diff --git a/ShadowLibrary/Core.hs b/ShadowLibrary/Core.hs index 678df33..cdcf1cc 100644 --- a/ShadowLibrary/Core.hs +++ b/ShadowLibrary/Core.hs @@ -33,7 +33,7 @@ import Data.Tree.NTree.TypeDefs import Data.Maybe import Control.Monad.Trans import Text.XML.HXT.XPath --- import Text.XML.HXT.Curl +import Text.XML.HXT.Curl import Text.XML.HXT.HTTP import Text.Regex.TDFA @@ -42,7 +42,7 @@ import Data.List (isInfixOf, intercalate) import Data.List.Utils (replace) --- import Network.Curl.Opts +--import Network.Curl.Opts polishTimeZone = TimeZone { timeZoneMinutes = 120, @@ -64,8 +64,8 @@ downloadDocument = readFromDocument [withParseHTML yes, withEncodingErrors no, withPreserveComment yes, withStrictInput yes, - withHTTP [] --- withCurl [("curl--user-agent","AMU Digital Libraries Indexing Agent")] +-- withHTTP [] + withCurl [("curl--user-agent","AMU Digital Libraries Indexing Agent")] ] downloadDocumentWithEncoding enc = readFromDocument [withParseHTML yes, @@ -73,13 +73,13 @@ downloadDocumentWithEncoding enc = readFromDocument [withParseHTML yes, withEncodingErrors no, withPreserveComment yes, withInputEncoding enc, - withHTTP []] --- withCurl []] +-- withHTTP []] + withCurl []] downloadXmlDocument = readFromDocument [withWarnings no, withEncodingErrors no, - withHTTP []] --- withCurl [] ] +-- withHTTP []] + withCurl [] ] data ShadowLibrary = ShadowLibrary { logoUrl :: Maybe String, diff --git a/app/pbsociety.hs b/app/pbsociety.hs new file mode 100644 index 0000000..4891207 --- /dev/null +++ b/app/pbsociety.hs @@ -0,0 +1,46 @@ + +{-# 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" -- pary adres-tytuł + -- >>> 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), String) -> ShadowItem +toShadowItem ((url, articleTitle), yearlyTitle) = + (defaultShadowItem url title) { + originalDate = Just date, + itype = "periodical", + format = Just "pdf", + finalUrl = url + } + where title = "Almanach Muszyny " ++ yearlyTitle ++ " " ++ (replace "\r\n" "" (replace "\r\n " "" articleTitle)) + date = getDate url + +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://pbsociety.org.pl/repository/" + let shadowLibrary = ShadowLibrary {logoUrl=Nothing, + lname="Polskie Towarzystwo Botaniczne", + abbrev="PBSociety", + lLevel=0, + webpage=start} + extractItemsStartingFromUrl shadowLibrary start extractRecords diff --git a/shadow-library.cabal b/shadow-library.cabal index cd77195..4090e2c 100644 --- a/shadow-library.cabal +++ b/shadow-library.cabal @@ -19,6 +19,7 @@ library build-depends: base >= 4.7 && < 5 , HTTP , hxt + , hxt-curl , hxt-http , hxt-xpath , MissingH @@ -59,6 +60,17 @@ executable almanachmuszyny , shadow-library default-language: Haskell2010 +executable pbsociety + hs-source-dirs: app + main-is: pbsociety.hs + ghc-options: -threaded -rtsopts -with-rtsopts=-N + build-depends: base + , hxt + , hxt-xpath + , MissingH + , regex-posix + , shadow-library + default-language: Haskell2010 source-repository head type: git