abhay-calais 0.0.7 → 0.0.8

data/README.markdown

@@ -32,7 +32,7 @@ This will return an object containing information extracted from the RDF respons
 ## Requirements ##
 
 * [Ruby 1.8.5 or better](http://ruby-lang.org)
-* [libxml-ruby](http://libxml.rubyforge.org/), [libxml2](http://xmlsoft.org/)
+* [nokogiri](http://nokogiri.rubyforge.org/nokogiri/), [libxml2](http://xmlsoft.org/), [libxslt](http://xmlsoft.org/xslt/)
 * [curb](http://curb.rubyforge.org/), [libcurl](http://curl.haxx.se/)
 * [json](http://json.rubyforge.org/)
 
@@ -46,4 +46,4 @@ You can install the Calais gem via Rubygems (`gem install calais`) or by buildin
 
 ## Acknowledgements ##
 
-* [Paul Legato](http://www.economaton.com/): Help all around with the new response processor and implementation of the 3.1 API.
+* [Paul Legato](http://www.economaton.com/): Help all around with the new response processor and implementation of the 3.1 API.

data/Rakefile

@@ -18,7 +18,7 @@ begin
     s.authors = ['Abhay Kumar']
     s.files = FileList["[A-Z]*", "{bin,generators,lib,test}/**/*"]
     s.rubyforge_project = 'calais'
-    s.add_dependency 'libxml-ruby', '>= 0.5.4'
+    s.add_dependency 'nokogiri', '>= 1.3.3'
     s.add_dependency 'json', '>= 1.1.3'
     s.add_dependency 'curb', '>= 0.1.4'
   end

data/VERSION.yml

@@ -1,4 +1,4 @@
 --- 
-:patch: 7
-:major: 0
 :minor: 0
+:patch: 8
+:major: 0

data/lib/calais.rb

@@ -5,7 +5,7 @@ require 'iconv'
 require 'set'
 
 require 'rubygems'
-require 'xml/libxml'
+require 'nokogiri'
 require 'json'
 require 'curb'
 

data/lib/calais/client.rb

@@ -38,36 +38,37 @@ module Calais
 
     def params_xml
       check_params
-
-      params_node = XML::Node.new('c:params')
+      document = Nokogiri::XML::Document.new
+      
+      params_node = Nokogiri::XML::Node.new('c:params', document)
       params_node['xmlns:c'] = 'http://s.opencalais.com/1/pred/'
       params_node['xmlns:rdf'] = 'http://www.w3.org/1999/02/22-rdf-syntax-ns#'
-
-      processing_node = XML::Node.new('c:processingDirectives')
+      
+      processing_node = Nokogiri::XML::Node.new('c:processingDirectives', document)
       processing_node['c:contentType'] = AVAILABLE_CONTENT_TYPES[@content_type] if @content_type
       processing_node['c:outputFormat'] = AVAILABLE_OUTPUT_FORMATS[@output_format] if @output_format
       processing_node['c:reltagBaseURL'] = @reltag_base_url.to_s if @reltag_base_url
-
+      
       processing_node['c:enableMetadataType'] = @metadata_enables.join(';') unless @metadata_enables.empty?
       processing_node['c:discardMetadata'] = @metadata_discards.join(';') unless @metadata_discards.empty?
       processing_node['c:omitOutputtingOriginalText'] = 'true' if @omit_outputting_original_text
-
-      user_node = XML::Node.new('c:userDirectives')
+      
+      user_node = Nokogiri::XML::Node.new('c:userDirectives', document)
       user_node['c:allowDistribution'] = @allow_distribution.to_s unless @allow_distribution.nil?
       user_node['c:allowSearch'] = @allow_search.to_s unless @allow_search.nil?
       user_node['c:externalID'] = @external_id.to_s if @external_id
       user_node['c:submitter'] = @submitter.to_s if @submitter
-
+      
       params_node << processing_node
       params_node << user_node
-
+      
       if @external_metadata
-        external_node = XML::Node.new('c:externalMetadata')
+        external_node = Nokogiri::XML::Node.new('c:externalMetadata', document)
         external_node << @external_metadata
         params_node << external_node
       end
-
-      params_node.to_s
+      
+      params_node.to_xml(:indent => 2)
     end
 
     private

data/lib/calais/response.rb

@@ -47,14 +47,14 @@ module Calais
     class Instance
       attr_accessor :prefix, :exact, :suffix, :offset, :length
 
-      # Makes a new Instance object from an appropriate LibXML::XML::Node.
+      # Makes a new Instance object from an appropriate Nokogiri::XML::Node.
       def self.from_node(node)
         instance = self.new
-        instance.prefix = node.find_first("c:prefix").content
-        instance.exact = node.find_first("c:exact").content
-        instance.suffix = node.find_first("c:suffix").content
-        instance.offset = node.find_first("c:offset").content.to_i
-        instance.length = node.find_first("c:length").content.to_i
+        instance.prefix = node.xpath("c:prefix[1]").first.content
+        instance.exact  = node.xpath("c:exact[1]").first.content
+        instance.suffix = node.xpath("c:suffix[1]").first.content
+        instance.offset = node.xpath("c:offset[1]").first.content.to_i
+        instance.length = node.xpath("c:length[1]").first.content.to_i
 
         instance
       end
@@ -76,112 +76,112 @@ module Calais
 
     private
       def extract_data
-        doc = XML::Parser.string(@raw_response).parse
+        doc = Nokogiri::XML(@raw_response)
 
-        if doc.root.find("/Error").first
-          raise Calais::Error, doc.root.find("/Error/Exception").first.content
+        if doc.root.xpath("/Error[1]").first
+          raise Calais::Error, doc.root.xpath("/Error/Exception").first.content
         end        
 
-        doc.root.find("rdf:Description/rdf:type[contains(@rdf:resource, '#{MATCHERS[:docinfometa]}')]/..").each do |node|
+        doc.root.xpath("rdf:Description/rdf:type[contains(@rdf:resource, '#{MATCHERS[:docinfometa]}')]/..").each do |node|
           @language = node['language']
           @submission_date =  DateTime.parse node['submissionDate']
 
-          attributes = extract_attributes(node.find("*[contains(name(), 'c:')]"))
+          attributes = extract_attributes(node.xpath("*[contains(name(), 'c:')]"))
 
           @signature = attributes.delete('signature')
           @submitter_code = attributes.delete('submitterCode')
 
-          node.remove!
+          node.remove
         end
 
-        doc.root.find("rdf:Description/rdf:type[contains(@rdf:resource, '#{MATCHERS[:docinfo]}')]/..").each do |node|
+        doc.root.xpath("rdf:Description/rdf:type[contains(@rdf:resource, '#{MATCHERS[:docinfo]}')]/..").each do |node|
           @request_id = node['calaisRequestID']
 
-          attributes = extract_attributes(node.find("*[contains(name(), 'c:')]"))
+          attributes = extract_attributes(node.xpath("*[contains(name(), 'c:')]"))
 
           @doc_title = attributes.delete('docTitle')
-          @doc_date = Date.parse attributes.delete('docDate')
+          @doc_date = Date.parse(attributes.delete('docDate')) 
 
-          node.remove!
+          node.remove
         end
 
-        @categories = doc.root.find("rdf:Description/rdf:type[contains(@rdf:resource, '#{MATCHERS[:doccat]}')]/..").map do |node|
+        @categories = doc.root.xpath("rdf:Description/rdf:type[contains(@rdf:resource, '#{MATCHERS[:doccat]}')]/..").map do |node|
           category = Category.new
-          category.name = node.find_first("c:categoryName").content
-          score = node.find_first("c:score")
+          category.name = node.xpath("c:categoryName[1]").first.content
+          score = node.xpath("c:score[1]").first
           category.score = score.content.to_f unless score.nil?
 
-          node.remove!
+          node.remove
           category
         end
 
-        @relevances = doc.root.find("rdf:Description/rdf:type[contains(@rdf:resource, '#{MATCHERS[:relevances]}')]/..").inject({}) do |acc, node|
-          subject_hash = node.find_first("c:subject")[:resource].split('/')[-1]
-          acc[subject_hash] = node.find_first("c:relevance").content.to_f
+        @relevances = doc.root.xpath("rdf:Description/rdf:type[contains(@rdf:resource, '#{MATCHERS[:relevances]}')]/..").inject({}) do |acc, node|
+          subject_hash = node.xpath("c:subject[1]").first[:resource].split('/')[-1]
+          acc[subject_hash] = node.xpath("c:relevance[1]").first.content.to_f
 
-          node.remove!
+          node.remove
           acc
         end
 
-        @entities = doc.root.find("rdf:Description/rdf:type[contains(@rdf:resource, '#{MATCHERS[:entities]}')]/..").map do |node|
+        @entities = doc.root.xpath("rdf:Description/rdf:type[contains(@rdf:resource, '#{MATCHERS[:entities]}')]/..").map do |node|
           extracted_hash = node['about'].split('/')[-1] rescue nil
 
           entity = Entity.new
           entity.calais_hash = CalaisHash.find_or_create(extracted_hash, @hashes)
           entity.type = extract_type(node)
-          entity.attributes = extract_attributes(node.find("*[contains(name(), 'c:')]"))
+          entity.attributes = extract_attributes(node.xpath("*[contains(name(), 'c:')]"))
 
           entity.relevance = @relevances[extracted_hash]
           entity.instances = extract_instances(doc, extracted_hash)
 
-          node.remove!
+          node.remove
           entity
         end
 
-        @relations = doc.root.find("rdf:Description/rdf:type[contains(@rdf:resource, '#{MATCHERS[:relations]}')]/..").map do |node|
+        @relations = doc.root.xpath("rdf:Description/rdf:type[contains(@rdf:resource, '#{MATCHERS[:relations]}')]/..").map do |node|
           extracted_hash = node['about'].split('/')[-1] rescue nil
 
           relation = Relation.new
           relation.calais_hash = CalaisHash.find_or_create(extracted_hash, @hashes)
           relation.type = extract_type(node)
-          relation.attributes = extract_attributes(node.find("*[contains(name(), 'c:')]"))
+          relation.attributes = extract_attributes(node.xpath("*[contains(name(), 'c:')]"))
           relation.instances = extract_instances(doc, extracted_hash)
 
-          node.remove!
+          node.remove
           relation
         end
 
-        @geographies = doc.root.find("rdf:Description/rdf:type[contains(@rdf:resource, '#{MATCHERS[:geographies]}')]/..").map do |node|
-          attributes = extract_attributes(node.find("*[contains(name(), 'c:')]"))
+        @geographies = doc.root.xpath("rdf:Description/rdf:type[contains(@rdf:resource, '#{MATCHERS[:geographies]}')]/..").map do |node|
+          attributes = extract_attributes(node.xpath("*[contains(name(), 'c:')]"))
 
           geography = Geography.new
           geography.name = attributes.delete('name')
           geography.calais_hash = attributes.delete('subject')
           geography.attributes = attributes
 
-          node.remove!
+          node.remove
           geography
         end
 
-        doc.root.find("rdf:Description/rdf:type[contains(@rdf:resource, '#{MATCHERS[:defaultlangid]}')]/..").each { |node| node.remove! }
-        doc.root.find("./*").each { |node| node.remove! }
+        doc.root.xpath("rdf:Description/rdf:type[contains(@rdf:resource, '#{MATCHERS[:defaultlangid]}')]/..").each { |node| node.remove }
+        doc.root.xpath("./*").each { |node| node.remove }
 
         return
       end
 
       def extract_instances(doc, hash)
-        doc.root.find("rdf:Description/rdf:type[contains(@rdf:resource, '#{MATCHERS[:instances]}')]/..").select do |instance_node|
-          instance_node.find_first("c:subject")[:resource].split("/")[-1] == hash
+        doc.root.xpath("rdf:Description/rdf:type[contains(@rdf:resource, '#{MATCHERS[:instances]}')]/..").select do |instance_node|
+          instance_node.xpath("c:subject[1]").first[:resource].split("/")[-1] == hash
         end.map do |instance_node|
           instance = Instance.from_node(instance_node)
-          instance_node.remove!
+          instance_node.remove
 
           instance
         end
       end
 
       def extract_type(node)
-        node.find("*[name()='rdf:type']")[0]['resource'].split('/')[-1]
+        node.xpath("*[name()='rdf:type']")[0]['resource'].split('/')[-1]
       rescue
         nil
       end

data/spec/calais/client_spec.rb

@@ -32,7 +32,7 @@ end
 describe Calais::Client, :params_xml do
   it 'returns an xml encoded string' do
     client = Calais::Client.new(:content => SAMPLE_DOCUMENT, :license_id => LICENSE_ID)
-    client.params_xml.should == %[<c:params xmlns:c="http://s.opencalais.com/1/pred/" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#">\n  <c:processingDirectives/>\n  <c:userDirectives/>\n</c:params>]
+    client.params_xml.should == %[<c:params xmlns:c=\"http://s.opencalais.com/1/pred/\" xmlns:rdf=\"http://www.w3.org/1999/02/22-rdf-syntax-ns#\">\n  <c:processingDirectives/>\n  <c:userDirectives/>\n</c:params>]
 
     client.content_type = :xml
     client.output_format = :json
@@ -76,4 +76,4 @@ describe Calais::Client, :enlighten do
     @client.enlighten
     @client.instance_variable_get(:@client).url.should == Calais::BETA_REST_ENDPOINT
   end
-end
+end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification 
 name: abhay-calais
 version: !ruby/object:Gem::Version 
-  version: 0.0.7
+  version: 0.0.8
 platform: ruby
 authors: 
 - Abhay Kumar
@@ -9,18 +9,18 @@ autorequire:
 bindir: bin
 cert_chain: []
 
-date: 2009-06-08 00:00:00 -07:00
+date: 2009-09-18 00:00:00 -07:00
 default_executable: 
 dependencies: 
 - !ruby/object:Gem::Dependency 
-  name: libxml-ruby
+  name: nokogiri
   type: :runtime
   version_requirement: 
   version_requirements: !ruby/object:Gem::Requirement 
     requirements: 
     - - ">="
       - !ruby/object:Gem::Version 
-        version: 0.5.4
+        version: 1.3.3
     version: 
 - !ruby/object:Gem::Dependency 
   name: json
@@ -62,6 +62,7 @@ files:
 - lib/calais/response.rb
 has_rdoc: true
 homepage: http://github.com/abhay/calais
+licenses: 
 post_install_message: 
 rdoc_options: 
 - --charset=UTF-8
@@ -82,11 +83,11 @@ required_rubygems_version: !ruby/object:Gem::Requirement
 requirements: []
 
 rubyforge_project: calais
-rubygems_version: 1.2.0
+rubygems_version: 1.3.5
 signing_key: 
 specification_version: 2
 summary: A Ruby interface to the Calais Web Service
 test_files: 
-- spec/helper.rb
-- spec/calais/response_spec.rb
 - spec/calais/client_spec.rb
+- spec/calais/response_spec.rb
+- spec/helper.rb