gkellogg-reddy 0.2.1

data/spec/rdfcore_helper.rb

@@ -0,0 +1,137 @@
+require 'matchers'
+
+module RdfCoreHelper
+  # Class representing test cases in format http://www.w3.org/2000/10/rdf-tests/rdfcore/testSchema#
+  class TestCase
+    include Matchers
+    
+    TEST_DIR = File.join(File.dirname(__FILE__), 'rdfcore')
+    NT_DIR = File.join(File.dirname(__FILE__), 'rdfa-triples')
+    BASE_MANIFEST_URL = "http://rdfa.digitalbazaar.com/test-suite/"
+    BASE_TEST_CASE_URL = "#{BASE_MANIFEST_URL}test-cases/"
+    
+    attr_accessor :about
+    attr_accessor :approval
+    attr_accessor :conclusionDocument
+    attr_accessor :data
+    attr_accessor :description
+    attr_accessor :discussion
+    attr_accessor :document
+    attr_accessor :entailmentRules
+    attr_accessor :inputDocument
+    attr_accessor :issue
+    attr_accessor :name
+    attr_accessor :outputDocument
+    attr_accessor :premiseDocument
+    attr_accessor :rdf_type
+    attr_accessor :status
+    attr_accessor :warning
+    attr_accessor :parser
+    
+    @@test_cases = []
+    
+    def initialize(triples)
+      triples.each do |statement|
+        next if statement.subject.is_a?(BNode)
+#        self.about ||= statement.subject
+#        self.name ||= statement.subject.short_name
+        
+        if statement.is_type?
+          self.rdf_type = statement.object.short_name
+        elsif statement.predicate.short_name =~ /Document\Z/i
+          #puts "#{statement.predicate.short_name}: #{statement.object.inspect}"
+          self.send("#{statement.predicate.short_name}=", statement.object.to_s.sub(/^.*rdfcore/, TEST_DIR))
+          if statement.predicate.short_name == "inputDocument"
+            self.about ||= statement.object
+            self.name ||= statement.object.short_name
+          end
+        elsif self.respond_to?("#{statement.predicate.short_name}=")
+          self.send("#{statement.predicate.short_name}=", statement.object.to_s)
+        end
+      end
+    end
+    
+    def inspect
+      "[Test Case " + %w(
+        about
+        name
+        inputDocument
+        outputDocument
+        issue
+        status
+        approval
+        description
+        discussion
+        issue
+        warning
+      ).map {|a| v = self.send(a); "#{a}='#{v}'" if v}.compact.join(", ") +
+      "]"
+    end
+    
+    # Read in file, and apply modifications reference either .html or .xhtml
+    def input
+      File.read(inputDocument)
+    end
+
+    def output
+      outputDocument && File.read(outputDocument)
+    end
+
+    # Run test case, yields input for parser to create triples
+    def run_test
+      rdf_string = input
+
+      # Run
+      @parser = RdfXmlParser.new
+      yield(rdf_string, @parser)
+
+      @parser.graph.should be_equivalent_graph(output, self) if output
+    end
+
+    def trace
+      @parser.debug.to_a.join("\n")
+    end
+    
+    def self.parse_test_cases
+      return unless @@test_cases.empty?
+      
+      @@positive_parser_tests = []
+      @@negative_parser_tests = []
+      @@positive_entailment_tests = []
+      @@negative_entailment_tests = []
+
+      manifest = File.read(File.join(TEST_DIR, "Manifest.rdf"))
+      parser = RdfXmlParser.new
+      begin
+        parser.parse(manifest, "http://www.w3.org/2000/10/rdf-tests/rdfcore/Manifest.rdf")
+      rescue
+        raise "Parse error: #{$!}\n\t#{parser.debug.join("\t\n")}\n\n"
+      end
+      graph = parser.graph
+      
+      # Group by subject
+      test_hash = graph.triples.inject({}) do |hash, st|
+        a = hash[st.subject] ||= []
+        a << st
+        hash
+      end
+      
+      @@test_cases = test_hash.values.map {|statements| TestCase.new(statements)}.compact.sort_by{|t| t.about.is_a?(URIRef) ? t.about.uri.to_s : "zzz"}
+      
+      @@test_cases.each do |tc|
+        next unless tc.status == "APPROVED"
+        case tc.rdf_type
+        when "PositiveParserTest" then @@positive_parser_tests << tc
+        when "NegativeParserTest" then @@negative_parser_tests << tc
+        when "PositiveEntailmentTest" then @@positive_entailment_tests << tc
+        when "NegativeEntailmentTest" then @@negative_entailment_tests << tc
+        end
+      end
+    end
+    def self.test_cases;                parse_test_cases; @@test_cases; end
+    def self.positive_parser_tests;     parse_test_cases; @@positive_parser_tests; end
+    def self.negative_parser_tests;     parse_test_cases; @@negative_parser_tests; end
+    def self.positive_entailment_tests; parse_test_cases; @@positive_entailment_tests; end
+    def self.negative_entailment_tests; parse_test_cases; @@negative_entailment_tests; end
+  end
+end

data/spec/rdfxml_spec.rb

@@ -0,0 +1,349 @@
+require File.join(File.dirname(__FILE__), 'spec_helper')
+include Reddy
+
+# w3c test suite: http://www.w3.org/TR/rdf-testcases/
+
+describe "RDF/XML Parser" do
+  before(:each) { @parser = RdfXmlParser.new }
+
+  it "should recognise and create single triple for empty non-RDF root" do
+    sampledoc = <<-EOF;
+<?xml version="1.0" ?>
+<NotRDF />
+EOF
+    graph = @parser.parse(sampledoc, "http://example.com", :strict => true)
+    graph.size.should == 1
+    statement = graph[0]
+    statement.subject.class.should == BNode
+    statement.predicate.to_s.should == RDF_TYPE.to_s
+    statement.object.to_s.should == XML_NS.uri.to_s + "NotRDF"
+  end
+  
+  it "should trigger parsing on XML documents with multiple RDF nodes" do
+    sampledoc = <<-EOF;
+<?xml version="1.0" ?>
+<GenericXML xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:ex="http://example.org/">
+  <rdf:RDF>
+    <rdf:Description rdf:about="http://example.org/one">
+      <ex:name>Foo</ex:name>
+    </rdf:Description>
+  </rdf:RDF>
+  <blablabla />
+  <rdf:RDF>
+    <rdf:Description rdf:about="http://example.org/two">
+      <ex:name>Bar</ex:name>
+    </rdf:Description>
+  </rdf:RDF>
+</GenericXML>
+    EOF
+    graph = @parser.parse(sampledoc, "http://example.com", :strict => true)
+    [graph[0].object.to_s, graph[1].object.to_s].sort.should == ["Bar", "Foo"].sort
+  end
+  
+  it "should be able to parse a simple single-triple document" do
+    sampledoc = <<-EOF;
+<?xml version="1.0" ?>
+<rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
+xmlns:ex="http://www.example.org/" xml:lang="en" xml:base="http://www.example.org/foo">
+  <ex:Thing rdf:about="http://example.org/joe" ex:name="bar">
+    <ex:belongsTo rdf:resource="http://tommorris.org/" />
+    <ex:sampleText rdf:datatype="http://www.w3.org/2001/XMLSchema#string">foo</ex:sampleText>
+    <ex:hadADodgyRelationshipWith>
+      <rdf:Description>
+        <ex:name>Tom</ex:name>
+        <ex:hadADodgyRelationshipWith>
+          <rdf:Description>
+            <ex:name>Rob</ex:name>
+            <ex:hadADodgyRelationshipWith>
+              <rdf:Description>
+                <ex:name>Mary</ex:name>
+              </rdf:Description>
+            </ex:hadADodgyRelationshipWith>
+          </rdf:Description>
+        </ex:hadADodgyRelationshipWith>
+      </rdf:Description>
+    </ex:hadADodgyRelationshipWith>
+  </ex:Thing>
+</rdf:RDF>
+    EOF
+
+    graph = @parser.parse(sampledoc, "http://example.com", :strict => true)
+    #puts @parser.debug
+    graph.size.should == 10
+    # print graph.to_ntriples
+    # TODO: add datatype parsing
+    # TODO: make sure the BNode forging is done correctly - an internal element->nodeID mapping
+    # TODO: proper test
+  end
+  
+  it "should raise an error if rdf:aboutEach is used, as per the negative parser test rdfms-abouteach-error001 (rdf:aboutEach attribute)" do
+    sampledoc = <<-EOF;
+<?xml version="1.0" ?>
+    <rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
+             xmlns:eg="http://example.org/">
+  
+      <rdf:Bag rdf:ID="node">
+        <rdf:li rdf:resource="http://example.org/node2"/>
+      </rdf:Bag>
+  
+      <rdf:Description rdf:aboutEach="#node">
+        <dc:rights xmlns:dc="http://purl.org/dc/elements/1.1/">me</dc:rights>
+  
+      </rdf:Description>
+  
+    </rdf:RDF>
+    EOF
+    
+    lambda do
+      graph = @parser.parse(sampledoc, "http://example.com", :strict => true)
+    end.should raise_error(InvalidPredicate, /Obsolete attribute .*aboutEach/)
+  end
+  
+  it "should raise an error if rdf:aboutEachPrefix is used, as per the negative parser test rdfms-abouteach-error002 (rdf:aboutEachPrefix attribute)" do
+    sampledoc = <<-EOF;
+<?xml version="1.0" ?>
+    <rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
+             xmlns:eg="http://example.org/">
+  
+      <rdf:Description rdf:about="http://example.org/node">
+        <eg:property>foo</eg:property>
+      </rdf:Description>
+  
+      <rdf:Description rdf:aboutEachPrefix="http://example.org/">
+        <dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">me</dc:creator>
+  
+      </rdf:Description>
+  
+    </rdf:RDF>
+    EOF
+    
+    lambda do
+      graph = @parser.parse(sampledoc, "http://example.com", :strict => true)
+    end.should raise_error(InvalidPredicate, /Obsolete attribute .*aboutEachPrefix/)
+  end
+  
+  it "should fail if given a non-ID as an ID (as per rdfcore-rdfms-rdf-id-error001)" do
+    sampledoc = <<-EOF;
+<?xml version="1.0"?>
+    <rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#">
+     <rdf:Description rdf:ID='333-555-666' />
+    </rdf:RDF>
+    EOF
+    
+    lambda do
+      graph = @parser.parse(sampledoc, "http://example.com", :strict => true)
+    end.should raise_error(ParserException, /ID addtribute '.*' must be a NCName/)
+  end
+  
+  it "should make sure that the value of rdf:ID attributes match the XML Name production (child-element version)" do
+    sampledoc = <<-EOF;
+<?xml version="1.0" ?>
+    <rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
+             xmlns:eg="http://example.org/">
+     <rdf:Description>
+       <eg:prop rdf:ID="q:name" />
+     </rdf:Description>
+    </rdf:RDF>
+    EOF
+    
+    lambda do
+      graph = @parser.parse(sampledoc, "http://example.com", :strict => true)
+    end.should raise_error(ParserException, /ID addtribute '.*' must be a NCName/)
+  end
+  
+  it "should be able to reify according to §2.17 of RDF/XML Syntax Specification" do
+    sampledoc = <<-EOF;
+<?xml version="1.0"?>
+    <rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
+             xmlns:ex="http://example.org/stuff/1.0/"
+             xml:base="http://example.org/triples/">
+      <rdf:Description rdf:about="http://example.org/">
+        <ex:prop rdf:ID="triple1">blah</ex:prop>
+      </rdf:Description>
+    </rdf:RDF>
+    EOF
+
+    triples = <<-EOF
+<http://example.org/> <http://example.org/stuff/1.0/prop> \"blah\" .
+<http://example.org/triples/#triple1> <http://www.w3.org/1999/02/22-rdf-syntax-ns#type> <http://www.w3.org/1999/02/22-rdf-syntax-ns#Statement> .
+<http://example.org/triples/#triple1> <http://www.w3.org/1999/02/22-rdf-syntax-ns#subject> <http://example.org/> .
+<http://example.org/triples/#triple1> <http://www.w3.org/1999/02/22-rdf-syntax-ns#predicate> <http://example.org/stuff/1.0/prop> .
+<http://example.org/triples/#triple1> <http://www.w3.org/1999/02/22-rdf-syntax-ns#object> \"blah\" .
+EOF
+    tc = RdfCoreHelper::TestCase.new([])
+    tc.about = "http://example.com"
+    tc.parser = @parser
+
+    graph = @parser.parse(sampledoc, tc.about, :strict => true)
+    graph.should be_equivalent_graph(triples, tc)
+  end
+  
+  it "should make sure that the value of rdf:ID attributes match the XML Name production (data attribute version)" do
+    sampledoc = <<-EOF;
+<?xml version="1.0" ?>
+    <rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
+             xmlns:eg="http://example.org/">
+     <rdf:Description rdf:ID="a/b" eg:prop="val" />
+    </rdf:RDF>
+    EOF
+    
+    lambda do
+      graph = @parser.parse(sampledoc, "http://example.com", :strict => true)
+    end.should raise_error(ParserException, "ID addtribute 'a/b' must be a NCName")
+  end
+  
+  it "should be able to handle Bags/Alts etc." do
+    sampledoc = <<-EOF;
+<?xml version="1.0" ?>
+<rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:eg="http://example.org/">
+  <rdf:Bag>
+    <rdf:li rdf:resource="http://tommorris.org/" />
+    <rdf:li rdf:resource="http://twitter.com/tommorris" />
+  </rdf:Bag>
+</rdf:RDF>
+    EOF
+    graph = @parser.parse(sampledoc, "http://example.com", :strict => true)
+    graph.predicates.should include("http://www.w3.org/1999/02/22-rdf-syntax-ns#_1", "http://www.w3.org/1999/02/22-rdf-syntax-ns#_2")
+  end
+  
+  # # when we have decent Unicode support, add http://www.w3.org/2000/10/rdf-tests/rdfcore/rdfms-rdf-id/error005.rdf
+  # 
+  # it "should support reification" do
+  #   pending
+  # end
+  # 
+  it "should detect bad bagIDs" do
+    sampledoc = <<-EOF;
+<?xml version="1.0" ?>
+    <rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#">
+     <rdf:Description rdf:bagID='333-555-666' />
+    </rdf:RDF>
+    EOF
+    
+    lambda do
+      graph = @parser.parse(sampledoc, "http://example.com", :strict => true)
+      puts @parser.debug
+    end.should raise_error(InvalidPredicate, /Obsolete attribute .*bagID/)
+  end
+
+  it "should parse testcase" do
+    sampledoc = <<-EOF;
+<?xml version="1.0" ?>
+<rdf:RDF
+		xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
+		xmlns:rdfs="http://www.w3.org/2000/01/rdf-schema#"
+		xmlns:test="http://www.w3.org/2000/10/rdf-tests/rdfcore/testSchema#"
+>
+<!-- amp-in-url/Manifest.rdf -->
+<test:PositiveParserTest rdf:about="http://www.w3.org/2000/10/rdf-tests/rdfcore/amp-in-url/Manifest.rdf#test001">
+
+   <test:status>APPROVED</test:status>
+   <test:approval rdf:resource="http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001Sep/0326.html" />
+   <!-- <test:discussion rdf:resource="pointer to archived email or other discussion" /> -->
+   <!-- <test:description>
+	-if we have a description, fill it in here -
+   </test:description> -->
+
+   <test:inputDocument>
+      <test:RDF-XML-Document rdf:about="http://www.w3.org/2000/10/rdf-tests/rdfcore/amp-in-url/test001.rdf" />
+   </test:inputDocument>
+
+   <test:outputDocument>
+      <test:NT-Document rdf:about="http://www.w3.org/2000/10/rdf-tests/rdfcore/amp-in-url/test001.nt" />
+   </test:outputDocument>
+
+</test:PositiveParserTest>
+</rdf:RDF>
+EOF
+
+    triples = <<-EOF
+<http://www.w3.org/2000/10/rdf-tests/rdfcore/amp-in-url/Manifest.rdf#test001> <http://www.w3.org/1999/02/22-rdf-syntax-ns#type> <http://www.w3.org/2000/10/rdf-tests/rdfcore/testSchema#PositiveParserTest> .
+<http://www.w3.org/2000/10/rdf-tests/rdfcore/amp-in-url/Manifest.rdf#test001> <http://www.w3.org/2000/10/rdf-tests/rdfcore/testSchema#approval> <http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001Sep/0326.html> .
+<http://www.w3.org/2000/10/rdf-tests/rdfcore/amp-in-url/Manifest.rdf#test001> <http://www.w3.org/2000/10/rdf-tests/rdfcore/testSchema#inputDocument> <http://www.w3.org/2000/10/rdf-tests/rdfcore/amp-in-url/test001.rdf> .
+<http://www.w3.org/2000/10/rdf-tests/rdfcore/amp-in-url/Manifest.rdf#test001> <http://www.w3.org/2000/10/rdf-tests/rdfcore/testSchema#outputDocument> <http://www.w3.org/2000/10/rdf-tests/rdfcore/amp-in-url/test001.nt> .
+<http://www.w3.org/2000/10/rdf-tests/rdfcore/amp-in-url/Manifest.rdf#test001> <http://www.w3.org/2000/10/rdf-tests/rdfcore/testSchema#status> "APPROVED" .
+<http://www.w3.org/2000/10/rdf-tests/rdfcore/amp-in-url/test001.nt> <http://www.w3.org/1999/02/22-rdf-syntax-ns#type> <http://www.w3.org/2000/10/rdf-tests/rdfcore/testSchema#NT-Document> .
+<http://www.w3.org/2000/10/rdf-tests/rdfcore/amp-in-url/test001.rdf> <http://www.w3.org/1999/02/22-rdf-syntax-ns#type> <http://www.w3.org/2000/10/rdf-tests/rdfcore/testSchema#RDF-XML-Document> .
+EOF
+    tc = RdfCoreHelper::TestCase.new([])
+    tc.about = "http://www.w3.org/2000/10/rdf-tests/rdfcore/xmlbase/Manifest.rdf#test001"
+    tc.parser = @parser
+    graph = @parser.parse(sampledoc, tc.about, :strict => true)
+    graph.should be_equivalent_graph(triples, tc)
+  end
+  
+  describe "parsing rdf files" do
+    def test_file(filepath, uri = nil)
+      n3_string = File.read(filepath)
+      graph = @parser.parse(n3_string, uri, :strict => true)
+      ntriples = graph.to_ntriples
+      ntriples.gsub!(/_:bn\d+/, '_:node1')
+      ntriples = ntriples.split("\n").sort.join("\n")
+
+      nt_string = File.read(filepath.sub('.rdf', '.nt'))
+      nt_string = nt_string.split("\n").sort.join("\n")
+
+      ntriples.should == nt_string
+    end
+
+    before(:all) do
+      @rdf_dir = File.join(File.dirname(__FILE__), '..', 'test', 'rdf_tests')
+    end
+
+    it "should parse Coldplay's BBC Music profile" do
+      gid = 'cc197bad-dc9c-440d-a5b5-d52ba2e14234'
+      file = File.join(@rdf_dir, "#{gid}.rdf")
+      test_file(file, "http://www.bbc.co.uk/music/artists/#{gid}")
+    end
+
+    # it "should parse xml literal test" do
+    #   file = File.join(@rdf_dir, "xml-literal-mixed.rdf")
+    #   test_file(file)
+    # end
+  end
+
+  # W3C Test suite from http://www.w3.org/2000/10/rdf-tests/rdfcore/
+  describe "w3c rdfcore tests" do
+    require 'rdfcore_helper'
+    include RdfCoreHelper
+    
+    def self.positive_tests
+      RdfCoreHelper::TestCase.positive_parser_tests rescue []
+    end
+
+    def self.negative_tests
+      []#RdfCoreHelper::TestCase.negative_parser_tests rescue []
+    end
+    
+    # Negative parser tests should raise errors.
+    describe "positive parser tests" do
+      positive_tests.each do |t|
+        #next unless t.about.uri.to_s =~ /rdfms-rdf-names-use/
+        #next unless t.name =~ /11/
+        #puts t.inspect
+        specify "test #{t.about.uri.to_s}" do
+          t.run_test do |rdf_string, parser|
+            parser.parse(rdf_string, t.about.uri.to_s, :strict => true, :debug => [])
+          end
+        end
+      end
+    end
+    
+    describe "negative parser tests" do
+      negative_tests.each do |t|
+        #next unless t.about.uri.to_s =~ /rdfms-empty-property-elements/
+        #next unless t.name =~ /1/
+        #puts t.inspect
+        specify "test #{t.about.uri.to_s}" do
+          t.run_test do |rdf_string, parser|
+            lambda do
+              parser.parse(rdf_string, t.about.uri.to_s, :strict => true, :debug => [])
+              parser.graph.should be_equivalent_graph("", t)
+            end.should raise_error(RdfException)
+          end
+        end
+      end
+    end
+  end
+end
+