2022-03-16 11:33:07 +01:00
|
|
|
|
|
|
|
{-# 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
|
|
|
|
|
|
|
|
|
2022-03-16 16:31:16 +01:00
|
|
|
extractRecords = extractLinksWithText "//a[font]" -- pary adres-tytuł
|
2022-03-16 11:33:07 +01:00
|
|
|
|
|
|
|
-- ... a tutaj te trójki przerabiamy do docelowej struktury ShadowItem
|
2022-03-16 16:31:16 +01:00
|
|
|
toShadowItem :: (String, String) -> ShadowItem
|
|
|
|
toShadowItem (url, title) =
|
2022-03-16 11:33:07 +01:00
|
|
|
(defaultShadowItem url title) {
|
2022-03-16 16:31:16 +01:00
|
|
|
originalDate = date,
|
2022-03-16 11:33:07 +01:00
|
|
|
itype = "periodical",
|
|
|
|
format = Just "pdf",
|
|
|
|
finalUrl = url
|
|
|
|
}
|
2022-03-16 16:31:16 +01:00
|
|
|
where date = extractYear title
|
|
|
|
|
|
|
|
extractMonth :: String -> Maybe String
|
|
|
|
extractMonth n =
|
|
|
|
case n =~~ ("/(styczeń|luty|marzec|kwiecien|maj|czerwiec|lipiec|sierpień|wrzesień|październik|listopad|grudzień)/" :: String) of
|
|
|
|
Just month -> Just month
|
|
|
|
otherwise -> Nothing
|
2022-03-16 11:33:07 +01:00
|
|
|
|
|
|
|
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
|
|
|
|
|
|
|
|
|
|
|
|
main = do
|
|
|
|
let start = "http://kobudo.pl/artykuly_jujutsu.html"
|
|
|
|
let shadowLibrary = ShadowLibrary {logoUrl=Nothing,
|
|
|
|
lname="Jujitsu",
|
|
|
|
abbrev="Juji",
|
|
|
|
lLevel=0,
|
|
|
|
webpage="http://kobudo.pl"}
|
|
|
|
|
2022-03-16 16:31:16 +01:00
|
|
|
extractItemsStartingFromUrl shadowLibrary start (extractRecords >>> arr toShadowItem)
|