Compare commits

...

5 Commits

4 changed files with 66 additions and 8 deletions

View File

@ -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
@ -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,

View File

@ -4,7 +4,7 @@ import ShadowLibrary.Core
import Text.XML.HXT.Core
import Text.XML.HXT.XPath
-- import Text.XML.HXT.Curl
--import Text.XML.HXT.Curl
import Data.List
import Data.List.Utils (replace)

View File

@ -0,0 +1,43 @@
{-# 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[@class='plik']" -- pary adres-tytuł
>>> second (arr $ replace "\r\n " " ") -- czyścimy drugi element pary, czyli tytuł z niepotrzebnych białych znaków
-- ostatecznie wyjdą dwójki ((adres URL, tytuł artykułu)
-- ... a tutaj te trójki przerabiamy do docelowej struktury ShadowItem
toShadowItem :: (String, String) -> ShadowItem
toShadowItem (url, articleTitle) =
(defaultShadowItem url title) {
originalDate = Just date,
itype = "periodical",
format = Just "pdf",
finalUrl = url
}
where title = articleTitle
date = getDate articleTitle
getDate articleTitle =
case articleTitle =~~ "(19[0-9][0-9]|20[0-9][0-9])" :: Maybe [[String]] of
Just [[_, year]] -> year
otherwise -> "No date for " ++ articleTitle
main = do
let start = "https://www.pipc.org.pl/publikacje/biuletyn-bezpieczna-chemia"
let shadowLibrary = ShadowLibrary {logoUrl=Nothing,
lname="Biuletyn Bezpieczna Chemia - Publikacje - PIPC",
lLevel=0,
webpage=start}
extractItemsStartingFromUrl shadowLibrary start (extractRecords >>> arr toShadowItem)

View File

@ -20,6 +20,7 @@ library
, HTTP
, hxt
, hxt-http
, hxt-curl
, hxt-xpath
, MissingH
, monad-logger
@ -54,6 +55,20 @@ executable almanachmuszyny
build-depends: base
, hxt
, hxt-xpath
, hxt-curl
, MissingH
, regex-posix
, shadow-library
default-language: Haskell2010
executable biuletynbezpiecznachemia
hs-source-dirs: app
main-is: biuletynbezpiecznachemia.hs
ghc-options: -threaded -rtsopts -with-rtsopts=-N
build-depends: base
, hxt
, hxt-xpath
, hxt-curl
, MissingH
, regex-posix
, shadow-library