63 lines
2.1 KiB
Haskell
63 lines
2.1 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, '.pdf')]" -- pary adres-tytuł
|
|
>>> second (arr $ replace "\n\t\t\t\t\t\t\t\t\t\t\t" "") -- czyścimy drugi element pary, czyli tytuł z niepotrzebnych białych znaków
|
|
>>> second (arr $ replace "\n\t\t\t\t\t\t\t\t\t\t" "") -- czyścimy drugi element pary, czyli tytuł z niepotrzebnych białych znaków
|
|
>>> second (arr $ replace "\324" "n")
|
|
>>> second (arr $ replace "\281" "e")
|
|
>>> second (arr $ replace "\380" "z")
|
|
>>> second (arr $ replace "\322" "l")
|
|
>>> second (arr $ replace "\243" "o")
|
|
>>> second (arr $ replace "\347" "s")
|
|
>>> second (arr $ replace "\263" "c")
|
|
>>> second (arr $ replace "\346" "S")
|
|
|
|
|
|
|
|
-- ... a tutaj te trójki przerabiamy do docelowej struktury ShadowItem
|
|
toShadowItem :: (String, String) -> ShadowItem
|
|
toShadowItem (url, articleTitle) =
|
|
(defaultShadowItem url title) {
|
|
originalDate = extractMonthAndYear title,
|
|
itype = "periodical",
|
|
format = Just "pdf",
|
|
finalUrl = url,
|
|
description = extractFileSize title
|
|
}
|
|
where title = articleTitle
|
|
|
|
|
|
extractMonthAndYear :: String -> Maybe String
|
|
extractMonthAndYear n =
|
|
case n =~~ ("[a-z]* (1[6789]|20)[0-9][0-9]" :: String) of
|
|
Just year -> Just year
|
|
otherwise -> Nothing
|
|
|
|
extractFileSize :: String -> Maybe String
|
|
extractFileSize n =
|
|
case n =~~ ("([0-9]*)\\.([0-9]*) (B|kB|MB)" :: String) of
|
|
Just fileSize -> Just fileSize
|
|
otherwise -> Nothing
|
|
|
|
main = do
|
|
let start = "http://podkarpacki.civitaschristiana.pl/formacja/zeszyty-formacyjne/"
|
|
let shadowLibrary = ShadowLibrary {
|
|
logoUrl=Nothing,
|
|
lname="Podkarpacki",
|
|
abbrev="podk",
|
|
lLevel=0,
|
|
webpage=start}
|
|
extractItemsStartingFromUrl shadowLibrary start (extractRecords >>> arr toShadowItem)
|