Biblioteka sejmowa crawler

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
@@ -64,8 +64,7 @@ downloadDocument = readFromDocument [withParseHTML yes,
                                      withEncodingErrors no,
                                      withPreserveComment yes,
                                      withStrictInput yes,
-                                     withHTTP []
+                                     withCurl [("curl--user-agent","AMU Digital Libraries Indexing Agent")]
---                                     withCurl [("curl--user-agent","AMU Digital Libraries Indexing Agent")]
                                     ]
 
 downloadDocumentWithEncoding enc = readFromDocument [withParseHTML yes,
@@ -73,8 +72,7 @@ downloadDocumentWithEncoding enc = readFromDocument [withParseHTML yes,
                                                      withEncodingErrors no,
                                                      withPreserveComment yes,
                                                      withInputEncoding enc,
-					             withHTTP []]		
+                                                     withCurl []]
---                                                     withCurl []]
 
 downloadXmlDocument = readFromDocument [withWarnings no,
                                         withEncodingErrors no,

app/bibliotekasejmowa.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
+
+extractAllPDFs = extractLinksWithText "/html/body/table[4]/tbody/tr[7]/td[2]/a"
+
+extractRecords = extractLinksWithText "//table[3]//a"
+                 >>> first (extractLinksWithText "//table[3]//a"
+                           >>> first (extractLinksWithText "//table[4]//a"
+                                     >>> first (extractAllPDFs
+                                               >>> first(arr $ (drop 23 ) . init . init . init))))
+
+toShadowItem :: ((((String, String), String), String), String) -> ShadowItem
+toShadowItem ((((url, typeName), articleTitle), _), category) =
+  (defaultShadowItem url title) {
+    originalDate = Just date,
+    itype = "periodical",
+    format = Just "pdf",
+    finalUrl = url
+    }
+  where title = "Biblioteka Sejmowa " ++ category ++ " " ++ typeName ++ " " ++ articleTitle
+        date = getDate articleTitle
+
+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://bs.sejm.gov.pl/F/?con_lng=POL&func=file&file_name=find-nowe&local_base=ars01"
+    let shadowLibrary = ShadowLibrary {logoUrl=Nothing,
+                                       lname="Biblioteka Sejmowa",
+                                       abbrev="BiblSejm",
+                                       lLevel=0,
+                                       webpage=start}
+    extractItemsStartingFromUrl shadowLibrary start (extractRecords >>> arr toShadowItem)

shadow-library.cabal

@@ -2,11 +2,11 @@ name:                shadow-library
 version:             0.1.0.0
 synopsis:            Initial project template from stack
 description:         Please see README.md
-homepage:            http://github.com/name/project
+homepage:            https://git.wmi.amu.edu.pl/s444409/twilight-library
 license:             Proprietary
 license-file:        LICENSE
-author:              Your name here
+author:              Marcin Kostrzewski
-maintainer:          your.address@example.com
+maintainer:          markos34@st.amu.edu.pl
 -- copyright:
 category:            Web
 build-type:          Simple
@@ -21,6 +21,7 @@ library
                      , hxt
                      , hxt-http
                      , hxt-xpath
+                     , hxt-curl
                      , MissingH
                      , monad-logger
                      , mtl
@@ -54,12 +55,26 @@ executable almanachmuszyny
    build-depends:       base
                       , hxt
                       , hxt-xpath
+                      , hxt-curl
                       , MissingH
                       , regex-posix
                       , shadow-library
    default-language:    Haskell2010
 
+executable bibliotekasejmowa
+  hs-source-dirs:      app
+  main-is:             bibliotekasejmowa.hs
+  ghc-options:         -threaded -rtsopts -with-rtsopts=-N
+  build-depends:       base
+                     , hxt
+                     , hxt-xpath
+                     , hxt-curl
+                     , MissingH
+                     , regex-posix
+                     , shadow-library
+  default-language:    Haskell2010
+
 
 source-repository head
   type:     git
-  location: https://github.com/name/project
+  location: https://git.wmi.amu.edu.pl/s444409/twilight-library

stack.yaml.lock

@@ -0,0 +1,19 @@
+# This file was autogenerated by Stack.
+# You should not edit this file by hand.
+# For more information, please see the documentation at:
+#   https://docs.haskellstack.org/en/stable/lock_files
+
+packages:
+- completed:
+    hackage: hxt-xpath-9.1.2.2@sha256:9cd590ae93a04573db8f90fa4094625ebd97dded45da7667c577ce6b38a42900,1999
+    pantry-tree:
+      size: 2225
+      sha256: aee2f75974e868ff429b8ff349a29667536c60397098f5dfedc968d1951511bb
+  original:
+    hackage: hxt-xpath-9.1.2.2
+snapshots:
+- completed:
+    size: 507596
+    url: https://raw.githubusercontent.com/commercialhaskell/stackage-snapshots/master/lts/11/9.yaml
+    sha256: 42f472dbf06482da1b3319241f3e3b3593a45bd7d4f537d2789f21386b9b2ad3
+  original: lts-11.9