rdf-n3 0.2.2 → 0.2.3

data/History.txt

@@ -1,3 +1,10 @@
+=== 0.2.3
+* In Writer, set @base_uri not @base, as :base_uri is an attribute.
+* Relativize URLs without matching as regexp.
+* Allow mixed case literal languages.
+* Improve N3 Unicode support for Ruby 1.9
+* Improve Turtle/N3 Writer to use unescaped and qname'd values
+
 === 0.2.2
 * Ruby 1.9.2 compatibility
 * Added script/tc to run test cases

data/README.rdoc

@@ -17,6 +17,9 @@ Install with 'gem install rdf-n3'
 
 Based on original parser by Patrick Sinclair (metade)
 
+== Limitations:
+* Full support of Unicode input requires Ruby version 1.9 or greater.
+
 == Usage:
 Instantiate a reader from a local file:
 

data/Rakefile

@@ -15,7 +15,7 @@ begin
     gemspec.authors = ["Gregg Kellogg"]
     gemspec.add_dependency('rdf', '>= 0.2.1')
     gemspec.add_dependency('treetop',  '>= 1.4.0')
-    gemspec.add_development_dependency('rspec')
+    gemspec.add_development_dependency('rspec', '= 1.3.0')
     gemspec.add_development_dependency('rdf-spec', '>= 0.2.1')
     gemspec.add_development_dependency('rdf-rdfxml', '>= 0.2.1')
     gemspec.add_development_dependency('rdf-isomorphic')
@@ -51,6 +51,12 @@ YARD::Rake::YardocTask.new do |t|
   t.files   = %w(lib/**/*.rb README.rdoc History.txt AUTHORS)   # optional
 end
 
+desc "Update N3 grammar"
+task :grammar do
+  sh "tt -o lib/rdf/n3/reader/n3_grammar.rb lib/rdf/n3/reader/n3_grammar.treetop"
+  sh "tt -o lib/rdf/n3/reader/n3_grammar_18.rb lib/rdf/n3/reader/n3_grammar_18.treetop"
+end
+
 desc "Generate test manifest yaml"
 namespace :spec do
   task :prepare do

data/VERSION

@@ -1 +1 @@
-0.2.2
+0.2.3

data/lib/rdf/n3.rb

@@ -31,5 +31,8 @@ module RDF
     autoload :Reader,  'rdf/n3/reader'
     autoload :VERSION, 'rdf/n3/version'
     autoload :Writer,  'rdf/n3/writer'
+    
+    def self.debug?; @debug; end
+    def self.debug=(value); @debug = value; end
   end
 end

data/lib/rdf/n3/reader.rb

@@ -1,6 +1,12 @@
 require 'treetop'
 
-Treetop.load(File.join(File.dirname(__FILE__), "reader", "n3_grammar"))
+if defined?(::Encoding)
+  # load full grammar
+  Treetop.load(File.join(File.dirname(__FILE__), "reader", "n3_grammar"))
+else
+  # load 1.8 grammar, doesn't include U00010000-\U000effff
+  Treetop.load(File.join(File.dirname(__FILE__), "reader", "n3_grammar_18"))
+end
 
 module RDF::N3
   ##
@@ -69,7 +75,7 @@ module RDF::N3
       parser = N3GrammerParser.new
       document = parser.parse(@doc)
       unless document
-        puts parser.inspect if $DEBUG
+        puts parser.inspect if ::RDF::N3::debug?
         reason = parser.failure_reason
         raise RDF::ReaderError, reason
       end
@@ -104,7 +110,7 @@ module RDF::N3
     # @param [XML Node, any] node:: XML Node or string for showing context
     # @param [String] message::
     def add_debug(node, message)
-      puts "#{node}: #{message}" if $DEBUG
+      puts "#{node}: #{message}" if ::RDF::N3::debug?
       @debug << "#{node}: #{message}" if @debug.is_a?(Array)
     end
 
@@ -412,7 +418,6 @@ module RDF::N3
       localname = expression.localname.text_value if expression.respond_to?(:localname)
       localname ||= (expression.respond_to?(:text_value) ? expression.text_value : expression).to_s.sub(/^:/, "")
       localname = nil if localname.empty? # In N3/Turtle "_:" is not named
-      escaped_localname = RDF::NTriples.escape(localname.to_s)
 
       if expression.respond_to?(:info)
         add_debug(*expression.info("build_uri(#{prefix.inspect}, #{localname.inspect})"))
@@ -421,23 +426,23 @@ module RDF::N3
       end
 
       uri = if @uri_mappings[prefix]
-        add_debug(*expression.info("build_uri: (ns): #{@uri_mappings[prefix]}, #{escaped_localname}")) if expression.respond_to?(:info)
-        ns(prefix, RDF::NTriples.escape(localname.to_s))
+        add_debug(*expression.info("build_uri: (ns): #{@uri_mappings[prefix]}, #{localname}")) if expression.respond_to?(:info)
+        ns(prefix, localname.to_s)
       elsif prefix == '_'
         add_debug(*expression.info("build_uri: (bnode)")) if expression.respond_to?(:info)
         bnode(localname)
       elsif prefix == "rdf"
         add_debug(*expression.info("build_uri: (rdf)")) if expression.respond_to?(:info)
         # A special case
-        RDF::RDF[escaped_localname]
+        RDF::RDF[localname.to_s]
       elsif prefix == "xsd"
         add_debug(*expression.info("build_uri: (xsd)")) if expression.respond_to?(:info)
         # A special case
-        RDF::XSD[escaped_localname]
+        RDF::XSD[localname.to_s]
       else
         add_debug(*expression.info("build_uri: (default_ns)")) if expression.respond_to?(:info)
         @default_ns ||= uri("#{@uri}#", nil)
-        ns(nil, escaped_localname)
+        ns(nil, localname.to_s)
       end
       add_debug(*expression.info("build_uri: #{uri.inspect}")) if expression.respond_to?(:info)
       uri

data/lib/rdf/n3/reader/n3_grammar.rb

@@ -27,10 +27,17 @@ module N3Grammer
     r0
   end
 
-  def _nt_alpha
+  module Barename0
+    def barename_start
+      elements[0]
+    end
+
+  end
+
+  def _nt_barename
     start_index = index
-    if node_cache[:alpha].has_key?(index)
-      cached = node_cache[:alpha][index]
+    if node_cache[:barename].has_key?(index)
+      cached = node_cache[:barename][index]
       if cached
         cached = SyntaxNode.new(input, index...(index + 1)) if cached == true
         @index = cached.interval.end
@@ -38,22 +45,62 @@ module N3Grammer
       return cached
     end
 
-    if has_terminal?('\G[A-Z_a-z\\p{Alpha}]', true, index)
+    i0, s0 = index, []
+    r1 = _nt_barename_start
+    s0 << r1
+    if r1
+      s2, i2 = [], index
+      loop do
+        r3 = _nt_barename_tail
+        if r3
+          s2 << r3
+        else
+          break
+        end
+      end
+      r2 = instantiate_node(SyntaxNode,input, i2...index, s2)
+      s0 << r2
+    end
+    if s0.last
+      r0 = instantiate_node(SyntaxNode,input, i0...index, s0)
+      r0.extend(Barename0)
+    else
+      @index = i0
+      r0 = nil
+    end
+
+    node_cache[:barename][start_index] = r0
+
+    r0
+  end
+
+  def _nt_barename_start
+    start_index = index
+    if node_cache[:barename_start].has_key?(index)
+      cached = node_cache[:barename_start][index]
+      if cached
+        cached = SyntaxNode.new(input, index...(index + 1)) if cached == true
+        @index = cached.interval.end
+      end
+      return cached
+    end
+
+    if has_terminal?('\G[A-Z_a-z\\u00c0-\\u00d6\\u00d8-\\u00f6\\u00f8-\\u02ff\\u0370-\\u037d\\u037f-\\u1fff\\u200c-\\u200d\\u2070-\\u218f\\u2c00-\\u2fef\\u3001-\\ud7ff\\uf900-\\ufdcf\\ufdf0-\\ufffd\\u{10000}-\\u{effff}]', true, index)
       r0 = instantiate_node(SyntaxNode,input, index...(index + 1))
       @index += 1
     else
       r0 = nil
     end
 
-    node_cache[:alpha][start_index] = r0
+    node_cache[:barename_start][start_index] = r0
 
     r0
   end
 
-  def _nt_alphanumeric
+  def _nt_barename_tail
     start_index = index
-    if node_cache[:alphanumeric].has_key?(index)
-      cached = node_cache[:alphanumeric][index]
+    if node_cache[:barename_tail].has_key?(index)
+      cached = node_cache[:barename_tail][index]
       if cached
         cached = SyntaxNode.new(input, index...(index + 1)) if cached == true
         @index = cached.interval.end
@@ -62,11 +109,11 @@ module N3Grammer
     end
 
     i0 = index
-    r1 = _nt_alpha
+    r1 = _nt_barename_start
     if r1
       r0 = r1
     else
-      if has_terminal?('\G[0-9]', true, index)
+      if has_terminal?('\G[-0-9\\u00b7\\u0300-\\u036f\\u203f-\\u2040\\\\]', true, index)
         r2 = true
         @index += 1
       else
@@ -80,7 +127,93 @@ module N3Grammer
       end
     end
 
-    node_cache[:alphanumeric][start_index] = r0
+    node_cache[:barename_tail][start_index] = r0
+
+    r0
+  end
+
+  def _nt_hexdigit
+    start_index = index
+    if node_cache[:hexdigit].has_key?(index)
+      cached = node_cache[:hexdigit][index]
+      if cached
+        cached = SyntaxNode.new(input, index...(index + 1)) if cached == true
+        @index = cached.interval.end
+      end
+      return cached
+    end
+
+    if has_terminal?('\G[0-9a-fA-F]', true, index)
+      r0 = instantiate_node(SyntaxNode,input, index...(index + 1))
+      @index += 1
+    else
+      r0 = nil
+    end
+
+    node_cache[:hexdigit][start_index] = r0
+
+    r0
+  end
+
+  module Integer0
+  end
+
+  def _nt_integer
+    start_index = index
+    if node_cache[:integer].has_key?(index)
+      cached = node_cache[:integer][index]
+      if cached
+        cached = SyntaxNode.new(input, index...(index + 1)) if cached == true
+        @index = cached.interval.end
+      end
+      return cached
+    end
+
+    i0, s0 = index, []
+    if has_terminal?('\G[+-]', true, index)
+      r2 = true
+      @index += 1
+    else
+      r2 = nil
+    end
+    if r2
+      r1 = r2
+    else
+      r1 = instantiate_node(SyntaxNode,input, index...index)
+    end
+    s0 << r1
+    if r1
+      s3, i3 = [], index
+      loop do
+        if has_terminal?('\G[0-9]', true, index)
+          r4 = true
+          @index += 1
+        else
+          r4 = nil
+        end
+        if r4
+          s3 << r4
+        else
+          break
+        end
+      end
+      if s3.empty?
+        @index = i3
+        r3 = nil
+      else
+        r3 = instantiate_node(SyntaxNode,input, i3...index, s3)
+      end
+      s0 << r3
+    end
+    if s0.last
+      r0 = instantiate_node(SyntaxNode,input, i0...index, s0)
+      r0.extend(Integer0)
+    else
+      @index = i0
+      r0 = nil
+    end
+
+    node_cache[:integer][start_index] = r0
 
     r0
   end
@@ -1126,157 +1259,6 @@ module N3Grammer
     r0
   end
 
-  module Barename0
-    def alpha
-      elements[0]
-    end
-
-  end
-
-  def _nt_barename
-    start_index = index
-    if node_cache[:barename].has_key?(index)
-      cached = node_cache[:barename][index]
-      if cached
-        cached = SyntaxNode.new(input, index...(index + 1)) if cached == true
-        @index = cached.interval.end
-      end
-      return cached
-    end
-
-    i0, s0 = index, []
-    r1 = _nt_alpha
-    s0 << r1
-    if r1
-      s2, i2 = [], index
-      loop do
-        i3 = index
-        r4 = _nt_alphanumeric
-        if r4
-          r3 = r4
-        else
-          if has_terminal?('-', false, index)
-            r5 = instantiate_node(SyntaxNode,input, index...(index + 1))
-            @index += 1
-          else
-            terminal_parse_failure('-')
-            r5 = nil
-          end
-          if r5
-            r3 = r5
-          else
-            @index = i3
-            r3 = nil
-          end
-        end
-        if r3
-          s2 << r3
-        else
-          break
-        end
-      end
-      r2 = instantiate_node(SyntaxNode,input, i2...index, s2)
-      s0 << r2
-    end
-    if s0.last
-      r0 = instantiate_node(SyntaxNode,input, i0...index, s0)
-      r0.extend(Barename0)
-    else
-      @index = i0
-      r0 = nil
-    end
-
-    node_cache[:barename][start_index] = r0
-
-    r0
-  end
-
-  def _nt_hexdigit
-    start_index = index
-    if node_cache[:hexdigit].has_key?(index)
-      cached = node_cache[:hexdigit][index]
-      if cached
-        cached = SyntaxNode.new(input, index...(index + 1)) if cached == true
-        @index = cached.interval.end
-      end
-      return cached
-    end
-
-    if has_terminal?('\G[0-9a-fA-F]', true, index)
-      r0 = instantiate_node(SyntaxNode,input, index...(index + 1))
-      @index += 1
-    else
-      r0 = nil
-    end
-
-    node_cache[:hexdigit][start_index] = r0
-
-    r0
-  end
-
-  module Integer0
-  end
-
-  def _nt_integer
-    start_index = index
-    if node_cache[:integer].has_key?(index)
-      cached = node_cache[:integer][index]
-      if cached
-        cached = SyntaxNode.new(input, index...(index + 1)) if cached == true
-        @index = cached.interval.end
-      end
-      return cached
-    end
-
-    i0, s0 = index, []
-    if has_terminal?('\G[+-]', true, index)
-      r2 = true
-      @index += 1
-    else
-      r2 = nil
-    end
-    if r2
-      r1 = r2
-    else
-      r1 = instantiate_node(SyntaxNode,input, index...index)
-    end
-    s0 << r1
-    if r1
-      s3, i3 = [], index
-      loop do
-        if has_terminal?('\G[0-9]', true, index)
-          r4 = true
-          @index += 1
-        else
-          r4 = nil
-        end
-        if r4
-          s3 << r4
-        else
-          break
-        end
-      end
-      if s3.empty?
-        @index = i3
-        r3 = nil
-      else
-        r3 = instantiate_node(SyntaxNode,input, i3...index, s3)
-      end
-      s0 << r3
-    end
-    if s0.last
-      r0 = instantiate_node(SyntaxNode,input, i0...index, s0)
-      r0.extend(Integer0)
-    else
-      @index = i0
-      r0 = nil
-    end
-
-    node_cache[:integer][start_index] = r0
-
-    r0
-  end
-
   module Language0
   end
 
@@ -1297,7 +1279,7 @@ module N3Grammer
     i0, s0 = index, []
     s1, i1 = [], index
     loop do
-      if has_terminal?('\G[a-z]', true, index)
+      if has_terminal?('\G[a-zA-Z]', true, index)
         r2 = true
         @index += 1
       else
@@ -1331,7 +1313,7 @@ module N3Grammer
         if r5
           s6, i6 = [], index
           loop do
-            if has_terminal?('\G[a-z0-9]', true, index)
+            if has_terminal?('\G[a-zA-Z0-9]', true, index)
               r7 = true
               @index += 1
             else
@@ -1493,13 +1475,6 @@ module N3Grammer
     r0
   end
 
-  module Localname0
-    def alpha
-      elements[0]
-    end
-
-  end
-
   def _nt_localname
     start_index = index
     if node_cache[:localname].has_key?(index)
@@ -1511,60 +1486,13 @@ module N3Grammer
       return cached
     end
 
-    i0, s0 = index, []
-    r1 = _nt_alpha
-    s0 << r1
-    if r1
-      s2, i2 = [], index
-      loop do
-        i3 = index
-        r4 = _nt_alphanumeric
-        if r4
-          r3 = r4
-        else
-          if has_terminal?('-', false, index)
-            r5 = instantiate_node(SyntaxNode,input, index...(index + 1))
-            @index += 1
-          else
-            terminal_parse_failure('-')
-            r5 = nil
-          end
-          if r5
-            r3 = r5
-          else
-            @index = i3
-            r3 = nil
-          end
-        end
-        if r3
-          s2 << r3
-        else
-          break
-        end
-      end
-      r2 = instantiate_node(SyntaxNode,input, i2...index, s2)
-      s0 << r2
-    end
-    if s0.last
-      r0 = instantiate_node(SyntaxNode,input, i0...index, s0)
-      r0.extend(Localname0)
-    else
-      @index = i0
-      r0 = nil
-    end
+    r0 = _nt_barename
 
     node_cache[:localname][start_index] = r0
 
     r0
   end
 
-  module Nprefix0
-    def alpha
-      elements[0]
-    end
-
-  end
-
   def _nt_nprefix
     start_index = index
     if node_cache[:nprefix].has_key?(index)
@@ -1576,47 +1504,7 @@ module N3Grammer
       return cached
     end
 
-    i0, s0 = index, []
-    r1 = _nt_alpha
-    s0 << r1
-    if r1
-      s2, i2 = [], index
-      loop do
-        i3 = index
-        r4 = _nt_alphanumeric
-        if r4
-          r3 = r4
-        else
-          if has_terminal?('-', false, index)
-            r5 = instantiate_node(SyntaxNode,input, index...(index + 1))
-            @index += 1
-          else
-            terminal_parse_failure('-')
-            r5 = nil
-          end
-          if r5
-            r3 = r5
-          else
-            @index = i3
-            r3 = nil
-          end
-        end
-        if r3
-          s2 << r3
-        else
-          break
-        end
-      end
-      r2 = instantiate_node(SyntaxNode,input, i2...index, s2)
-      s0 << r2
-    end
-    if s0.last
-      r0 = instantiate_node(SyntaxNode,input, i0...index, s0)
-      r0.extend(Nprefix0)
-    else
-      @index = i0
-      r0 = nil
-    end
+    r0 = _nt_barename
 
     node_cache[:nprefix][start_index] = r0