forked from filipg/twilight-library
61 lines
2.2 KiB
Haskell
61 lines
2.2 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 = extractLinks "//a[@class='action' and text() = 'View Journal']"
|
|
-- >>> arr ((++"/issue/archive") . init)
|
|
-- >>> extractLinksWithText "//div/h4[contains(@class, 'Vol')]/a"
|
|
|
|
extractRecords = extractLinksWithText "//a[@class='action' and text() = 'View Journal']" -- pary adres-tytuł
|
|
>>> second (arr $ replace "\r\n " " ") -- czyścimy drugi element pary, czyli tytuł z niepotrzebnych białych znaków
|
|
>>> first (arr ((++"/issue/archive") . init)) -- modyfikujemy pierwszy element pary, czyli adres URL
|
|
>>> first (extractLinksWithText "//div/h4[contains(@class, 'Vol')]/a")
|
|
|
|
|
|
-- >>> first (extractLinksWithText "//div[@class='tocTitle']/a")
|
|
-- >>> first (arr ((++"/pdf") . init))
|
|
-- >>> second (arr $ replace "2020")
|
|
|
|
-- 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 title = "Geo Journal OOO" ++ yearlyTitle ++ " AAAAA" ++ (replace "\r\n" "" (replace "\r\n " "" articleTitle))
|
|
date = "202020"
|
|
|
|
|
|
-- 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 = "https://geojournals.pgi.gov.pl"
|
|
|
|
let shadowLibrary = ShadowLibrary {logoUrl=Nothing,
|
|
lname="Polish Geological Journals",
|
|
abbrev="geojour",
|
|
lLevel=0,
|
|
webpage=start}
|
|
extractItemsStartingFromUrl shadowLibrary start (extractRecords >>> arr toShadowItem)
|