title extraction

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 +64,8 @@ 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,

app/almanachmuszyny.hs

@@ -4,18 +4,21 @@ 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)
 
 import Text.Regex.Posix
 import Text.Printf
 
-extractRecords = extractLinksWithText "//a[@class='roczniki']"
-                 >>> second (arr $ replace "\r\n              " "")
-                 >>> first (arr ((++"tr") . init))
-                 >>> first (extractLinksWithText "//li/a[contains(@href,'.pdf')]")
 
+extractRecords = extractLinksWithText "//a[@class='roczniki']"  -- 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) {

app/teatrLalek.hs

@@ -0,0 +1,39 @@
+
+{-# 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 = extractLinks "//a[contains(@href, '.pdf')]"
+
+
+toShadowItem :: String -> ShadowItem
+toShadowItem url =
+  (defaultShadowItem url title) {
+    originalDate = Just date,
+    itype = "periodical",
+    format = Just "pdf",
+    finalUrl = url
+    }
+  where
+    date = last $ getAllTextMatches $ url =~ "(19[0-9][0-9]|20[0-9][0-9])" :: String
+    titleToProcess = last $ getAllTextMatches $ url =~ "/[a-zA-Z ]+[-_]" :: String
+    title = titleToProcess =~ "[a-zA-Z ]+" :: String
+
+
+main = do
+    let start = "http://polunima.pl/teatr-lalek/"
+    let shadowLibrary = ShadowLibrary {logoUrl=Nothing,
+                                       lname="Teatr Lalek",
+                                       abbrev="Teatr",
+                                       lLevel=0,
+                                       webpage=start}
+    extractItemsStartingFromUrl shadowLibrary start (extractRecords >>> arr toShadowItem)

shadow-library.cabal

@@ -20,7 +20,6 @@ library
                      , HTTP
                      , hxt
                      , hxt-http
-                     , hxt-curl
                      , hxt-xpath
                      , MissingH
                      , monad-logger
@@ -35,18 +34,19 @@ library
                      , tz
   default-language:    Haskell2010
 
--- executable maly-modelarz-exe
---    hs-source-dirs:      app
---    main-is:             malymodelarz.hs
---    ghc-options:         -threaded -rtsopts -with-rtsopts=-N
---    build-depends:       base
---                       , hxt
---                       , hxt-curl
---                       , hxt-xpath
---                       , MissingH
---                       , regex-posix
---                       , shadow-library
---    default-language:    Haskell2010
+executable teatrLalek
+    hs-source-dirs:      app
+    main-is:             teatrLalek.hs
+    ghc-options:         -threaded -rtsopts -with-rtsopts=-N
+    build-depends:       base
+                       , hxt
+                       , hxt-curl
+                       , hxt-xpath
+                       , MissingH
+                       , regex-posix
+                       , shadow-library
+                       , regex-tdfa
+    default-language:    Haskell2010
 
 executable almanachmuszyny
    hs-source-dirs:      app
@@ -55,7 +55,6 @@ executable almanachmuszyny
    build-depends:       base
                       , hxt
                       , hxt-xpath
-                      , hxt-curl
                       , MissingH
                       , regex-posix
                       , shadow-library