forked from filipg/twilight-library
44 lines
1.5 KiB
Haskell
44 lines
1.5 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 "//a[contains(@href, 'index.php?download=gazetka')]" -- pary adres-tytuł
|
||
|
|
>>> second (arr $ replace "\r\n " " ") -- czyścimy drugi element pary, czyli tytuł z niepotrzebnych białych znaków
|
||
|
|
>>> first (arr $ replace "index.php?download=" "pliki/") -- modyfikujemy pierwszy element pary, czyli adres URL
|
||
|
|
|
||
|
|
toShadowItem :: (String, String) -> ShadowItem
|
||
|
|
toShadowItem (url, articleTitle) =
|
||
|
|
(defaultShadowItem url title) {
|
||
|
|
originalDate = Just date,
|
||
|
|
itype = "periodical",
|
||
|
|
format = Just "pdf",
|
||
|
|
finalUrl = url
|
||
|
|
}
|
||
|
|
where title = "Parafia Srem " ++ articleTitle
|
||
|
|
date = getDate url
|
||
|
|
|
||
|
|
getDate url =
|
||
|
|
case url =~~ "(20[0-1][0-9])" :: Maybe [[String]] of
|
||
|
|
Just [[_, year]] -> year
|
||
|
|
otherwise -> error $ "unexpected url: " ++ url
|
||
|
|
|
||
|
|
|
||
|
|
main = do
|
||
|
|
let start = "https://sremfara.pl/miesiecznik_30.html"
|
||
|
|
let shadowLibrary = ShadowLibrary {logoUrl=Nothing,
|
||
|
|
lname="Parafia Sremm",
|
||
|
|
abbrev="ParSre",
|
||
|
|
lLevel=0,
|
||
|
|
webpage=start}
|
||
|
|
extractItemsStartingFromUrl shadowLibrary start (extractRecords >>> arr toShadowItem)
|