Make script work for 'nasze popowice' pdfs

This commit is contained in:
Jakub Kolasiński 2021-03-17 21:07:39 +01:00
parent 4ae6066ccf
commit 8ba5a297b0

View File

@ -4,7 +4,6 @@ import ShadowLibrary.Core
import Text.XML.HXT.Core import Text.XML.HXT.Core
import Text.XML.HXT.XPath import Text.XML.HXT.XPath
-- import Text.XML.HXT.Curl
import Data.List import Data.List
import Data.List.Utils (replace) import Data.List.Utils (replace)
@ -12,35 +11,38 @@ import Text.Regex.Posix
import Text.Printf import Text.Printf
extractRecords = extractLinksWithText "//a[@class='attachmentContainer']" -- pary adres-tytuł extractRecords = extractLinksWithText "//a[contains(@href,'.pdf') and contains(text(),'popowice')]"
>>> second (arr $ replace "\r\n " " ") -- czyścimy drugi element pary, czyli tytuł z niepotrzebnych białych znaków
>>> first (arr ((++"tr") . init)) -- modyfikujemy pierwszy element pary, czyli adres URL
>>> first (extractLinksWithText "//li/a[contains(@href,'.pdf')]") -- pobieramy stronę z adresu URL i wyciągamy linki z tej strony pasujące do wyrażenia XPathowego
-- ostatecznie wyjdą trójki ((adres URL, tytuł artykułu), tytuł rocznika)
-- ... a tutaj te trójki przerabiamy do docelowej struktury ShadowItem toShadowItem :: (String, String) -> ShadowItem
toShadowItem :: ((String, String), String) -> ShadowItem toShadowItem (url, articleTitle) =
toShadowItem ((url, articleTitle), yearlyTitle) =
(defaultShadowItem url title) { (defaultShadowItem url title) {
originalDate = Just date, originalDate = Just date,
itype = "periodical", itype = "periodical",
format = Just "pdf", format = Just "pdf",
finalUrl = url finalUrl = url
} }
where title = "Almanach Muszyny " ++ yearlyTitle ++ " " ++ (replace "\r\n" "" (replace "\r\n " "" articleTitle)) where title = "Nasze Popowice " ++ getNr url
date = getDate 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 = getDate url =
case url =~~ "/(19[0-9][0-9]|20[0-9][0-9])/" :: Maybe [[String]] of case url =~~ "20[0-9][0-9]_[0-1][0-9]" :: Maybe String of
Just [[_, year]] -> year Just date -> replaceSeparatorWithDot date
otherwise -> error $ "unexpected url: " ++ url otherwise -> error $ "unexpected url: " ++ url
main = do main = do
let start = "https://smpopowice.pl/index.php/numery-archiwalne" let start = "https://smpopowice.pl/index.php/numery-archiwalne"
let shadowLibrary = ShadowLibrary {logoUrl=Nothing, let shadowLibrary = ShadowLibrary {logoUrl=Nothing,
lname="Almanach Muszyny", lname="Nasze Popowice",
abbrev="AlmMusz", abbrev="NaszPop",
lLevel=0, lLevel=0,
webpage=start} webpage=start}
extractItemsStartingFromUrl shadowLibrary start (extractRecords >>> arr toShadowItem) extractItemsStartingFromUrl shadowLibrary start (extractRecords >>> arr toShadowItem)