
lname="Czasopisma PTTK",
|
|
abbrev="czasopisma",
|
|
lLevel=0,
|
|
webpage=start}
|
|
putStrLn "Program started"
|
|
extractItemsStartingFromUrl shadowLibrary start (extractRecords >>> arr toShadowItem)
|
|
putStrLn "Page sucesfully crawled"
|