RubyGems - nokogiri - Versions diffs - 1.6.7.2 → 1.6.8.rc1 - Mend

nokogiri 1.6.7.2 → 1.6.8.rc1

Potentially problematic release.

This version of nokogiri might be problematic. Click here for more details.

Files changed (74) hide show

checksums.yaml +4 -4
data/.travis.yml +12 -9
data/CHANGELOG.ja.rdoc +18 -0
data/CHANGELOG.rdoc +12 -7
data/CONTRIBUTING.md +42 -0
data/Gemfile +1 -1
data/Manifest.txt +6 -0
data/README.md +1 -1
data/Rakefile +1 -1
data/bin/nokogiri +2 -2
data/dependencies.yml +1 -1
data/ext/nokogiri/extconf.rb +3 -3
data/ext/nokogiri/nokogiri.c +0 -7
data/ext/nokogiri/nokogiri.h +1 -34
data/ext/nokogiri/xml_document.c +2 -4
data/ext/nokogiri/xml_namespace.c +56 -17
data/ext/nokogiri/xml_node.c +12 -36
data/ext/nokogiri/xml_node_set.c +169 -143
data/ext/nokogiri/xml_node_set.h +3 -4
data/ext/nokogiri/xml_sax_parser.c +2 -5
data/ext/nokogiri/xml_syntax_error.c +0 -4
data/ext/nokogiri/xml_syntax_error.h +0 -1
data/ext/nokogiri/xml_xpath_context.c +9 -18
data/lib/nokogiri.rb +3 -0
data/lib/nokogiri/css/parser.rb +8 -2
data/lib/nokogiri/css/parser.y +7 -2
data/lib/nokogiri/version.rb +1 -1
data/lib/nokogiri/xml/document.rb +7 -1
data/lib/nokogiri/xml/dtd.rb +4 -4
data/lib/nokogiri/xml/node.rb +2 -2
data/ports/archives/libxml2-2.9.3.tar.gz +0 -0
data/test/css/test_parser.rb +7 -1
data/test/files/GH_1042.html +18 -0
data/test/files/namespace_pressure_test.xml +1684 -0
data/test/files/tlm.html +2 -1
data/test/html/sax/test_parser.rb +2 -2
data/test/html/test_document.rb +18 -8
data/test/html/test_document_encoding.rb +46 -54
data/test/html/test_document_fragment.rb +21 -22
data/test/html/test_node.rb +16 -0
data/test/html/test_node_encoding.rb +12 -14
data/test/namespaces/test_namespaces_in_parsed_doc.rb +14 -0
data/test/test_reader.rb +19 -0
data/test/test_xslt_transforms.rb +5 -3
data/test/xml/sax/test_parser.rb +36 -39
data/test/xml/test_document.rb +7 -2
data/test/xml/test_document_encoding.rb +14 -16
data/test/xml/test_dtd_encoding.rb +0 -2
data/test/xml/test_node_encoding.rb +78 -80
data/test/xml/test_reader_encoding.rb +100 -102
data/test/xslt/test_exception_handling.rb +1 -1
metadata +11 -28
data/patches/libxml2/0001-Revert-Missing-initialization-for-the-catalog-module.patch +0 -29
data/patches/libxml2/0002-Fix-missing-entities-after-CVE-2014-3660-fix.patch +0 -31
data/patches/libxml2/0003-Stop-parsing-on-entities-boundaries-errors.patch +0 -32
data/patches/libxml2/0004-Cleanup-conditional-section-error-handling.patch +0 -49
data/patches/libxml2/0005-CVE-2015-1819-Enforce-the-reader-to-run-in-constant-.patch +0 -177
data/patches/libxml2/0006-Another-variation-of-overflow-in-Conditional-section.patch +0 -32
data/patches/libxml2/0007-Fix-an-error-in-previous-Conditional-section-patch.patch +0 -28
data/patches/libxml2/0008-CVE-2015-8035-Fix-XZ-compression-support-loop.patch +0 -31
data/patches/libxml2/0009-Updated-config.guess.patch +0 -397
data/patches/libxml2/0010-Fix-parsering-short-unclosed-comment-uninitialized-access.patch +0 -64
data/patches/libxml2/0011-Avoid-extra-processing-of-MarkupDecl-when-EOF.patch +0 -35
data/patches/libxml2/0012-Avoid-processing-entities-after-encoding-conversion-.patch +0 -83
data/patches/libxml2/0013-CVE-2015-7497-Avoid-an-heap-buffer-overflow-in-xmlDi.patch +0 -33
data/patches/libxml2/0014-CVE-2015-5312-Another-entity-expansion-issue.patch +0 -32
data/patches/libxml2/0015-Add-xmlHaltParser-to-stop-the-parser.patch +0 -81
data/patches/libxml2/0016-Detect-incoherency-on-GROW.patch +0 -36
data/patches/libxml2/0017-CVE-2015-7500-Fix-memory-access-error-due-to-incorre.patch +0 -105
data/patches/libxml2/0018-CVE-2015-8242-Buffer-overead-with-HTML-parser-in-pus.patch +0 -43
data/patches/libxml2/0019-Do-not-print-error-context-when-there-is-none.patch +0 -28
data/patches/libxml2/0020-xmlStopParser-reset-errNo.patch +0 -41
data/patches/libxml2/0021-Reuse-xmlHaltParser-where-it-makes-sense.patch +0 -175
data/ports/archives/libxml2-2.9.2.tar.gz +0 -0

@@ -46,7 +46,7 @@
 .codesnip-container  {border:1px solid #ccc; background:#eee; padding: 5px;margin:10px;}
 </style>
 <link rel="EditURI" type="application/rsd+xml" title="RSD" href="http://tenderlovemaking.com/xmlrpc.php?rsd" />
-<link rel="wlwmanifest" type="application/wlwmanifest+xml" href="http://tenderlovemaking.com/wp-includes/wlwmanifest.xml" />
+<link rel="wlwmanifest" type="application/wlwmanifest+xml" href="http://tenderlovemaking.com/wp-includes/wlwmanifest.xml" />
 <meta name="generator" content="WordPress 2.6" />
 	<link rel="stylesheet" type="text/css" href="http://tenderlovemaking.com/wp-content/plugins/spell_checker/spell_checker.css" />
@@ -826,6 +826,7 @@ page.<span class="me1">body</span> =~ /&lt;textarea<span class="br0">&#91;</span
 </ul>
 </div>
+<div id="abc.123" class='special.character'>Special character div</div>
 <div id="footer">
 A design by <a href="http://blog.geminigeek.com/wordpress-theme">GeminiGeek</a> &bull; Powered by <a href="http://wordpress.org">Wordpress</a><!--&bull; <a href="#">CSS</a> &bull; <a href="#">xHTML 1.0</a>-->
 </div>

data/test/html/sax/test_parser.rb CHANGED

@@ -29,9 +29,9 @@ module Nokogiri
           # Take a look at the comment in test_parse_document to know
           # a possible reason to this difference.
           if Nokogiri.uses_libxml?
-            assert_equal 1110, @parser.document.end_elements.length
+            assert_equal 1111, @parser.document.end_elements.length
           else
-            assert_equal 1119, @parser.document.end_elements.length
+            assert_equal 1120, @parser.document.end_elements.length
           end
         end

data/test/html/test_document.rb CHANGED

@@ -479,6 +479,15 @@ eohtml
         assert_equal 1, found.length
       end
+      def test_find_by_css_with_escaped_characters
+        found_without_escape = @html.css("div[@id='abc.123']")
+        found_by_id = @html.css('#abc\.123')
+        found_by_class = @html.css('.special\.character')
+        assert_equal 1, found_without_escape.length
+        assert_equal found_by_id, found_without_escape
+        assert_equal found_by_class, found_without_escape
+      end
       def test_find_with_function
         assert @html.css("div:awesome() h1", Class.new {
           def awesome divs
@@ -628,19 +637,20 @@ eohtml
       end
       def test_capturing_nonparse_errors_during_node_copy_between_docs
-        skip("JRuby HTML parse errors are different than libxml2's") if Nokogiri.jruby?
-        doc1 = Nokogiri::HTML("<div id='unique'>one</div>")
-        doc2 = Nokogiri::HTML("<div id='unique'>two</div>")
+        # Errors should be emitted while parsing only, and should not change when moving nodes.
+        doc1 = Nokogiri::HTML("<html><body><diva id='unique'>one</diva></body></html>")
+        doc2 = Nokogiri::HTML("<html><body><dive id='unique'>two</dive></body></html>")
         node1 = doc1.at_css("#unique")
         node2 = doc2.at_css("#unique")
-        original_errors = doc1.errors.dup
+        original_errors1 = doc1.errors.dup
+        original_errors2 = doc2.errors.dup
+        assert original_errors1.any?{|e| e.to_s =~ /Tag diva invalid/ }, "it should complain about the tag name"
+        assert original_errors2.any?{|e| e.to_s =~ /Tag dive invalid/ }, "it should complain about the tag name"
         node1.add_child node2
-        assert_equal original_errors.length+1, doc1.errors.length
-        assert_match(/ID unique already defined/, doc1.errors.last.to_s)
+        assert_equal original_errors1, doc1.errors
+        assert_equal original_errors2, doc2.errors
       end
       def test_silencing_nonparse_errors_during_attribute_insertion_1262

data/test/html/test_document_encoding.rb CHANGED

@@ -3,34 +3,33 @@ require "helper"
 module Nokogiri
   module HTML
-    if RUBY_VERSION =~ /^1\.9/
-      class TestDocumentEncoding < Nokogiri::TestCase
-        def test_encoding
-          doc = Nokogiri::HTML File.open(SHIFT_JIS_HTML, 'rb')
+    class TestDocumentEncoding < Nokogiri::TestCase
+      def test_encoding
+        doc = Nokogiri::HTML File.open(SHIFT_JIS_HTML, 'rb')
-          hello = "こんにちは"
+        hello = "こんにちは"
-          assert_match doc.encoding, doc.to_html
-          assert_match hello.encode('Shift_JIS'), doc.to_html
-          assert_equal 'Shift_JIS', doc.to_html.encoding.name
+        assert_match doc.encoding, doc.to_html
+        assert_match hello.encode('Shift_JIS'), doc.to_html
+        assert_equal 'Shift_JIS', doc.to_html.encoding.name
-          assert_match hello, doc.to_html(:encoding => 'UTF-8')
-          assert_match 'UTF-8', doc.to_html(:encoding => 'UTF-8')
-          assert_match 'UTF-8', doc.to_html(:encoding => 'UTF-8').encoding.name
-        end
+        assert_match hello, doc.to_html(:encoding => 'UTF-8')
+        assert_match 'UTF-8', doc.to_html(:encoding => 'UTF-8')
+        assert_match 'UTF-8', doc.to_html(:encoding => 'UTF-8').encoding.name
+      end
-        def test_encoding_without_charset
-          doc = Nokogiri::HTML File.open(SHIFT_JIS_NO_CHARSET, 'r:cp932:cp932').read
+      def test_encoding_without_charset
+        doc = Nokogiri::HTML File.open(SHIFT_JIS_NO_CHARSET, 'r:cp932:cp932').read
-          hello = "こんにちは"
+        hello = "こんにちは"
-          assert_match hello, doc.content
-          assert_match hello, doc.to_html(:encoding => 'UTF-8')
-          assert_match 'UTF-8', doc.to_html(:encoding => 'UTF-8').encoding.name
-        end
+        assert_match hello, doc.content
+        assert_match hello, doc.to_html(:encoding => 'UTF-8')
+        assert_match 'UTF-8', doc.to_html(:encoding => 'UTF-8').encoding.name
+      end
-        def test_default_to_encoding_from_string
-          bad_charset = <<-eohtml
+      def test_default_to_encoding_from_string
+        bad_charset = <<-eohtml
 <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"   "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
 <html>
 <head>
@@ -40,33 +39,33 @@ module Nokogiri
   <a href="http://tenderlovemaking.com/">blah!</a>
 </body>
 </html>
-          eohtml
-          doc = Nokogiri::HTML(bad_charset)
-          assert_equal bad_charset.encoding.name, doc.encoding
+        eohtml
+        doc = Nokogiri::HTML(bad_charset)
+        assert_equal bad_charset.encoding.name, doc.encoding
-          doc = Nokogiri.parse(bad_charset)
-          assert_equal bad_charset.encoding.name, doc.encoding
-        end
+        doc = Nokogiri.parse(bad_charset)
+        assert_equal bad_charset.encoding.name, doc.encoding
+      end
-        def test_encoding_non_utf8
-          orig = '日本語が上手です'
-          bin = Encoding::ASCII_8BIT
-          [Encoding::Shift_JIS, Encoding::EUC_JP].each do |enc|
-            html = <<-eohtml.encode(enc)
+      def test_encoding_non_utf8
+        orig = '日本語が上手です'
+        bin = Encoding::ASCII_8BIT
+        [Encoding::Shift_JIS, Encoding::EUC_JP].each do |enc|
+          html = <<-eohtml.encode(enc)
 <html>
 <meta http-equiv="Content-Type" content="text/html; charset=#{enc.name}">
 <title xml:lang="ja">#{orig}</title></html>
-            eohtml
-            text = Nokogiri::HTML.parse(html).at('title').inner_text
-            assert_equal(
-              orig.encode(enc).force_encoding(bin),
-              text.encode(enc).force_encoding(bin)
-            )
-          end
+          eohtml
+          text = Nokogiri::HTML.parse(html).at('title').inner_text
+          assert_equal(
+            orig.encode(enc).force_encoding(bin),
+            text.encode(enc).force_encoding(bin)
+          )
         end
+      end
-        def test_encoding_with_a_bad_name
-          bad_charset = <<-eohtml
+      def test_encoding_with_a_bad_name
+        bad_charset = <<-eohtml
 <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"   "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
 <html>
 <head>
@@ -76,23 +75,16 @@ module Nokogiri
   <a href="http://tenderlovemaking.com/">blah!</a>
 </body>
 </html>
-          eohtml
-          doc = Nokogiri::HTML(bad_charset, nil, 'askldjfhalsdfjhlkasdfjh')
-          assert_equal ['http://tenderlovemaking.com/'],
-            doc.css('a').map { |a| a['href'] }
-        end
+        eohtml
+        doc = Nokogiri::HTML(bad_charset, nil, 'askldjfhalsdfjhlkasdfjh')
+        assert_equal ['http://tenderlovemaking.com/'],
+          doc.css('a').map { |a| a['href'] }
       end
     end
     class TestDocumentEncodingDetection < Nokogiri::TestCase
-      if IO.respond_to?(:binread)
-        def binread(file)
-          IO.binread(file)
-        end
-      else
-        def binread(file)
-          IO.read(file)
-        end
+      def binread(file)
+        IO.binread(file)
       end
       def binopen(file)

data/test/html/test_document_fragment.rb CHANGED

@@ -9,19 +9,17 @@ module Nokogiri
         @html = Nokogiri::HTML.parse(File.read(HTML_FILE), HTML_FILE)
       end
-      if RUBY_VERSION >= '1.9'
-        def test_inspect_encoding
-          fragment = "<div>こんにちは！</div>".encode('EUC-JP')
-          f = Nokogiri::HTML::DocumentFragment.parse fragment
-          assert_equal "こんにちは！", f.content
-        end
+      def test_inspect_encoding
+        fragment = "<div>こんにちは！</div>".encode('EUC-JP')
+        f = Nokogiri::HTML::DocumentFragment.parse fragment
+        assert_equal "こんにちは！", f.content
+      end
-        def test_html_parse_encoding
-          fragment = "<div>こんにちは！</div>".encode 'EUC-JP'
-          f = Nokogiri::HTML.fragment fragment
-          assert_equal 'EUC-JP', f.document.encoding
-          assert_equal "こんにちは！", f.content
-        end
+      def test_html_parse_encoding
+        fragment = "<div>こんにちは！</div>".encode 'EUC-JP'
+        f = Nokogiri::HTML.fragment fragment
+        assert_equal 'EUC-JP', f.document.encoding
+        assert_equal "こんにちは！", f.content
       end
       def test_colons_are_not_removed
@@ -270,7 +268,7 @@ module Nokogiri
       def test_capturing_nonparse_errors_during_fragment_clone
         # see https://github.com/sparklemotion/nokogiri/issues/1196 for background
-        original = Nokogiri::HTML.fragment("<div id='unique'></div>")
+        original = Nokogiri::HTML.fragment("<div id='unique'></div><div id='unique'></div>")
         original_errors = original.errors.dup
         copy = original.dup
@@ -278,19 +276,20 @@ module Nokogiri
       end
       def test_capturing_nonparse_errors_during_node_copy_between_fragments
-        skip("JRuby HTML parse errors are different than libxml2's") if Nokogiri.jruby?
-        frag1 = Nokogiri::HTML.fragment("<div id='unique'>one</div>")
-        frag2 = Nokogiri::HTML.fragment("<div id='unique'>two</div>")
+        # Errors should be emitted while parsing only, and should not change when moving nodes.
+        frag1 = Nokogiri::HTML.fragment("<diva id='unique'>one</diva>")
+        frag2 = Nokogiri::HTML.fragment("<dive id='unique'>two</dive>")
         node1 = frag1.at_css("#unique")
         node2 = frag2.at_css("#unique")
+        original_errors1 = frag1.errors.dup
+        original_errors2 = frag2.errors.dup
+        assert original_errors1.any?{|e| e.to_s =~ /Tag diva invalid/ }, "it should complain about the tag name"
+        assert original_errors2.any?{|e| e.to_s =~ /Tag dive invalid/ }, "it should complain about the tag name"
-        original_errors = frag1.errors.dup
-        node1.add_child node2 # we should also not see an error on stderr
+        node1.add_child node2
-        assert_equal original_errors.length+1, frag1.errors.length
-        assert_match(/ID unique already defined/, frag1.errors.last.to_s)
+        assert_equal original_errors1, frag1.errors
+        assert_equal original_errors2, frag2.errors
       end
     end
   end

data/test/html/test_node.rb CHANGED

@@ -192,5 +192,21 @@ module Nokogiri
         end
       end
     end
+    def test_GH_1042
+      file = File.join(ASSETS_DIR, 'GH_1042.html');
+      html = Nokogiri::HTML(File.read(file))
+      table = html.xpath("//table")[1]
+      trs = table.xpath("tr").drop(1)
+      # the jruby inplementation of drop uses dup() on the IRubyObject (which
+      # is NOT the same dup() method on the ruby Object) which produces a
+      # shallow clone. a shallow of valid XMLNode triggers several
+      # NullPointerException on inspect() since loads of invariants
+      # are not set. the fix for GH1042 ensures a proper working clone.
+      assert_nothing_raised do
+        trs.inspect
+      end
+    end
   end
 end

data/test/html/test_node_encoding.rb CHANGED

@@ -3,24 +3,22 @@ require "helper"
 module Nokogiri
   module HTML
-    if RUBY_VERSION =~ /^1\.9/
-      class TestNodeEncoding < Nokogiri::TestCase
-        def test_inner_html
-          doc = Nokogiri::HTML File.open(SHIFT_JIS_HTML, 'rb')
+    class TestNodeEncoding < Nokogiri::TestCase
+      def test_inner_html
+        doc = Nokogiri::HTML File.open(SHIFT_JIS_HTML, 'rb')
-          hello = "こんにちは"
+        hello = "こんにちは"
-          contents = doc.at('h2').inner_html
-          assert_equal doc.encoding, contents.encoding.name
-          assert_match hello.encode('Shift_JIS'), contents
+        contents = doc.at('h2').inner_html
+        assert_equal doc.encoding, contents.encoding.name
+        assert_match hello.encode('Shift_JIS'), contents
-          contents = doc.at('h2').inner_html(:encoding => 'UTF-8')
-          assert_match hello, contents
+        contents = doc.at('h2').inner_html(:encoding => 'UTF-8')
+        assert_match hello, contents
-          doc.encoding = 'UTF-8'
-          contents = doc.at('h2').inner_html
-          assert_match hello, contents
-        end
+        doc.encoding = 'UTF-8'
+        contents = doc.at('h2').inner_html
+        assert_match hello, contents
       end
     end
   end

data/test/namespaces/test_namespaces_in_parsed_doc.rb CHANGED

@@ -61,6 +61,20 @@ module Nokogiri
         ns_attrs = n.to_xml.scan(/\bxmlns(?::.+?)?=/)
         assert_equal 3, ns_attrs.length
       end
+      def test_namespaces_under_memory_pressure_issue1155
+        skip("JRuby doesn't do GC.") if Nokogiri.jruby?
+        # this test is here to emit warnings when run under valgrind
+        # see https://github.com/sparklemotion/nokogiri/issues/1155 for background
+        filename = File.join ASSETS_DIR, 'namespace_pressure_test.xml'
+        doc = Nokogiri::XML File.open(filename)
+        # bizarrely, can't repro without the call to #to_a
+        doc.xpath('//namespace::*').to_a.each do |ns|
+          ns.inspect
+        end
+      end
     end
   end
 end

data/test/test_reader.rb CHANGED

@@ -386,6 +386,25 @@ class TestReader < Nokogiri::TestCase
                   reader.map { |n| n.namespace_uri })
   end
+  def test_namespaced_attributes
+    reader = Nokogiri::XML::Reader.from_memory(<<-eoxml)
+    <x xmlns:edi='http://ecommerce.example.org/schema' xmlns:commons="http://rets.org/xsd/RETSCommons">
+      <edi:foo commons:street-number="43">hello</edi:foo>
+      <y edi:name="francis" bacon="87"/>
+    </x>
+    eoxml
+    attr_ns = []
+    while reader.read
+      if reader.node_type == Nokogiri::XML::Node::ELEMENT_NODE
+        reader.attribute_nodes.each {|attr| attr_ns << (attr.namespace.nil? ? nil : attr.namespace.prefix) }
+      end
+    end
+    assert_equal(['commons',
+                  'edi',
+                  nil],
+                 attr_ns)
+  end
   def test_local_name
     reader = Nokogiri::XML::Reader.from_memory(<<-eoxml)
     <x xmlns:edi='http://ecommerce.example.org/schema'>

data/test/test_xslt_transforms.rb CHANGED

@@ -32,7 +32,7 @@ class TestXsltTransforms < Nokogiri::TestCase
     assert_match %r{<h1>Grandma</h1>}, result
     assert result = style.apply_to(@doc)
-    assert_match %r{<h1></h1>}, result
+    assert_match %r{<h1></h1>|<h1/>}, result
   end
   def test_transform_with_output_style
@@ -95,7 +95,9 @@ encoding="iso-8859-1" indent="yes"/>
 </xsl:stylesheet>
       eoxslt
     end
-    assert_no_match(/<td>/, xslt.apply_to(@doc, ['title', 'foo']))
+    result = xslt.apply_to(@doc, ['title', 'foo'])
+    assert_no_match(/<td>/, result)
+    assert_match(/This is an adjacent/, result)
   end
   def test_transform_arg_error
@@ -268,7 +270,7 @@ encoding="iso-8859-1" indent="yes"/>
         <xsl:output encoding="UTF-8" indent="yes" method="xml" />
         <xsl:template match="/">
-          <xsl:value-of select="/a" />
+          <a><xsl:value-of select="/a" /></a>
         </xsl:template>
       </xsl:stylesheet>
     EOXSL

data/test/xml/sax/test_parser.rb CHANGED

@@ -167,10 +167,8 @@ module Nokogiri
           assert @parser.document.errors
           assert @parser.document.errors.length > 0
-          if RUBY_VERSION =~ /^1\.9/
-            doc.errors.each do |error|
-              assert_equal 'UTF-8', error.message.encoding.name
-            end
+          doc.errors.each do |error|
+            assert_equal 'UTF-8', error.message.encoding.name
           end
           # when using JRuby Nokogiri, more errors will be generated as the DOM
@@ -207,42 +205,41 @@ module Nokogiri
             @parser.parse_io(f, encoding)
           }
           assert(@parser.document.cdata_blocks.length > 0)
-          if RUBY_VERSION =~ /^1\.9/
-            called = false
-            @parser.document.start_elements.flatten.each do |thing|
-              assert_equal 'UTF-8', thing.encoding.name
-              called = true
-            end
-            assert called
-            called = false
-            @parser.document.end_elements.flatten.each do |thing|
-              assert_equal 'UTF-8', thing.encoding.name
-              called = true
-            end
-            assert called
-            called = false
-            @parser.document.data.each do |thing|
-              assert_equal 'UTF-8', thing.encoding.name
-              called = true
-            end
-            assert called
-            called = false
-            @parser.document.comments.flatten.each do |thing|
-              assert_equal 'UTF-8', thing.encoding.name
-              called = true
-            end
-            assert called
-            called = false
-            @parser.document.cdata_blocks.flatten.each do |thing|
-              assert_equal 'UTF-8', thing.encoding.name
-              called = true
-            end
-            assert called
+          called = false
+          @parser.document.start_elements.flatten.each do |thing|
+            assert_equal 'UTF-8', thing.encoding.name
+            called = true
+          end
+          assert called
+          called = false
+          @parser.document.end_elements.flatten.each do |thing|
+            assert_equal 'UTF-8', thing.encoding.name
+            called = true
+          end
+          assert called
+          called = false
+          @parser.document.data.each do |thing|
+            assert_equal 'UTF-8', thing.encoding.name
+            called = true
+          end
+          assert called
+          called = false
+          @parser.document.comments.flatten.each do |thing|
+            assert_equal 'UTF-8', thing.encoding.name
+            called = true
+          end
+          assert called
+          called = false
+          @parser.document.cdata_blocks.flatten.each do |thing|
+            assert_equal 'UTF-8', thing.encoding.name
+            called = true
           end
+          assert called
         end
         def test_parse_file