53 lines
1.7 KiB
Haskell
53 lines
1.7 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 "//ul[@class='menu-topmenu menu-iconmenu']//li[@class='first' or @class='last' or not(@class)]//a"
|
|
>>> second (arr $ replace "\n\t \n\t\t" "")
|
|
>>> second (arr $ replace "\t \n " "")
|
|
>>> first (extractLinksWithText "//div[@class='jsn-article-content']//a[contains(.,'nr')]")
|
|
>>> first (second ( arr $ replace "ś" "s"))
|
|
>>> first (second ( arr $ replace "ń" "n"))
|
|
>>> first (second ( arr $ replace "ź" "z"))
|
|
|
|
toShadowItem :: ((String, String), String) -> ShadowItem
|
|
toShadowItem ((url, articleTitle), yearlyTitle) =
|
|
(defaultShadowItem url title) {
|
|
originalDate = extractDate articleTitle,
|
|
itype = "periodical",
|
|
format = Just "pdf",
|
|
finalUrl = url
|
|
}
|
|
where title = articleTitle
|
|
|
|
extractDate :: String -> Maybe String
|
|
extractDate url =
|
|
case url =~~ ("[A-Za-z]+ [0-9]{4}" :: String) of
|
|
Just date -> Just date
|
|
otherwise -> extractOtherDate url
|
|
|
|
extractOtherDate :: String -> Maybe String
|
|
extractOtherDate url =
|
|
case url =~~ ("[0-9]{4}_[0-9]{2}" :: String) of
|
|
Just date -> Just (replace "_" "-" date)
|
|
otherwise -> Just url
|
|
|
|
|
|
main = do
|
|
let start = "https://www.smpopowice.pl/"
|
|
let shadowLibrary = ShadowLibrary {logoUrl=Nothing,
|
|
lname="Nasze Popowice",
|
|
abbrev="NaszPop",
|
|
lLevel=0,
|
|
webpage=start}
|
|
extractItemsStartingFromUrl shadowLibrary start (extractRecords >>> arr toShadowItem)
|