czottmann · tony2nite · Jul 12, 2011 · Jul 12, 2011 · Jul 12, 2011 · Jul 12, 2012
diff --git a/Classes/article.rb b/Classes/article.rb
@@ -8,13 +8,14 @@
 
 require "constants"
 require "data_fetcher"
+require "cgi"
 
 
 module Instapaper
 
   class Article < DataFetcher
 
-    attr_accessor :ip_id, :title, :site, :status, :html
+      attr_accessor :ip_id, :title, :site, :status, :html, :url
 
 
     def initialize(id, title = "")
@@ -48,11 +49,12 @@ def fetch
       log("fetch")
 
       @loading = true
-      init_fetch
+      connection = init_fetch
     end
 
 
     def connection(connection, didFailWithError: error)
+      log("connection:didFailWithError: " + @ip_id.to_s)
       if super
         @status = { :error => error.localizedDescription }
         @html   = ""
@@ -63,65 +65,79 @@ def connection(connection, didFailWithError: error)
 
     def connectionDidFinishLoading(connection)
       super
+      log("connection:connectionDidFinishLoading: " + @ip_id.to_s)
 
-      process_raw
+      error = Pointer.new_with_type("@")
+      doc = NSXMLDocument.alloc.initWithData(@data, options: NSXMLDocumentTidyHTML, error: error)
+
+      if NSString.alloc.initWithData(@data, encoding: NSUTF8StringEncoding).empty?
+        @html = ""
+      else
+        @html = process_raw doc, error
+      end
+
       notify("article.fetched")
     end
 
 
-  private
+
 
     # Wandelt die rohen Daten in HTML um, räumt das HTML auf, erzeugt einige
     # weitere Attribute (+@site+ etc.).
 
-    def process_raw
-      log("process_raw")
-      error = Pointer.new_with_type("@")
-      @doc = NSXMLDocument.alloc.initWithData(@data, options: NSXMLDocumentTidyHTML, error: error)
+    def process_raw(doc, error)
+      result = NSXMLDocument.alloc.initWithXMLString("<html><body></body></html>", options: NSXMLDocumentTidyHTML, error: error)
 
-      if NSString.alloc.initWithData(@data, encoding: NSUTF8StringEncoding).empty?
-        @html = ""
-        return
-       end
-
-      # Titel etc. extrahieren
-      @site = get_xpath_value_from_doc("//div[@class='sm']")
+      body = result.nodesForXPath("//body", error: error).first
 
-      # Unnützes Zeug löschen
+      # Titel etc. extrahieren
+      @site = get_xpath_value_from_doc(doc, "//*[@id='titlebar']/span/a")
+      @title = get_xpath_value_from_doc(doc, "//h1")
+
+      # Unnützes Zeug löschen - strip out redundant stuff
       xpath_selectors = [
         "//style",
         "//link",
         "//meta",
+        "//script",
         "//noscript",
-        "//@*[local-name()='style']",
-        "//div[contains(@class, 'top')]",
-        "//div[contains(@class, 'bar')][1]"
+		"//img"
       ].join("|")
 
-      @doc.nodesForXPath(xpath_selectors, error: error).each { |n| n.detach }
-
-      adjust_bottom_bar
-      adjust_blockquotes
+      doc.nodesForXPath(xpath_selectors, error: error).each { |n| n.detach }
+
+        #adjust_blockquotes doc
 
-      @html = NSMutableString.alloc.initWithData(@doc.XMLData, encoding: NSUTF8StringEncoding)
-    end
+      if doc.nodesForXPath("//h1", error: error).first.nil?
+          log('process_raw:unexpected title format for ' + @title + "/" + @url)
+      else
+          title = doc.nodesForXPath("//h1", error: error).first.detach
+          body.addChild(title)
+      end
+
+      if doc.nodesForXPath("//div[@id='story']", error: error).first.nil?
+          log('process_raw:unexpected story format for ' + @url)
+      else
+          story = doc.nodesForXPath("//div[@id='story']", error: error).first.detach
+          body.addChild(story)
+
+          add_bottom_bar body
+      end
 
+      NSMutableString.alloc.initWithData(result.XMLData, encoding: NSUTF8StringEncoding)
+    end
 
-    def get_xpath_value_from_doc(xpath_selector)
+  private
+    def get_xpath_value_from_doc(doc, xpath_selector)
       error = Pointer.new_with_type("@")
-      value = @doc.nodesForXPath(xpath_selector, error: error).first.objectValue rescue ""
+      value = doc.nodesForXPath(xpath_selector, error: error).first.objectValue rescue ""
+      value = CGI.unescapeHTML(value)
+      value = value.gsub('’','\'')
       normalize_string(value)
     end
 
 
-    def adjust_bottom_bar
-      error = Pointer.new_with_type("@")
-      bb = @doc.nodesForXPath("//div[contains(@class, 'bottom')] | //div[contains(@class, 'bar')]", error: error).first
-
-      return if bb.nil?
-
-      bb.setChildren(nil)
-
+    def add_bottom_bar(doc)
       tags = [
         ["hr"],
         ["p", "You're reading an Instapaper article which was copied to your reader by Ephemera, the Mac tool for IP enthusiasts. Delete this article on your reading device, and during the next sync it'll be archived on Instapaper.com."],
@@ -138,14 +154,14 @@ def adjust_bottom_bar
           node = NSXMLNode.elementWithName(n)
         end
 
-        bb.insertChild(node, atIndex: i)
+        doc.addChild(node)
       end
     end
 
 
-    def adjust_blockquotes
+    def adjust_blockquotes doc
       error = Pointer.new_with_type("@")
-      @doc.nodesForXPath("//blockquote/p", error: error).each do |node|
+      doc.nodesForXPath("//blockquote/p", error: error).each do |node|
         node.setName("div")
       end
     end
@@ -154,8 +170,6 @@ def adjust_blockquotes
     def normalize_string(string)
       string.gsub(/<[^>]*>/, " ").squeeze(" ").strip
     end
-
   end
-
 end
 
diff --git a/Classes/data_fetcher.rb b/Classes/data_fetcher.rb
@@ -6,9 +6,9 @@
 #  Copyright (c) 2010 Carlo Zottmann. All rights reserved.
 #
 
+require "bootstrap"
 require "notifiable"
 
-
 module Instapaper
 
   class DataFetcher
@@ -41,7 +41,6 @@ def init_fetch
         cachePolicy: NSURLRequestReloadIgnoringCacheData,
         timeoutInterval: @timeout
       )
-
       @connection = NSURLConnection.connectionWithRequest(@request, delegate: self)
     end
 
@@ -53,6 +52,7 @@ def connection(connection, didReceiveResponse: response)
 
 
     def connection(connection, didReceiveData:data)
+      log("connection:didReceiveData")
       # NSLog("#{self.class}#didReceiveData")
       @data.appendData(data)
     end

diff --git a/Classes/unread_articles_list.rb b/Classes/unread_articles_list.rb
@@ -86,7 +86,7 @@ def extract_articles_and_bundles
 
         links.each do |l|
           node  = NSXMLDocument.alloc.initWithXMLString( l.XMLString, options: NSXMLDocumentTidyXML, error: error )
-          first_subnode = node.nodesForXPath( "//a[@class='actionLink']/@href", error: error ).first
+          first_subnode = node.nodesForXPath( "//a[@class='actionLink' and @title='Edit']/@href", error: error ).first
 
           next if first_subnode.nil?
 

diff --git a/Classes/window_controller.rb b/Classes/window_controller.rb
@@ -199,7 +199,7 @@ def singlearticlesprocessor_status(notification)
         article = a[:article]
         msg = article.site + ": " + article.title
         holler(msg, "→")
-        log("singlearticlesprocessor_status", "- #{msg}")
+        log("singlearticlesprocessor_status", "- " + a[:id].to_s + ": " + msg)
       end
 
       log("singlearticlesprocessor_status", "processing unread articles")