dq-readability.rb in dq-readability-1.0.3

- old
+ new

@@ -94,11 +94,15 @@
 
 	  # changing img src
       @html.css("img").each do |elem|
         begin
 		  if elem['src'][0] == '/' 
-			elem['src'] = URI.join(base,elem['src']).to_s if URI.parse(elem['src']).host == nil 
+			if elem['src'][1] == '/'
+				elem['src'] = 'http:'+elem['src']
+			else
+				elem['src'] = URI.join(base,elem['src']).to_s if URI.parse(elem['src']).host == nil 
+			end
 		  else
 			if @url.split('').last == '/'
 				elem['src'] = URI.join(@url,elem['src']).to_s if URI.parse(elem['src']).host == nil
 			else
 				x = @url.split('/')
@@ -111,12 +115,21 @@
         rescue
           elem.remove
         end
       end
 
-      #changing the 'a' href
+	# changing certain tags to <p> tags
+	
+	x = @html.css("ol")
+	x.each do |t|
+		t.name = "p"
+	end
+	len = @html.css('ol').length
+	debug("length of ol tag #{len}")
 
+     #changing the 'a' href
+
 	 @html.css("a").each do |elem|
         begin
 		  if elem['href'][0] == '/' 
 			elem['href'] = URI.join(base,elem['href']).to_s if URI.parse(elem['href']).host == nil 
 		  else
@@ -523,11 +536,11 @@
       s = Nokogiri::XML::Node::SaveOptions
       save_opts = s::NO_DECLARATION | s::NO_EMPTY_TAGS | s::AS_XHTML
       html = node.serialize(:save_with => save_opts)
 
       # Get rid of duplicate whitespace
-      return html.gsub(/[\r\n\f]+/, "\n" )
+      return "<head><meta http-equiv='Content-Type' content='text/html; charset=utf-8'></head>" + "\n" + html.gsub(/[\r\n\f]+/, "\n" )
     end
 
     def clean_conditionally(node, candidates, selector)
       return unless @clean_conditionally
       node.css(selector).each do |el|
@@ -548,13 +561,13 @@
           content_length = el.text.strip.length  # Count the text length excluding any surrounding whitespace
           link_density = get_link_density(el)
           to_remove = false
           reason = ""
           
-          if (counts["img"] > counts["p"]) && (counts["img"] > 1)
+          if (counts["img"] > counts["p"]+2)
             reason = "too many images"
             to_remove = true
-          elsif counts["li"] > counts["p"] && name != "ul" && name != "ol"
+        elsif counts["li"] > counts["p"] && name != "ul" && name != "ol"
             reason = "more <li>s than <p>s"
             to_remove = true
 		elsif counts["input"] > (counts["p"] / 3).to_i
             reason = "less than 3x <p>s than <input>s"
            to_remove = true