lib/grom/reader.rb in grom-0.3.5 vs lib/grom/reader.rb in grom-0.3.6

@@ -1,33 +1,41 @@
 module Grom
+  # Reads n-triple data and passes it to a Grom::Builder instance to create objects
+  #
+  # @since 0.1.0
+  # @attr_reader [String] data n-triple data.
+  # @attr_reader [Hash] statements_by_subject statements grouped by subject.
+  # @attr_reader [Hash] edges_by_subject subjects connected to objects which are uris via their predicates.
+  # @attr_reader [Array] objects Grom::Node objects generated from n-triple data.
   class Reader
-    attr_reader :data, :statements_by_subject, :subjects_by_type, :edges_by_subject, :objects
+    attr_reader :data, :statements_by_subject, :edges_by_subject, :objects
 
+    # @param [String] data n-triple data.
     def initialize(data)
       @data = data
 
       read_data
 
       @objects = Grom::Builder.new(self).objects
     end
 
+    # Reads the n-triple data and separates the statements by subject.
+    #
+    # @return [Grom::Reader] an instance of self.
     def read_data
-      # Reset all our hashes just in case
-      @statements_by_subject  = {}
+      @statements_by_subject = {}
 
-      # TODO: Find a better name for this hash!
       @edges_by_subject = {}
 
       RDF::NTriples::Reader.new(@data) do |reader|
         reader.each_statement do |statement|
           subject = statement.subject.to_s
 
-          # TODO: Use Ruby key value syntax in below method.
           Grom::Helper.lazy_array_insert(@statements_by_subject, subject, statement)
 
           predicate = statement.predicate.to_s
 
-          if (statement.object =~ URI.regexp) == 0 && predicate != RDF.type.to_s
+          if statement.object.uri? && predicate != RDF.type.to_s
             predicate = Grom::Helper.get_id(predicate)
             @edges_by_subject[subject] ||= {}
             @edges_by_subject[subject][predicate] ||= []
             @edges_by_subject[subject][predicate] << statement.object.to_s
           end