adding parser context and replace_entities accessor. closes #76

CHANGELOG.rdoc

@@ -17,6 +17,7 @@
   * Added XML::Node#create_external_subset
   * Added XML::Node#create_internal_subset
   * XML Builder can append raw strings (GH #141, patch from dudleyf)
+  * XML::SAX::ParserContext added
 
 * Bugfixes
 

ext/nokogiri/xml_sax_parser_context.c

@@ -13,6 +13,12 @@ static void deallocate(xmlParserCtxtPtr ctxt)
   NOKOGIRI_DEBUG_END(handler);
 }
 
+/*
+ * call-seq:
+ *  parse_io(io, encoding)
+ *
+ * Parse +io+ object with +encoding+
+ */
 static VALUE parse_io(VALUE klass, VALUE io, VALUE encoding)
 {
   xmlCharEncoding enc = (xmlCharEncoding)NUM2INT(encoding); 
@@ -29,12 +35,24 @@ static VALUE parse_io(VALUE klass, VALUE io, VALUE encoding)
   return Data_Wrap_Struct(klass, NULL, deallocate, ctxt);
 }
 
+/*
+ * call-seq:
+ *  parse_file(filename)
+ *
+ * Parse file given +filename+
+ */
 static VALUE parse_file(VALUE klass, VALUE filename)
 {
   xmlParserCtxtPtr ctxt = xmlCreateFileParserCtxt(StringValuePtr(filename));
   return Data_Wrap_Struct(klass, NULL, deallocate, ctxt);
 }
 
+/*
+ * call-seq:
+ *  parse_memory(data)
+ *
+ * Parse the XML stored in memory in +data+
+ */
 static VALUE parse_memory(VALUE klass, VALUE data)
 {
   if(Qnil == data) rb_raise(rb_eArgError, "data cannot be nil");
@@ -49,6 +67,12 @@ static VALUE parse_memory(VALUE klass, VALUE data)
   return Data_Wrap_Struct(klass, NULL, deallocate, ctxt);
 }
 
+/*
+ * call-seq:
+ *  parse_with(sax_handler)
+ *
+ * Use +sax_handler+ and parse the current document
+ */
 static VALUE parse_with(VALUE self, VALUE sax_handler)
 {
   if(!rb_obj_is_kind_of(sax_handler, cNokogiriXmlSaxParser))
@@ -74,6 +98,44 @@ static VALUE parse_with(VALUE self, VALUE sax_handler)
   NOKOGIRI_SAX_TUPLE_DESTROY(ctxt->userData);
 }
 
+/*
+ * call-seq:
+ *  replace_entities=(boolean)
+ *
+ * Should this parser replace entities?  & will get converted to '&' if
+ * set to true
+ */
+static VALUE set_replace_entities(VALUE self, VALUE value)
+{
+  xmlParserCtxtPtr ctxt;
+  Data_Get_Struct(self, xmlParserCtxt, ctxt);
+
+  if(Qfalse == value)
+    ctxt->replaceEntities = 0;
+  else
+    ctxt->replaceEntities = 1;
+
+  return value;
+}
+
+/*
+ * call-seq:
+ *  replace_entities
+ *
+ * Should this parser replace entities?  & will get converted to '&' if
+ * set to true
+ */
+static VALUE get_replace_entities(VALUE self)
+{
+  xmlParserCtxtPtr ctxt;
+  Data_Get_Struct(self, xmlParserCtxt, ctxt);
+
+  if(0 == ctxt->replaceEntities)
+    return Qfalse;
+  else
+    return Qtrue;
+}
+
 void init_xml_sax_parser_context()
 {
   VALUE nokogiri  = rb_define_module("Nokogiri");
@@ -88,4 +150,6 @@ void init_xml_sax_parser_context()
   rb_define_singleton_method(klass, "file", parse_file, 1);
 
   rb_define_method(klass, "parse_with", parse_with, 1);
+  rb_define_method(klass, "replace_entities=", set_replace_entities, 1);
+  rb_define_method(klass, "replace_entities", get_replace_entities, 0);
 }

lib/nokogiri/xml/sax/parser.rb

@@ -76,19 +76,21 @@ def initialize doc = Nokogiri::XML::SAX::Document.new, encoding = 'UTF-8'
         ###
         # Parse given +thing+ which may be a string containing xml, or an
         # IO object.
-        def parse thing
+        def parse thing, &block
           if thing.respond_to?(:read) && thing.respond_to?(:close)
-            parse_io(thing)
+            parse_io(thing, &block)
           else
-            parse_memory(thing)
+            parse_memory(thing, &block)
           end
         end
 
         ###
         # Parse given +io+
         def parse_io io, encoding = 'ASCII'
           @encoding = encoding
-          ParserContext.io(io, ENCODINGS[encoding]).parse_with self
+          ctx = ParserContext.io(io, ENCODINGS[encoding])
+          yield ctx if block_given?
+          ctx.parse_with self
         end
 
         ###
@@ -97,11 +99,15 @@ def parse_file filename
           raise ArgumentError unless filename
           raise Errno::ENOENT unless File.exists?(filename)
           raise Errno::EISDIR if File.directory?(filename)
-          ParserContext.file(filename).parse_with self
+          ctx = ParserContext.file filename
+          yield ctx if block_given?
+          ctx.parse_with self
         end
 
         def parse_memory data
-          ParserContext.memory(data).parse_with(self)
+          ctx = ParserContext.memory data
+          yield ctx if block_given?
+          ctx.parse_with self
         end
       end
     end

test/xml/sax/test_parser.rb

@@ -11,6 +11,38 @@ def setup
           @parser = XML::SAX::Parser.new(Doc.new)
         end
 
+        def test_parser_context_yielded_io
+          doc = Doc.new
+          parser = XML::SAX::Parser.new doc
+          xml = "<foo a='&amp;b'/>"
+
+          block_called = false
+          parser.parse(StringIO.new(xml)) { |ctx|
+            block_called = true
+            ctx.replace_entities = true
+          }
+
+          assert block_called
+
+          assert_equal ['a', '&b'], doc.start_elements.first.last
+        end
+
+        def test_parser_context_yielded_in_memory
+          doc = Doc.new
+          parser = XML::SAX::Parser.new doc
+          xml = "<foo a='&amp;b'/>"
+
+          block_called = false
+          parser.parse(xml) { |ctx|
+            block_called = true
+            ctx.replace_entities = true
+          }
+
+          assert block_called
+
+          assert_equal ['a', '&b'], doc.start_elements.first.last
+        end
+
         def test_xml_decl
           {
             ''          => nil,

test/xml/sax/test_parser_context.rb

@@ -6,6 +6,15 @@ module Nokogiri
   module XML
     module SAX
       class TestParserContext < Nokogiri::SAX::TestCase
+        def test_replace_entities
+          pc = ParserContext.new StringIO.new('<root />'), 'UTF-8'
+          pc.replace_entities = false
+          assert_equal false, pc.replace_entities
+
+          pc.replace_entities = true
+          assert_equal true, pc.replace_entities
+        end
+
         def test_from_io
           assert_nothing_raised do
             ParserContext.new StringIO.new('fo'), 'UTF-8'