Parsing entries

lib/opds/entry.rb

@@ -1,7 +1,87 @@
 module OPDS
 	class Entry
-		def self.from_nokogiri(content)
-			self.new
+		include Logging
+		attr_reader :raw_doc
+		attr_reader :title
+		attr_reader :id
+		attr_reader :updated_at
+		attr_reader :summary
+		attr_reader :authors
+		attr_reader :links
+		attr_reader :dcmetas
+		def self.from_nokogiri(content,namespaces=nil)
+			z=self.new
+			z.instance_variable_set('@raw_doc',content)
+			z.instance_variable_set('@namespaces',namespaces)
+			z.serialize!
+			z
+		end
+
+
+		def serialize!
+			@namespaces=raw_doc.root.namespaces if @namespaces.nil?
+			@authors=[]
+			@raw_doc=raw_doc.at('./xmlns:entry',@namespaces) if raw_doc.at('./xmlns:entry',@namespaces)
+			@title=text(raw_doc.at('./xmlns:title',@namespaces))
+			@id=text(raw_doc.at('./xmlns:id',@namespaces))
+			@summary=text(raw_doc.at('./xmlns:summary',@namespaces))
+			d=text(raw_doc.at('./xmlns:updated',@namespaces))
+			@updated_at=DateTime.parse(d) unless d.nil?
+
+			@authors=raw_doc.xpath('./xmlns:author',@namespaces).collect do |auth|
+				{
+					:name => text(raw_doc.at('./xmlns:author/xmlns:name',@namespaces)),
+					:uri => text(raw_doc.at('./xmlns:author/xmlns:uri',@namespaces)),
+					:email => text(raw_doc.at('./xmlns:author/xmlns:email',@namespaces))
+				}
+			end
+
+			@links=OPDS::Support::LinkSet.new
+			raw_doc.xpath('./xmlns:link',@namespaces).each do |n|
+				text=nil
+				text=n.attributes['title'].value unless n.attributes['title'].nil?
+				link=n.attributes['href'].value
+				type=n.attributes['type'].value unless n.attributes['type'].nil?
+				unless n.attributes['rel'].nil?
+					n.attributes['rel'].value.split.each do |rel|
+						@links.push(rel,link,text,type)
+					end
+				else
+					@links.push(nil,link,text,type)
+				end
+			end
+			@dcmetas=Hash.new
+			prefs=@namespaces.reject{|_,v| !%W[http://purl.org/dc/terms/ http://purl.org/dc/elements/1.1/].include?v}
+			prefs.keys.map{|p| p.split(':').last}.each do |pref|
+				raw_doc.xpath('./'+pref+':*',@namespaces).each do |n|
+					@dcmetas[n.name]=[] unless  @dcmetas[n.name]
+					@dcmetas[n.name].push [n.text, n]
+				end
+			end
+
+		end
+
+
+		def author
+			authors.first
+		end
+
+		def partial?
+			links.by(:rel)['alternate'].any? do |l|
+				l[3]=='application/atom+xml'||l[3]=='application/atom+xml;type=entry'
+			end
+		end
+
+		def complete_url
+			links.by(:rel)['alternate'].any? do |l|
+				l[3]=='application/atom+xml'
+			end unless !partial?
+		end
+
+		protected 
+		def text(t)
+			return t.text unless t.nil?
+			t
 		end
 	end
 end

lib/opds/feed.rb

@@ -48,16 +48,16 @@ def self.from_nokogiri(content)
 		#read xml entries into entry struct
 		def serialize!
 			@entries=raw_doc.xpath('/xmlns:feed/xmlns:entry',raw_doc.root.namespaces).map do |el|
-				OPDS::Entry.from_nokogiri(el)
+				OPDS::Entry.from_nokogiri(el,raw_doc.root.namespaces)
 			end
 		end
 
 		def title
-			raw_doc.at('/xmlns:feed/xmlns:title',raw_doc.root.namespaces).text
+			text(raw_doc.at('/xmlns:feed/xmlns:title',raw_doc.root.namespaces))
 		end
 
 		def icon
-			raw_doc.at('/xmlns:feed/xmlns:icon',raw_doc.root.namespaces).text
+			text(raw_doc.at('/xmlns:feed/xmlns:icon',raw_doc.root.namespaces))
 		end
 
 		def links
@@ -81,14 +81,14 @@ def links
 		end
 
 		def id
-			raw_doc.at('/xmlns:feed/xmlns:id',raw_doc.root.namespaces).text
+			text(raw_doc.at('/xmlns:feed/xmlns:id',raw_doc.root.namespaces))
 		end
 
 		def author
 			{
-			:name => raw_doc.at('/xmlns:feed/xmlns:author/xmlns:name',raw_doc.root.namespaces).text,
-			:uri => raw_doc.at('/xmlns:feed/xmlns:author/xmlns:uri',raw_doc.root.namespaces).text,
-			:email => raw_doc.at('/xmlns:feed/xmlns:author/xmlns:email',raw_doc.root.namespaces).text
+			:name => text(raw_doc.at('/xmlns:feed/xmlns:author/xmlns:name',raw_doc.root.namespaces)),
+			:uri => text(raw_doc.at('/xmlns:feed/xmlns:author/xmlns:uri',raw_doc.root.namespaces)),
+			:email => text(raw_doc.at('/xmlns:feed/xmlns:author/xmlns:email',raw_doc.root.namespaces))
 			}
 		end
 
@@ -121,5 +121,11 @@ def prev_page
 			Feed.parse_url(prev_page_url,@browser)
 		end
 
+		protected
+		def text(t)
+			return t.text unless t.nil?
+			t
+		end
+
 	end
 end

lib/opds/support/linkset.rb

@@ -6,6 +6,7 @@ def initialize
 				@rel_store=Hash.new
 				@txt_store=Hash.new
 				@lnk_store=Hash.new
+				@typ_store=Hash.new
 				@store=[]
 			end
 
@@ -14,10 +15,12 @@ def []=(k,v)
 				i=@store.size-1
 				@rel_store[k]=[] unless @rel_store[k]
 				@rel_store[k].push i
-				@txt_store[v.last]=[] unless @txt_store[v.last]
-				@txt_store[v.last].push i
+				@txt_store[v[1]]=[] unless @txt_store[v[1]]
+				@txt_store[v[1]].push i
 				@lnk_store[v.first]=[] unless @lnk_store[v.first]
 				@lnk_store[v.first].push i
+				@typ_store[v.last]=[] unless @typ_store[v.last]
+				@typ_store[v.last].push i
 
 			end
 
@@ -29,8 +32,8 @@ def each(&block)
 				@store.each(&block)
 			end
 
-			def push(rel,link,text=nil)
-				self[rel]=[link,text]
+			def push(rel,link,text=nil,type=nil)
+				self[rel]=[link,text,type]
 			end
 
 			def link_url(k)
@@ -50,6 +53,12 @@ def link_text(k)
 				t=remap(collection(ty)[v])
 				t.first[2] unless t.nil?
 			end
+
+			def link_type(k)
+				ty,v=k.first
+				t=remap(collection(ty)[v])
+				t.first[3] unless t.nil?
+			end
 
 			def size
 				@store.size
@@ -77,6 +86,7 @@ def collection(type)
 				when 'link' then @lnk_store
 				when 'rel' then @rel_store
 				when 'txt' then @txt_store
+				when 'type' then @typ_store
 				end
 			end
 

spec/entry_spec.rb

@@ -0,0 +1,46 @@
+require File.expand_path(File.dirname(__FILE__) + '/spec_helper')
+
+def sample(type)
+	File.open( File.expand_path(File.dirname(__FILE__) + "/../samples/#{type}.txt"))
+end
+
+describe OPDS::Entry do
+
+
+	subject do 
+		 OPDS::Entry.from_nokogiri(Nokogiri::XML(sample(:entry)))
+	end
+
+	it "should have a title "do
+			subject.title.size.should_not be(0)
+	end
+
+	it "should have an id"do
+			subject.id.size.should_not be(0)
+	end
+
+	it "should have a summary "do
+			subject.summary.size.should_not be(0)
+	end
+
+	it "should have a update date "do
+			subject.updated_at.day.should be(13)
+	end
+
+	it "should have an author "do
+			subject.author.size.should_not be(0)
+	end
+
+	it "should have  links "do
+			subject.links.size.should_not be(0)
+	end
+
+	it "should have dc:meta" do
+
+			subject.dcmetas.size.should_not be(0)
+	end
+
+	it "should not be partial" do
+			subject.should_not be_partial()
+	end
+end

spec/linkset_spec.rb

@@ -2,9 +2,9 @@
 
 describe OPDS::Support::LinkSet do 
 	before(:each) do 
-		subject.push('root','http://feedbooks.com','Racine')
-		subject.push('subsection','http://feedbooks.com/publicdomain','Domaine pub')
-		subject.push('subsection','http://feedbooks.com/original','Original')
+		subject.push('root','http://feedbooks.com','Racine','application/atom+xml')
+		subject.push('subsection','http://feedbooks.com/publicdomain','Domaine pub','application/atom+xml')
+		subject.push('subsection','http://feedbooks.com/original','Original','application/atom+xml')
 		subject.push('subsection','http://feedbooks.com/feed','feeds')
 		subject.push('http://opds-spec.org/shelf','http://feedbooks.com/shelf','shelf')
 		subject.push('related','http://feedbooks.com/shelf',nil)

spec/opdsparser_spec.rb

@@ -47,6 +47,7 @@ def sample(type)
 		it "should have entries" do 
 			subject.parse(sample(feed_type)).entries.size.should_not be(0)
 		end
+
 
 
 	end
@@ -59,6 +60,10 @@ def sample(type)
 			subject.parse(sample(:acquisition)).should be_first_page()
 		end
 
+		it "should have partial entries" do 
+			subject.parse(sample(:acquisition)).entries.any?(&:partial?).should be()
+		end
+
 		it do
 			feed=nil
 			lambda { feed=subject.parse(sample(:acquisition)).next_page  }.should_not raise_error