calais 0.0.3 → 0.0.5

data/History.txt

@@ -1,3 +1,14 @@
+== 0.0.5
+
+* fixed error where classes weren't being required in the proper order on Ubuntu (reported by Jon Moses)
+* New things coming back from the API. Fixing in tests.
+
+== 0.0.4
+
+* changed dependency from hpricot to libxml
+* utf fun
+* cleanup all around
+
 == 0.0.3
 
 * pluginized the library for Rails (via pius: http://gitorious.org/projects/calais-au-rails)

data/README.txt

@@ -26,7 +26,7 @@ This will return an object containing the RDF representation of the text, the na
 
 * Ruby 1.8.5 or better
   * Uses the following standard libraries: digest/sha1, net/http, yaml, cgi
-* Hpricot
+* libxml2 / libxml-ruby
 
 == Install
 

data/lib/calais.rb

@@ -1,14 +1,20 @@
 require 'digest/sha1'
 require 'net/http'
-require 'yaml'
 require 'cgi'
+require 'iconv'
 
 require 'rubygems'
-require 'hpricot'
+require 'xml/libxml'
 
 $KCODE = "UTF8"
+require 'jcode'
 
-Dir.glob(File.join(File.dirname(__FILE__), 'calais/*.rb')).each { |f| require f }
+$:.unshift File.expand_path(File.dirname(__FILE__)) + '/calais'
+
+require 'name'
+require 'relationship'
+require 'response'
+require 'client'
 
 module Calais
   POST_URL = "http://api.opencalais.com"
@@ -35,13 +41,16 @@ module Calais
   
   class << self
     def enlighten(*args, &block) Client.new(*args, &block).call(:enlighten) end
+
     def process_document(*args, &block) 
       data, error = Calais.enlighten(*args, &block)
-      Client.process_data(data, error)
+      process_data(data, error)
     end
+
+    def process_data(data, error=nil) Response.new(data, error) end
   end
 end
 
 module Calais
-  VERSION = '0.0.3'
+  VERSION = '0.0.5'
 end

data/lib/calais/client.rb

@@ -11,37 +11,23 @@ module Calais
       yield(self) if block_given?
     end
     
-    def call(method, times=1)
+    def call(method)
       method = method.intern unless method.is_a?(Symbol)
       raise ArgumentError.new("Unknown method: #{method}") unless AVAILABLE_METHODS.keys.include? method
       
       post_args = {
         "licenseID" => @license_id,
-        "content" => @content,
+        "content" => Iconv.iconv('UTF-8//IGNORE', 'UTF-8',  "#{@content} ").first[0..-2],
         "paramsXML" => params_xml
       }
       
       url = URI.parse(POST_URL + AVAILABLE_METHODS[method])
       resp, data = Net::HTTP.post_form(url, post_args)
       
-      handle_response(resp, data, method, times)
-    end
-    
-    def self.process_data(data, error=nil)
-      Calais::Response.new(data, error)
+      return resp.is_a?(Net::HTTPOK) ? data : [data, "API Error: #{resp}"]
     end
     
     private
-      def handle_response(resp, data, method, times)
-        if resp.is_a? Net::HTTPOK
-          [data, nil]
-        elsif times >= MAX_RETRIES
-          [data, "Too many retries: #{times}"]
-        else
-          call(method, times+1)
-        end
-      end
-      
       def params_xml
         content_type = @content_type && AVAILABLE_CONTENT_TYPES.keys.include?(@content_type) ? AVAILABLE_CONTENT_TYPES[@content_type] : AVAILABLE_CONTENT_TYPES[DEFAULT_CONTENT_TYPE]
         output_format = @output_format && AVAILABLE_OUTPUT_FORMATS.keys.include?(@output_format) ? AVAILABLE_OUTPUT_FORMATS[@output_format] : AVAILABLE_OUTPUT_FORMATS[DEFAULT_OUTPUT_FORMAT]

data/lib/calais/response.rb

@@ -7,15 +7,11 @@ module Calais
       @names = []
       @relationships = []
 
-      parse_rdf(raw)
+      parse_raw(raw)
       return if @error
-
-      h_doc = Hpricot.XML(@rdf)
-      document_node = h_doc.root.search("//rdf:Description//c:document//..").remove.first
-      signature_node = h_doc.root.search("//rdf:Description//c:signature//..").remove.first
-      language_node = h_doc.root.search("//rdf:Description//c:lang//..").remove.first
-      h_doc = parse_names(h_doc)
-      h_doc = parse_relationships(h_doc)
+      
+      parse_names
+      parse_relationships
     end
     
     Name::TYPES.each_pair do |method_name, type|
@@ -25,30 +21,23 @@ module Calais
     end
     
     private
-      def parse_rdf(raw)
-        @rdf = CGI::unescapeHTML Hpricot.XML(raw).at("/string").inner_html
-        @hpricot = Hpricot.XML(@rdf)
-        @error = Hpricot.XML(response).at("/Error/Exception").inner_html rescue @error
+      def parse_raw(raw)
+        @libxml = XML::Parser.string(XML::Parser.string(raw).parse.root.child.content).parse
+        @rdf = @libxml.to_s
+        @error = @libxml.find("/Error/Exception").first.content rescue @error
       end
       
-      def parse_names(doc)
-        name_elements = doc.root.search("//rdf:Description//c:name//..")
-        @names = name_elements.map do |ele|
-          name = ele.at("c:name").inner_html
-          type = ele.at("rdf:type").attributes["rdf:resource"].split("/").last
-          hash = ele.attributes["rdf:about"].split("/").last
-          
-          detection_nodes = doc.root.search("//rdf:Description//c:subject//..").collect! do |ele|
-            ele unless ele.at("c:subject").attributes["rdf:resource"].match(hash).nil?
-          end.compact
+      def parse_names
+        @names = @libxml.root.find("rdf:Description/rdf:type[contains(@rdf:resource, '/em/e/')]/..").map do |n|
+          name = n.find_first("c:name").content
+          type = n.find_first("rdf:type").properties.to_a.assoc("resource").last.split('/').last
+          hash = n.properties.to_a.assoc("about").last.split("/").last
           
-          locations = detection_nodes.map do |ele|
-            start = ele.at("c:offset").inner_html.to_i
-            Range.new(start, start+ele.at("c:length").inner_html.to_i)
+          locations = @libxml.root.find("rdf:Description/c:subject[contains(@rdf:resource, '#{hash}')]/..").map do |n2|
+            start = n2.find_first("c:offset").content.to_i
+            Range.new(start, start+n2.find_first("c:length").content.to_i)
           end
           
-          detection_nodes.remove
-          
           Name.new(
             :name => name,
             :hash => hash,
@@ -56,47 +45,34 @@ module Calais
             :locations => locations
           )
         end
-        name_elements.remove
-        
-        doc
       end
       
-      def parse_relationships(doc)
-        relationship_elements = doc.root.search("rdf:Description")
-        @relationships = relationship_elements.map do |ele|
-          next if ele.at("c:docId")
-
-          hash = ele.attributes["rdf:about"].split("/").last
-          type = ele.at("rdf:type").attributes["rdf:resource"].split("/").last
+      def parse_relationships
+        @libxml.root.find("rdf:Description/rdf:type[contains(@rdf:resource, '/em/r')]/..").each do |n|
+          hash = n.properties.to_a.assoc("about").last.split("/").last
+          type = n.find_first("rdf:type").properties.to_a.assoc("resource").last.split('/').last
+          
           metadata = {}
-          ele.children.each do |child|
-            next if child.comment? || child.name == "rdf:type"
-            
-            value = if child.attributes["rdf:resource"]
-                Name.find_in_names(child.attributes["rdf:resource"].split("/").last, @names) rescue nil
-              else
-                child.inner_html.strip
-              end
-            metadata[child.name.split(":").last] = value
+          
+          n.to_a.each do |n2|
+            next if n2.name == "type" or n2.comment?
+            resource = n2.properties.to_a.assoc("resource")
+            metadata[n2.name] = resource ? Name.find_in_names(resource.last.split("/").last, @names) : n2.content.strip
           end
           
-          locations = doc.root.search("//rdf:Description//c:docId//..").collect! do |ele|
-            ele unless ele.at("c:subject").attributes["rdf:resource"].match(hash).nil?
-          end.compact.map do |ele|
-            start = ele.at("c:offset").inner_html.to_i
-            Range.new(start, start+ele.at("c:length").inner_html.to_i)
+          locations = @libxml.root.find("rdf:Description/c:subject[contains(@rdf:resource, '#{hash}')]/..").map do |n2|
+            start = n2.find_first("c:offset").content.to_i
+            Range.new(start, start+n2.find_first("c:length").content.to_i)
           end
           
-          Relationship.new(
+          
+          @relationships << Relationship.new(
             :type => type,
             :hash => hash,
             :metadata => metadata,
             :locations => locations
           )
-        end.compact
-        relationship_elements.remove
-        
-        doc
+        end
       end
   end
 end

data/spec/calais_spec.rb

@@ -52,7 +52,7 @@ describe Calais, ".process_document" do
   it "returns relationships" do
     @response.relationships.should_not be_nil
     @response.relationships.should_not be_empty
-    @response.relationships.map {|r| r.type }.should == ["PersonProfessional"]
+    @response.relationships.map {|r| r.type }.should == ["Quotation", "Quotation", "PersonProfessional", "Quotation", "Quotation", "Quotation", "Quotation"]
   end
 
 end

metadata

@@ -1,57 +1,33 @@
 --- !ruby/object:Gem::Specification 
+rubygems_version: 0.9.2
+specification_version: 1
 name: calais
 version: !ruby/object:Gem::Version 
-  version: 0.0.3
-platform: ruby
-authors: 
-- Abhay Kumar
+  version: 0.0.5
+date: 2008-03-12 00:00:00 -07:00
+summary: A Ruby interface to the Calais Web Service
+require_paths: 
+- lib
+email: info@opensynapse.net
+homepage: http://calais.rubyforge.org
+rubyforge_project: calais
+description: "== Features  * Accepts documents in text/plain, text/xml and text/html format. * Basic access to the Open Calais API's Enlighten action. * Output is RDF representation of input document. * Single function ability to tag a document and receive a response in RDF format, names in the document, and their relationships.  == Synopsis  This is a very basic wrapper to the Open Calais API. It uses the POST endpoint and currently supports the Enlighten action. Here's a simple call:  Calais.enlighten(:content => \"The government of the United Kingdom has given corporations like fast food chain McDonald's the right to award high school qualifications to employees who complete a company training program.\", :content_type => :text, :license_id => LICENSE_ID)  This is the easiest way to get the RDF-formated response from the OpenCalais service.  If you want to do something more fun like getting all sorts of fun information about a document, you can try this:  Calais.process_document(:content => \"The government of the United Kingdom has given corporations like fast food chain McDonald's the right to award high school qualifications to employees who complete a company training program.\", :content_type => :text, :license_id => LICENSE_ID)  This will return an object containing the RDF representation of the text, the names in the text, and any relationships that exist there."
 autorequire: 
+default_executable: 
 bindir: bin
+has_rdoc: true
+required_ruby_version: !ruby/object:Gem::Version::Requirement 
+  requirements: 
+  - - ">"
+    - !ruby/object:Gem::Version 
+      version: 0.0.0
+  version: 
+platform: ruby
+signing_key: 
 cert_chain: 
-- |
-  -----BEGIN CERTIFICATE-----
-  MIIDNjCCAh6gAwIBAgIBADANBgkqhkiG9w0BAQUFADBBMQ0wCwYDVQQDDARpbmZv
-  MRswGQYKCZImiZPyLGQBGRYLb3BlbnN5bmFwc2UxEzARBgoJkiaJk/IsZAEZFgNu
-  ZXQwHhcNMDgwMjAzMDUwODQzWhcNMDkwMjAyMDUwODQzWjBBMQ0wCwYDVQQDDARp
-  bmZvMRswGQYKCZImiZPyLGQBGRYLb3BlbnN5bmFwc2UxEzARBgoJkiaJk/IsZAEZ
-  FgNuZXQwggEiMA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIBAQCmaA3Od1p42luz
-  zDJepXD3VBFEmmeCUCOjs8rkGIlhRibBvAU8GB0hhkTUykeF6JvAp68FYtIqyTqM
-  EY7bnyYTWsvX7HrX/wGRshSKZPnxn2b0AnZ9T3QQZyUut1YQ5G+kBQrI76hz9ynA
-  l0mPCiGxrh+yUNTKt7KzOAzQbtPlqGiIzj+aYvzmdEsj24Ekm/11A/ntPnz+N/Wj
-  yS5c2tbfZdU8NfwfHCZQUBE4PROYCCjoly0QChvBQzKSZPrEpJB3EedMUyBc5m5E
-  TQ0u5aItr3isQchwo410x7ixzVveVzn4mchaGCZ3ZuPwaQkuI/7KSSWWH1LCouct
-  N7LsWR7jAgMBAAGjOTA3MAkGA1UdEwQCMAAwCwYDVR0PBAQDAgSwMB0GA1UdDgQW
-  BBRsRhkAWj4iWaut121ZcaOAKXG27DANBgkqhkiG9w0BAQUFAAOCAQEAFuqEVgKC
-  U6f237SZ/hzevOwRkaErF1EcaCEVzuNj+KNdbQOK9oOo+hHyos3jUo17TiUNDi+3
-  VJhw3cOkA/PEpa0ou0Vm8VIfXdp6dh62NhTKHBVwQ/qXHnn3aVuV/zIfOmi9WQ+t
-  mr7ehGTw7URly95GOESW4NKQ95p+iquAh/NGhtHGFt+nxjJGUkkYlnGVaxmmgof3
-  sP2hOrejIrD9jAoejiRhiA+IyEoaYJvlh+D+3MngvnyDFqHiFZgngM0fvTnMTsgT
-  avOOKhLsesocjiElkLMv8mwuY+L8P4tSvDTDKXxM9Bx/YagwgzYCqPoGtFdWI/GI
-  +keKvrmaTOJ7CQ==
-  -----END CERTIFICATE-----
-
-date: 2008-02-07 00:00:00 -08:00
-default_executable: 
-dependencies: 
-- !ruby/object:Gem::Dependency 
-  name: hoe
-  version_requirement: 
-  version_requirements: !ruby/object:Gem::Requirement 
-    requirements: 
-    - - ">="
-      - !ruby/object:Gem::Version 
-        version: 1.5.0
-    version: 
-description: "== Features  * Accepts documents in text/plain, text/xml and text/html format. * Basic access to the Open Calais API's Enlighten action. * Output is RDF representation of input document. * Single function ability to tag a document and receive a response in RDF format, names in the document, and their relationships.  == Synopsis  This is a very basic wrapper to the Open Calais API. It uses the POST endpoint and currently supports the Enlighten action. Here's a simple call:  Calais.enlighten(:content => \"The government of the United Kingdom has given corporations like fast food chain McDonald's the right to award high school qualifications to employees who complete a company training program.\", :content_type => :text, :license_id => LICENSE_ID)  This is the easiest way to get the RDF-formated response from the OpenCalais service.  If you want to do something more fun like getting all sorts of fun information about a document, you can try this:  Calais.process_document(:content => \"The government of the United Kingdom has given corporations like fast food chain McDonald's the right to award high school qualifications to employees who complete a company training program.\", :content_type => :text, :license_id => LICENSE_ID)  This will return an object containing the RDF representation of the text, the names in the text, and any relationships that exist there."
-email: info@opensynapse.net
-executables: []
-
-extensions: []
-
-extra_rdoc_files: 
-- History.txt
-- Manifest.txt
-- README.txt
+post_install_message: 
+authors: 
+- Abhay Kumar
 files: 
 - History.txt
 - MIT-LICENSE
@@ -71,32 +47,28 @@ files:
 - spec/fixtures/slovenia_euro.xml
 - spec/helper.rb
 - spec/spec.opts
-has_rdoc: true
-homepage: http://calais.rubyforge.org
-post_install_message: 
+test_files: []
+
 rdoc_options: 
 - --main
 - README.txt
-require_paths: 
-- lib
-required_ruby_version: !ruby/object:Gem::Requirement 
-  requirements: 
-  - - ">="
-    - !ruby/object:Gem::Version 
-      version: "0"
-  version: 
-required_rubygems_version: !ruby/object:Gem::Requirement 
-  requirements: 
-  - - ">="
-    - !ruby/object:Gem::Version 
-      version: "0"
-  version: 
-requirements: []
+extra_rdoc_files: 
+- History.txt
+- Manifest.txt
+- README.txt
+executables: []
 
-rubyforge_project: calais
-rubygems_version: 1.0.1
-signing_key: 
-specification_version: 2
-summary: A Ruby interface to the Calais Web Service
-test_files: []
+extensions: []
+
+requirements: []
 
+dependencies: 
+- !ruby/object:Gem::Dependency 
+  name: hoe
+  version_requirement: 
+  version_requirements: !ruby/object:Gem::Version::Requirement 
+    requirements: 
+    - - ">="
+      - !ruby/object:Gem::Version 
+        version: 1.4.0
+    version: 

data.tar.gz.sig

@@ -1 +0,0 @@
-��i/>�[ɵ:��m��<����