lib/gutenberg_rdf/rdf.rb in gutenberg_rdf-0.3.1 vs lib/gutenberg_rdf/rdf.rb in gutenberg_rdf-0.4.0

@@ -46,25 +46,26 @@
     def publisher
       xml.elements['pgterms:ebook/dcterms:publisher'].text
     end
 
     def language
-      xml.elements['pgterms:ebook/dcterms:language'].text
+      xml.elements['pgterms:ebook/dcterms:language/rdf:Description/rdf:value'].text
     end
 
     def rights
       xml.elements['pgterms:ebook/dcterms:rights'].text
     end
 
     def covers
-      official_cover_images.concat(other_cover_images).sort.uniq
+      official_cover_images.concat(other_cover_images).uniq
     end
 
     def ebooks
       files = Array.new
-      xml.elements.each('pgterms:file') do |file|
-        files << Media.new(file)
+      xml.elements.each('pgterms:ebook/dcterms:hasFormat') do |format|
+        file = format.elements['pgterms:file']
+        files << Media.new(file) if file.elements['dcterms:format/rdf:Description/rdf:value'].text.match(/\Atext|\Aapplication/)
       end
       files
     end
 
   private
@@ -83,41 +84,32 @@
       title_array = title_array.first.split(/, or,/) if title_array.count == 1
 
       title_array.map(&:strip)
     end
 
-    def roles
-      @roles ||= extract_roles
-    end
-
-    def extract_roles
-      entries = Hash.new
-      xml.elements.each('pgterms:ebook/dcterms:creator') do |entry|
-        entries["#{entry.attributes['rdf:resource'].sub('2009/agents/', '')}"] = 'aut'
+    def extract_authors
+      agents = Array.new
+      xml.elements.each('pgterms:ebook/dcterms:creator') do |contributor|
+        agent = Agent.new(contributor.elements['pgterms:agent'])
+        agent.role = 'aut'
+        agents << agent
       end
-      xml.elements.each('pgterms:ebook/marcrel:*') do |entry|
-        entries["#{entry.attributes['rdf:resource'].sub('2009/agents/', '')}"] = entry.name
+      xml.elements.each('pgterms:ebook/marcrel:*') do |contributor|
+        agent = Agent.new(contributor.elements['pgterms:agent'])
+        agent.role = contributor.name
+        agents << agent
       end
-      entries
+      agents
     end
 
-    def extract_authors
-      entries = Array.new
-      xml.elements.each('pgterms:agent') do |agent|
-        entry = Agent.new(agent)
-        entry.assign_role(roles)
-        entries << entry
-      end
-      entries
-    end
-
     def official_cover_images
       entries = Array.new
-      xml.elements.each('pgterms:file') do |file|
+      xml.elements.each('pgterms:ebook/dcterms:hasFormat') do |format|
+        file = format.elements['pgterms:file']
         entries << file.attributes['about'] if file_is_image?(file)
       end
-      entries
+      entries.sort
     end
 
     def file_is_image?(node)
       node.elements.each('dcterms:format/rdf:Description/rdf:value') do |value|
         return true if value.text.match(/image/)
@@ -130,10 +122,10 @@
       xml.elements.each('pgterms:ebook/pgterms:marc901') do |node|
         cover = node.text
         cover.sub!(/\Afile:\/\/\/public\/vhost\/g\/gutenberg\/html/, 'http://www.gutenberg.org')
         entries << cover
       end
-      entries
+      entries.sort
     end
 
   end
 end