C14n fix and new parameters

ext/nokogiri/xml_document.c

@@ -428,9 +428,14 @@ static int block_caller(void * ctx, xmlNodePtr _node, xmlNodePtr _parent)
   VALUE parent;
   VALUE ret;
 
-  block  = (VALUE)ctx;
-  node   = Nokogiri_wrap_xml_node(Qnil, _node);
+  if(_node->type == XML_NAMESPACE_DECL){
+    node = Nokogiri_wrap_xml_namespace(_parent->doc, (xmlNsPtr) _node);
+  }
+  else{
+    node   = Nokogiri_wrap_xml_node(Qnil, _node);
+  }
   parent = _parent ? Nokogiri_wrap_xml_node(Qnil, _parent) : Qnil;
+  block  = (VALUE)ctx;
 
   ret = rb_funcall(block, rb_intern("call"), 2, node, parent);
 
@@ -440,16 +445,25 @@ static int block_caller(void * ctx, xmlNodePtr _node, xmlNodePtr _parent)
 }
 
 /* call-seq:
- *  doc.canonicalize
- *  doc.canonicalize { |node, parent| ... }
+ *  doc.canonicalize(mode=XML_C14N_1_0,inclusive_namespaces=nil,with_comments=false)
+ *  doc.canonicalize { |obj, parent| ... }
  *
  * Canonicalize a document and return the results.  Takes an optional block
- * that takes two parameters the +node+ and that node's +parent+.  The block
- * must return a non-nil, non-false value if the +node+ passed in should be
- * included in the canonicalized document.
+ * that takes two parameters: the +obj+ and that node's +parent+.  
+ * The  +obj+ will be either a Nokogiri::XML::Node, or a Nokogiri::XML::Namespace
+ * The block must return a non-nil, non-false value if the +obj+ passed in 
+ * should be included in the canonicalized document.
  */
-static VALUE canonicalize(VALUE self)
+static VALUE canonicalize(int argc, VALUE* argv, VALUE self)
 {
+  VALUE mode;
+  VALUE incl_ns;
+  VALUE with_comments;
+  xmlChar **ns;
+  long ns_len, i;
+
+  rb_scan_args(argc, argv, "03", &mode, &incl_ns, &with_comments);
+
   xmlDocPtr doc;
   xmlOutputBufferPtr buf;
   xmlC14NIsVisibleCallback cb = NULL;
@@ -473,7 +487,25 @@ static VALUE canonicalize(VALUE self)
     ctx = (void *)rb_block_proc();
   }
 
-  xmlC14NExecute(doc, cb, ctx, 0, NULL, 0, buf);
+  if(NIL_P(incl_ns)){
+    ns = NULL;
+  }
+  else{
+    ns_len = RARRAY_LEN(incl_ns);
+    ns = calloc((size_t)ns_len+1, sizeof(xmlChar *));
+    for (i = 0 ; i < ns_len ; i++) {
+      VALUE entry = rb_ary_entry(incl_ns, i);
+      const char * ptr = StringValuePtr(entry);
+      ns[i] = (xmlChar*) ptr;
+    }
+  }
+
+
+  xmlC14NExecute(doc, cb, ctx, 
+    (int)      (NIL_P(mode)        ? 0 : NUM2INT(mode)), 
+    ns,
+    (int)      (NIL_P(with_comments)        ? 0 : 1),
+    buf);
 
   xmlOutputBufferClose(buf);
 
@@ -503,7 +535,7 @@ void init_xml_document()
   rb_define_method(klass, "encoding", encoding, 0);
   rb_define_method(klass, "encoding=", set_encoding, 1);
   rb_define_method(klass, "version", version, 0);
-  rb_define_method(klass, "canonicalize", canonicalize, 0);
+  rb_define_method(klass, "canonicalize", canonicalize, -1);
   rb_define_method(klass, "dup", duplicate_node, -1);
   rb_define_method(klass, "url", url, 0);
   rb_define_method(klass, "create_entity", create_entity, -1);

lib/nokogiri/xml.rb

@@ -35,6 +35,12 @@ def XML thing, url = nil, encoding = nil, options = XML::ParseOptions::DEFAULT_X
   end
 
   module XML
+    # Original C14N 1.0 spec canonicalization
+    XML_C14N_1_0 =       0
+    # Exclusive C14N 1.0 spec canonicalization
+    XML_C14N_EXCLUSIVE_1_0 =     1
+    # C14N 1.1 spec canonicalization
+    XML_C14N_1_1 = 2
     class << self
       ###
       # Parse an XML document using the Nokogiri::XML::Reader API.  See

lib/nokogiri/xml/node.rb

@@ -879,6 +879,14 @@ def do_xinclude options = XML::ParseOptions::DEFAULT_XML, &block
         process_xincludes(options.to_i)
       end
 
+      def canonicalize(mode=XML::XML_C14N_1_0,inclusive_namespaces=nil,with_comments=false)
+        c14n_root = self
+        document.canonicalize(mode, inclusive_namespaces, with_comments) do |node, parent|
+          tn = node.is_a?(XML::Node) ? node : parent
+          tn == c14n_root || tn.ancestors.include?(c14n_root)
+        end
+      end
+
       private
 
       def extract_params params # :nodoc:

test/xml/test_c14n.rb

@@ -25,6 +25,9 @@ def test_3_1
         c14n = doc.canonicalize
         assert_no_match(/version=/, c14n)
         assert_match(/Hello, world/, c14n)
+        assert_no_match(/Comment/, c14n)
+        c14n = doc.canonicalize(nil, nil, true)
+        assert_match(/Comment/, c14n)
       end
 
       def test_exclude_block_params
@@ -81,6 +84,60 @@ def test_exclude_block_object
         end
         assert_equal xml, c14n
       end
+
+      def test_c14n_node
+        xml = '<a><b><c></c></b></a>'
+        doc = Nokogiri.XML xml
+        c14n = doc.at_xpath('//b').canonicalize
+        assert_equal '<b><c></c></b>', c14n
+      end
+
+      def test_c14_modes
+        # http://www.w3.org/TR/xml-exc-c14n/#sec-Enveloping
+
+        doc1 = Nokogiri.XML <<-eoxml
+<n0:local xmlns:n0="http://foobar.org" xmlns:n3="ftp://example.org">
+  <n1:elem2 xmlns:n1="http://example.net" xml:lang="en">
+    <n3:stuff xmlns:n3="ftp://example.org"/>
+  </n1:elem2>
+</n0:local>        
+        eoxml
+        doc2 = Nokogiri.XML <<-eoxml
+<n2:pdu xmlns:n1="http://example.com"
+           xmlns:n2="http://foo.example"
+           xml:lang="fr"
+           xml:space="retain">
+  <n1:elem2 xmlns:n1="http://example.net" xml:lang="en">
+    <n3:stuff xmlns:n3="ftp://example.org"/>
+  </n1:elem2>
+</n2:pdu>        
+        eoxml
+
+        c14n = doc1.at_xpath('//n1:elem2', {'n1' => 'http://example.net'}).canonicalize
+        assert_equal '<n1:elem2 xmlns:n0="http://foobar.org" xmlns:n1="http://example.net" xmlns:n3="ftp://example.org" xml:lang="en">
+    <n3:stuff></n3:stuff>
+  </n1:elem2>', c14n
+        c14n = doc2.at_xpath('//n1:elem2', {'n1' => 'http://example.net'}).canonicalize
+        assert_equal '<n1:elem2 xmlns:n1="http://example.net" xmlns:n2="http://foo.example" xml:lang="en" xml:space="retain">
+    <n3:stuff xmlns:n3="ftp://example.org"></n3:stuff>
+  </n1:elem2>', c14n
+
+        excl_c14n = '<n1:elem2 xmlns:n1="http://example.net" xml:lang="en">
+    <n3:stuff xmlns:n3="ftp://example.org"></n3:stuff>
+  </n1:elem2>'
+        c14n = doc1.at_xpath('//n1:elem2', {'n1' => 'http://example.net'}).canonicalize(XML::XML_C14N_EXCLUSIVE_1_0)
+        assert_equal excl_c14n, c14n
+        c14n = doc2.at_xpath('//n1:elem2', {'n1' => 'http://example.net'}).canonicalize(XML::XML_C14N_EXCLUSIVE_1_0)
+        assert_equal excl_c14n, c14n
+
+        c14n = doc2.at_xpath('//n1:elem2', {'n1' => 'http://example.net'}).canonicalize(XML::XML_C14N_EXCLUSIVE_1_0, ['n2'])
+        assert_equal '<n1:elem2 xmlns:n1="http://example.net" xmlns:n2="http://foo.example" xml:lang="en">
+    <n3:stuff xmlns:n3="ftp://example.org"></n3:stuff>
+  </n1:elem2>', c14n
+
+      end
+
+
     end
   end
 end