calais 0.0.1

data.tar.gz.sig

@@ -0,0 +1 @@
+ ���Vw&�+�@6l8g8�����uɂ��H�L�E�[a�-&,���v�l���uo���#{
�3�T�
���v��%ơ�������[����\գ�\p��%0�K���A�PfB�Z�&����X����$�	M�`�W_������F	eD*�ɖ��,�^t�{�8̮$���n�V�Y���9�(c@j���F̹k |��ȏ�d�j4��ߤo�8���g��xG�wkd���X���F᭞/��

data/History.txt

@@ -0,0 +1,5 @@
+== 0.0.1
+
+* Access to Calais's Enlighten action
+* Single method to process a document
+  * Get relationships and names from a document

data/Manifest.txt

@@ -0,0 +1,14 @@
+History.txt
+Manifest.txt
+README.txt
+Rakefile
+lib/calais.rb
+lib/calais/client.rb
+lib/calais/response.rb
+spec/calais_spec.rb
+spec/fixtures/.gitignore
+spec/fixtures/bicycles_austrailia.xml
+spec/fixtures/calais.yml.sample
+spec/fixtures/slovenia_euro.xml
+spec/helper.rb
+spec/spec.opts

data/README.txt

@@ -0,0 +1,39 @@
+== Calais
+A Ruby interface to the Open Calais Web Service (http://opencalais.com)
+
+== 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.
+
+== Requirements
+
+* Ruby 1.8.5 or better
+  * Uses the following standard libraries: digest/sha1, net/http, yaml, cgi
+* Hpricot
+
+== Install
+
+TODO
+
+== Authors
+
+Abhay Kumar
+http://opensynapse.net
+

data/Rakefile

@@ -0,0 +1,37 @@
+# -*- ruby -*-
+
+require 'rubygems'
+require 'hoe'
+require 'spec/rake/spectask'
+
+require './lib/calais.rb'
+
+Hoe.new('calais', Calais::VERSION) do |p|
+  p.rubyforge_name = 'calais'
+  p.author = 'Abhay Kumar'
+  p.summary = 'A Ruby interface to the Calais Web Service'
+  p.description = p.paragraphs_of('README.txt', 1..9).join("\n\n")
+  p.email = 'info@opensynapse.net'
+  p.url = 'http://calais.rubyforge.org'
+  p.changes = p.paragraphs_of('History.txt', 0..1).join("\n\n")
+  p.remote_rdoc_dir = ''
+end
+
+desc "Run all specs"
+Spec::Rake::SpecTask.new do |t|
+  t.spec_files = FileList["spec/*_spec.rb"].sort
+  t.spec_opts = ["--options", "spec/spec.opts"]
+end
+
+desc "Run all specs and get coverage statistics"
+Spec::Rake::SpecTask.new('coverage') do |t|
+  t.spec_opts = ["--options", "spec/spec.opts"]
+  t.spec_files = FileList["spec/*_spec.rb"].sort
+  t.rcov_opts = ["--exclude", "spec", "--exclude", "gems"]
+  t.rcov = true
+end
+
+Rake::Task[:default].prerequisites.clear
+task :default => :spec
+
+# vim: syntax=Ruby

data/lib/calais.rb

@@ -0,0 +1,47 @@
+require 'digest/sha1'
+require 'net/http'
+require 'yaml'
+require 'cgi'
+
+require 'rubygems'
+require 'hpricot'
+
+$KCODE = "UTF8"
+
+Dir.glob(File.join(File.dirname(__FILE__), 'calais/*.rb')).each { |f| require f }
+
+module Calais
+  POST_URL = "http://api.opencalais.com"
+  
+  AVAILABLE_OUTPUT_FORMATS = {
+    :rdf => "XML/RDF"
+  }
+  DEFAULT_OUTPUT_FORMAT = :rdf
+  
+  AVAILABLE_CONTENT_TYPES = {
+    :xml => "TEXT/XML",
+    :html => "TEXT/HTML",
+    :text => "TEXT/TXT"
+  }
+  DEFAULT_CONTENT_TYPE = :xml
+  
+  DEFAULT_SUBMITTER = "calais.rb"
+  
+  AVAILABLE_METHODS = {
+    :enlighten => "/enlighten/calais.asmx/Enlighten"
+  }
+  
+  MAX_RETRIES = 5
+  
+  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)
+    end
+  end
+end
+
+module Calais
+  VERSION = '0.0.1'
+end

data/lib/calais/client.rb

@@ -0,0 +1,61 @@
+module Calais
+  class Client
+    attr_accessor :license_id
+    attr_accessor :content
+    attr_accessor :content_type, :output_format
+    attr_accessor :allow_distribution, :allow_search, :submitter, :external_id
+    attr_accessor :external_metadata
+    
+    def initialize(options={}, &block)
+      options.each {|k,v| send("#{k}=", v)}
+      yield(self) if block_given?
+    end
+    
+    def call(method, times=1)
+      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,
+        "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)
+    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]
+        allow_distribution = @allow_distribution ? "true" : "false"
+        allow_search = @allow_search ? "true" : "false"
+        submitter = @submitter || DEFAULT_SUBMITTER
+        external_id = @external_id || Digest::SHA1.hexdigest(@content.inspect)
+        external_metadata = @external_metadata || ""
+        
+        xml  = %[<c:params xmlns:c="http://s.opencalais.com/1/pred/" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#">]
+        xml += %[<c:processingDirectives c:contentType="#{content_type}" c:outputFormat="#{output_format}"></c:processingDirectives>]
+        xml += %[<c:userDirectives c:allowDistribution="#{allow_distribution}" c:allowSearch="#{allow_search}" c:externalID="#{external_id}" c:submitter="#{submitter}"></c:userDirectives>]
+        xml += %[<c:externalMetadata>#{external_metadata}</c:externalMetadata>]
+        xml += %[</c:params>]
+      end
+  end
+end

data/lib/calais/response.rb

@@ -0,0 +1,107 @@
+module Calais
+  class Response
+    attr_reader :rdf, :names, :relationships, :error
+    
+    def initialize(raw, error=nil)
+      @error = error
+      @names = []
+      @relationships = []
+
+      parse_rdf(raw)
+      parse_names
+      parse_relationships
+    end
+    
+    def to_dot
+      used = []
+      id = @hpricot.at("rdf:Description//c:document//..").attributes["c:externalID"]
+      
+      dot = "digraph \"#{id}\"\n"
+      dot += "{\n"
+      dot += "\tgraph [rankdir=LR, overlap=false];\n"
+      dot += "\tnode [shape = circle];"
+
+      @relationships.each do |rel|
+        dot += "\t\"#{rel.actor.name}\" -> \"#{rel.target.name}\""
+        dot += " ["
+        dot += "label=\""
+        dot += "#{rel.metadata} " if rel.metadata
+        dot += "(#{rel.type})"
+        dot += "\"];\n"
+        used |= [rel.actor.hash, rel.target.hash]
+      end
+      
+      @names.each do |name|
+        dot += "\t\"#{name.name}\";\n" unless used.include?(name.hash)
+      end
+      dot += "}\n"
+      
+      f = File.open("#{id}.dot", 'w')
+      f.puts dot
+      f.close
+      
+    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
+      end
+      
+      def parse_names
+        @names = @hpricot.root.search("rdf:Description//c:name//..").map do |ele|
+          Calais::Response::Name.new(
+            :name => ele.at("c:name").inner_html,
+            :hash => ele.attributes["rdf:about"].split('/').last,
+            :type => ele.at("rdf:type").attributes["rdf:resource"].split('/').last
+          )
+        end unless @error
+      end
+      
+      def parse_relationships
+        doc = Hpricot.XML(@rdf)
+        doc.search("rdf:Description//c:docId//..").remove
+        doc.search("rdf:Description//c:document//..").remove
+        doc.search("rdf:Description//c:name//..").remove
+        
+        @relationships = doc.root.search("rdf:Description").map do |ele|
+          relationship = ele.at("rdf:type")
+          actor = relationship.next_sibling
+          metadata = actor.next_sibling.attributes["rdf:resource"] ? nil : actor.next_sibling.inner_html.strip
+          target = metadata ? actor.next_sibling.next_sibling : actor.next_sibling
+          
+          actor_name = actor ? Name.find_in_names(actor.attributes["rdf:resource"].split('/').last, @names) : nil
+          target_name = target ? Name.find_in_names(target.attributes["rdf:resource"].split('/').last, @names) :  nil
+
+          Calais::Response::Relationship.new(
+            :type => relationship.attributes["rdf:resource"].split('/').last,
+            :actor => actor_name,
+            :target => target_name,
+            :metadata => metadata
+          )
+        end
+      end
+    
+    class Name
+      include Comparable
+      attr_accessor :name, :type, :hash
+      
+      def initialize(args={})
+        args.each {|k,v| send("#{k}=", v)}
+      end
+      
+      def self.find_in_names(hash, names)
+        names.select {|name| name.hash == hash }.first
+      end
+    end
+    
+    class Relationship
+      attr_accessor :type, :actor, :target, :metadata
+      
+      def initialize(args={})
+        args.each {|k,v| send("#{k}=", v)}
+      end
+    end
+  end
+end

data/spec/calais_spec.rb

@@ -0,0 +1,72 @@
+require File.expand_path("#{File.dirname(__FILE__)}/helper")
+
+describe Calais do
+  it "provides a version number" do
+    Calais::VERSION.should_not be_nil
+  end
+end
+
+describe Calais::Client, ".new" do
+  it "accepts arguments as a hash" do
+    client = nil
+    
+    lambda { client = Calais::Client.new(:content => SAMPLE_DOCUMENT, :license_id => LICENSE_ID) }.should_not raise_error(ArgumentError)
+    
+    client.license_id.should == LICENSE_ID
+    client.content.should == SAMPLE_DOCUMENT
+  end
+  
+  it "accepts arguments as a block" do
+    client = nil
+    
+    lambda {
+      client = Calais::Client.new do |c|
+        c.content = SAMPLE_DOCUMENT
+        c.license_id = LICENSE_ID
+      end
+    }.should_not raise_error(ArgumentError)
+    
+    client.license_id.should == LICENSE_ID
+    client.content.should == SAMPLE_DOCUMENT
+  end
+  
+  it "should not accept unknown attributes" do
+    lambda { Calais::Client.new(:monkey => "monkey", :license_id => LICENSE_ID) }.should raise_error(NoMethodError)
+  end
+end
+
+describe Calais, ".process_document" do
+  it "returns a Calais::Response" do
+    response = Calais.process_document(:content => SAMPLE_DOCUMENT, :content_type => :xml, :license_id => LICENSE_ID)
+    response.should_not be_nil
+    response.should be_a_kind_of(Calais::Response)
+  end
+
+  it "returns a Calais::Response (with relationships)" do
+    response = Calais.process_document(:content => File.read(File.join(File.dirname(__FILE__), 'fixtures', 'bicycles_austrailia.xml')), :content_type => :xml, :license_id => LICENSE_ID)
+    response.should_not be_nil
+    response.should be_a_kind_of(Calais::Response)
+  end
+end
+
+describe Calais::Client, ".call" do
+  before(:all) do
+    @client = Calais::Client.new(:content => SAMPLE_DOCUMENT, :license_id => LICENSE_ID)
+  end
+
+  it "accepts known methods" do
+    lambda { @client.call('enlighten') }.should_not raise_error(ArgumentError)
+  end
+
+  it "complains about unkown methods" do
+    lambda { @client.call('monkey') }.should raise_error(ArgumentError)
+  end
+end
+
+describe Calais::Client, ".params_xml" do
+  it "returns an xml encoded string" do
+    client = Calais::Client.new(:content => SAMPLE_DOCUMENT, :content_type => :xml, :license_id => LICENSE_ID)
+    client.send("params_xml").should_not be_nil
+    client.send("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#\"><c:processingDirectives c:contentType=\"TEXT/XML\" c:outputFormat=\"XML/RDF\"></c:processingDirectives><c:userDirectives c:allowDistribution=\"false\" c:allowSearch=\"false\" c:externalID=\"dc68d5a382724c2238d9f22ba9c0b4d2581569d8\" c:submitter=\"calais.rb\"></c:userDirectives><c:externalMetadata></c:externalMetadata></c:params>]
+  end
+end

data/spec/fixtures/.gitignore

@@ -0,0 +1 @@
+calais.yml

data/spec/fixtures/bicycles_austrailia.xml

@@ -0,0 +1,18 @@
+<document>
+<id>57831</id>
+<title>Record number of bicycles sold in Australia in 2006</title>
+<date>January 4, 2007</date>
+<body>
+Bicycle sales in Australia have recorded record sales of 1,273,781 units for 2006, exceeding car sales by 32 percent. It is the fifth year in a row that the bicycle industry has sold more than one million units, a figure yet to be realised by car manufacturers.
+
+The Cycling Promotion Fund (CPF) spokesman Ian Christie said Australians were increasingly using bicycles as an alternative to cars. Sales rose nine percent in 2006 while the car market stalled. Mr Christie said people were looking to cut their fuel costs and improve their fitness.
+
+Mr Christie said organisations were beginning to supply bicycles as a company vehicle. "There is an emerging trend towards people using bikes as their official company-supplied vehicle in place of the traditional company car," he said.
+
+"Some of Australia's biggest corporations now have bicycle fleets, and when you add in government organisations, we now know of at least 50 organisations which operate fleets of bikes."
+
+"Although the company bicycle is a long way from taking over from the company car, it's an important trend when you consider that nearly half of all cars sold are to company fleets."
+
+The CPF claims most commutes to work are less than 5 kilometres (3 miles) making bicycle travel a viable alternative.
+
+</body></document>

data/spec/fixtures/calais.yml.sample

@@ -0,0 +1 @@
+key: asdlfkjasg08ahng2g08agds

data/spec/fixtures/slovenia_euro.xml

@@ -0,0 +1,14 @@
+<document>
+<id>57507</id>
+<title>Slovenia adopts euro</title>
+<date>January 1, 2007</date>
+<body>
+ in Europe.]]
+
+On January 1 2007, [[Slovenia]] officially joined the Eurozone and adopted the euro as its new official currency. At the same time, Slovenian euro coins, which were available as a "starter kit" from December 15, became legal tender everywhere in the Eurozone.
+
+A period of dual circulation, when both the former currency, the tolar and the euro will be accepted, will only last until January 14. All banks are going to exchange tolars into euros at the official rate of 239.64 SIT per 1 EUR at no commission until March 1, 2007. After that date it will be possible to exchange tolar banknotes (unlimited) and coins (until 2016) only at the Bank of Slovenia.
+
+Most ATMs already dispense euro banknotes with the rest to follow in the following days. Commercial banks automatically converted all tolar accounts into euros and online banking systems will be accessible to their users again on January 3. Even though January 1 and 2 are non-working days in Slovenia, many banks will be opened from 10:00 to 14:00 on both days, available only for one service: exchanging tolars to euros and larger euro banknotes into smaller units.
+
+</body></document>

data/spec/helper.rb

@@ -0,0 +1,8 @@
+require 'rubygems'
+require 'spec'
+require 'yaml'
+
+require File.dirname(__FILE__) + '/../lib/calais'
+
+SAMPLE_DOCUMENT = File.read(File.join(File.dirname(__FILE__), 'fixtures/slovenia_euro.xml'))
+LICENSE_ID = YAML.load(File.read(File.join(File.dirname(__FILE__), 'fixtures/calais.yml')))['key']

data/spec/spec.opts

@@ -0,0 +1,4 @@
+--colour
+--format s
+--backtrace
+--loadby mtime

metadata

@@ -0,0 +1,98 @@
+--- !ruby/object:Gem::Specification 
+name: calais
+version: !ruby/object:Gem::Version 
+  version: 0.0.1
+platform: ruby
+authors: 
+- Abhay Kumar
+autorequire: 
+bindir: bin
+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: 2008-02-02 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
+files: 
+- History.txt
+- Manifest.txt
+- README.txt
+- Rakefile
+- lib/calais.rb
+- lib/calais/client.rb
+- lib/calais/response.rb
+- spec/calais_spec.rb
+- spec/fixtures/.gitignore
+- spec/fixtures/bicycles_austrailia.xml
+- spec/fixtures/calais.yml.sample
+- spec/fixtures/slovenia_euro.xml
+- spec/helper.rb
+- spec/spec.opts
+has_rdoc: true
+homepage: http://calais.rubyforge.org
+post_install_message: 
+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: []
+
+rubyforge_project: calais
+rubygems_version: 1.0.1
+signing_key: 
+specification_version: 2
+summary: A Ruby interface to the Calais Web Service
+test_files: []
+