twilight-library/app/czasopisma_francja.hs
2021-03-20 12:06:27 +01:00

50 lines
1.6 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') and contains(text(),'czasopisma_francja')]"
toShadowItem :: (String, String) -> ShadowItem
toShadowItem (url, articleTitle) =
(defaultShadowItem url title) {
originalDate = Just date,
itype = "periodical",
format = Just "pdf",
finalUrl = url
}
where title = "czasopisma_francja " ++ getNr url
date = getDate url
replaceSeparatorWithSpace = map (\c -> if c=='_' then ' '; else if c=='-' then ' '; else c)
replaceSeparatorWithDot = map (\c -> if c=='_' then '.'; else c)
getNr url =
case url =~~ "nr[-_][0-9][0-9]" :: Maybe String of
Just nr -> replaceSeparatorWithSpace nr
otherwise -> error $ "unexpected url: " ++ url
getDate url =
case url =~~ "20[0-9][0-9]_[0-1][0-9]" :: Maybe String of
Just date -> replaceSeparatorWithDot date
otherwise -> error $ "unexpected url: " ++ url
main = do
let start = "http://www.almanachmuszyny.pl/"
let shadowLibrary = ShadowLibrary {logoUrl=Nothing,
lname="czasopisma_francja",
abbrev="CzasFr",
lLevel=0,
webpage=start}
extractItemsStartingFromUrl shadowLibrary start (extractRecords >>> arr toShadowItem)