nokogiri 1.6.2.1-java → 1.6.3-java

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 86e7f8df4f2ded92061f78e5de04e32b30968a4d
-  data.tar.gz: 53bd91a477753dfd493b81fad44ae7ca235d19c0
+  metadata.gz: 56b670c9b1574c610cb83034fee9f044ab7f2b29
+  data.tar.gz: b6d4586a68f531110b29f96f75317b464e53e033
 SHA512:
-  metadata.gz: 4faeceb14d3a2ab3feb8f6092281bc0c254b6c2908f9ff8d3d30c039c0643f0ac94d0fd9ea35e303decad55596d37030d3c6e4cb830b96d4c4f1f3764941f1d8
-  data.tar.gz: a46e13fb03da556e98a9d7822df48315993799c4f446d57f7dd51eae65cbba5e7aa22962e2a71d675b1a67fba4fa7b75215d602481a9c25223d3eee1f39062d4
+  metadata.gz: 7c642cad9650139cbea702fc74635d21d9c634216c7b0e09eff22fda426d5a5dd685256c5fd187fcd028253cd9a633d46d3883730c723b5fb379473b98eda390
+  data.tar.gz: abdf35dcd629550cf9930520b29cbb965bad38ad8abd715482f47127610fee4c8b807b70f8478da979ac65f77b95beb3cfbc4e78c551fd19d648e624ad9ec443

data/.travis.yml

@@ -2,6 +2,7 @@ language: ruby
 rvm:
   - 1.9.3
   - 2.0.0
+  - 2.1.2
   - ruby-head
   - jruby-19mode
   - jruby-20mode
@@ -21,6 +22,8 @@ matrix:
       jdk: openjdk7
     - rvm: 2.0.0
       jdk: openjdk7
+    - rvm: 2.1.2
+      jdk: openjdk7
     - rvm: ruby-head
       jdk: openjdk7
     - rvm: rbx-19mode

data/CHANGELOG.ja.rdoc

@@ -1,3 +1,19 @@
+=== 1.6.3 / 2014年7月20日
+
+==== 機能
+
+* Added Node#document? and Node#processing_instruction?
+
+
+==== バグ修正
+
+* [JRuby] Fix Ruby memory exhaustion vulnerability. #1087 (Thanks, @ocher)
+* [MRI] Fix segfault during GC when using `libxml-ruby` and `nokogiri` together in multi-threaded environment. #895 (Thanks, @ender672!)
+* Building on OSX 10.9 stock ruby 2.0.0 now works. #1101 (Thanks, @zenspider!)
+* Node#parse now works again for HTML document nodes (broken in 1.6.2+).
+* Processing instructions can now be added via Node#add_next_sibling.
+
+
 === 1.6.2.1 / 2014年5月13日
 
 ==== バグ修正

data/CHANGELOG.rdoc

@@ -1,3 +1,19 @@
+=== 1.6.3 / 2014-07-20
+
+==== Features
+
+* Added Node#document? and Node#processing_instruction?
+
+
+==== Bug fixes
+
+* [JRuby] Fix Ruby memory exhaustion vulnerability. #1087 (Thanks, @ocher)
+* [MRI] Fix segfault during GC when using `libxml-ruby` and `nokogiri` together in multi-threaded environment. #895 (Thanks, @ender672!)
+* Building on OSX 10.9 stock ruby 2.0.0 now works. #1101 (Thanks, @zenspider!)
+* Node#parse now works again for HTML document nodes (broken in 1.6.2+).
+* Processing instructions can now be added via Node#add_next_sibling.
+
+
 === 1.6.2.1 / 2014-05-13
 
 ==== Bug fixes

data/ext/java/nokogiri/internals/NokogiriNonStrictErrorHandler.java

@@ -90,9 +90,11 @@ public class NokogiriNonStrictErrorHandler extends NokogiriErrorHandler{
      * the parsing to stop, or an error that can be ignored.
      */
     private static boolean isFatal(String msg) {
+        String msgLowerCase = msg.toLowerCase();
         return
-          msg.toLowerCase().contains("in prolog") ||
-          msg.toLowerCase().contains("limit") ||
-          msg.toLowerCase().contains("preceding the root element must be well-formed");
+          msgLowerCase.contains("in prolog") ||
+          msgLowerCase.contains("limit") ||
+          msgLowerCase.contains("preceding the root element must be well-formed") ||
+          msgLowerCase.contains("following the root element must be well-formed");
     }
 }

data/ext/nokogiri/extconf.rb

@@ -311,6 +311,9 @@ when 'mingw32', /mswin/
   $CFLAGS << " -DXP_WIN -DXP_WIN32 -DUSE_INCLUDED_VASPRINTF"
 when /solaris/
   $CFLAGS << " -DUSE_INCLUDED_VASPRINTF"
+when /darwin/
+  # TODO: maybe make it stricter to only work on stock ruby? How to tell?
+  $CFLAGS << " -Wno-error=unused-command-line-argument-hard-error-in-future"
 else
   $CFLAGS << " -g -DXP_UNIX"
 end
@@ -469,6 +472,7 @@ else
         }
       }
 
+      # Defining a macro that expands to a C string; double quotes are significant.
       $CPPFLAGS << ' ' << "-DNOKOGIRI_#{recipe.name.upcase}_PATH=\"#{recipe.path}\"".shellescape
 
       case libname

data/ext/nokogiri/xml_document.c

@@ -17,13 +17,29 @@ static int dealloc_node_i(xmlNodePtr key, xmlNodePtr node, xmlDocPtr doc)
   return ST_CONTINUE;
 }
 
+static void remove_private(xmlNodePtr node)
+{
+  xmlNodePtr child;
+
+  for (child = node->children; child; child = child->next)
+    remove_private(child);
+
+  if ((node->type == XML_ELEMENT_NODE ||
+       node->type == XML_XINCLUDE_START ||
+       node->type == XML_XINCLUDE_END) &&
+      node->properties) {
+    for (child = (xmlNodePtr)node->properties; child; child = child->next)
+      remove_private(child);
+  }
+
+  node->_private = NULL;
+}
+
 static void dealloc(xmlDocPtr doc)
 {
-  xmlDeregisterNodeFunc func;
   st_table *node_hash;
 
   NOKOGIRI_DEBUG_START(doc);
-  func = xmlDeregisterNodeDefault(NULL);
 
   node_hash  = DOC_UNLINKED_NODE_HASH(doc);
 
@@ -31,10 +47,17 @@ static void dealloc(xmlDocPtr doc)
   st_free_table(node_hash);
 
   free(doc->_private);
-  doc->_private = NULL;
+
+  /* When both Nokogiri and libxml-ruby are loaded, make sure that all nodes
+   * have their _private pointers cleared. This is to avoid libxml-ruby's
+   * xmlDeregisterNode callback from accessing VALUE pointers from ruby's GC
+   * free context, which can result in segfaults.
+   */
+  if (xmlDeregisterNodeDefaultValue)
+    remove_private((xmlNodePtr)doc);
+
   xmlFreeDoc(doc);
 
-  xmlDeregisterNodeDefault(func);
   NOKOGIRI_DEBUG_END(doc);
 }
 

data/lib/nokogiri/version.rb

@@ -1,6 +1,6 @@
 module Nokogiri
   # The version of Nokogiri you are using
-  VERSION = '1.6.2.1'
+  VERSION = '1.6.3'
 
   class VersionInfo # :nodoc:
     def jruby?

data/lib/nokogiri/xml/node.rb

@@ -287,7 +287,7 @@ module Nokogiri
       def prepend_child node_or_tags
         if first = children.first
           # Mimic the error add_child would raise.
-          raise RuntimeError, "Document already has a root node" if is_a?(XML::Document) && !node_or_tags.is_a?(XML::ProcessingInstruction)
+          raise RuntimeError, "Document already has a root node" if document? && !node_or_tags.processing_instruction?
           first.__send__(:add_sibling, :previous, node_or_tags)
         else
           add_child(node_or_tags)
@@ -313,7 +313,7 @@ module Nokogiri
       #
       # Also see related method +before+.
       def add_previous_sibling node_or_tags
-        raise ArgumentError.new("A document may not have multiple root nodes.") if parent.is_a?(XML::Document) && !node_or_tags.is_a?(XML::ProcessingInstruction)
+        raise ArgumentError.new("A document may not have multiple root nodes.") if (parent && parent.document?) && !node_or_tags.processing_instruction?
 
         add_sibling :previous, node_or_tags
       end
@@ -326,7 +326,7 @@ module Nokogiri
       #
       # Also see related method +after+.
       def add_next_sibling node_or_tags
-        raise ArgumentError.new("A document may not have multiple root nodes.") if parent.is_a?(XML::Document)
+        raise ArgumentError.new("A document may not have multiple root nodes.") if (parent && parent.document?) && !node_or_tags.processing_instruction?
         
         add_sibling :next, node_or_tags
       end
@@ -513,7 +513,7 @@ module Nokogiri
         # document as the parsing context instead. Otherwise, the in-context
         # parser cannot find an element or a document node.
         # Document Fragments are also not usable by the in-context parser.
-        if !element? && !xml? && (!parent || parent.fragment?)
+        if !element? && !document? && (!parent || parent.fragment?)
           return document.parse(string_or_io, options)
         end
 
@@ -601,6 +601,16 @@ module Nokogiri
         type == HTML_DOCUMENT_NODE
       end
 
+      # Returns true if this is a Document
+      def document?
+        is_a? XML::Document
+      end
+
+      # Returns true if this is a ProcessingInstruction node
+      def processing_instruction?
+        type == PI_NODE
+      end
+
       # Returns true if this is a Text node
       def text?
         type == TEXT_NODE

data/lib/nokogiri/xml/sax/document.rb

@@ -8,7 +8,7 @@ module Nokogiri
     # The basic way a SAX style parser works is by creating a parser,
     # telling the parser about the events we're interested in, then giving
     # the parser some XML to process.  The parser will notify you when
-    # it encounters events your said you would like to know about.
+    # it encounters events you said you would like to know about.
     #
     # To register for events, you simply subclass Nokogiri::XML::SAX::Document,
     # and implement the methods for which you would like notification.

data/test/xml/test_document.rb

@@ -625,6 +625,12 @@ module Nokogiri
         refute_empty doc.errors
       end
 
+      def test_memory_explosion_on_wrong_formatted_element_following_the_root_element
+        doc = Nokogiri::XML("<a/><\n")
+        refute_nil doc
+        refute_empty doc.errors
+      end
+
       def test_document_has_errors
         doc = Nokogiri::XML(<<-eoxml)
           <foo><bar></foo>

data/test/xml/test_node.rb

@@ -136,7 +136,29 @@ module Nokogiri
       def test_parse_with_unparented_text_context_node
         doc = XML::Document.new
         elem = XML::Text.new("foo", doc)
-        elem.parse("<bar/>")
+        x = elem.parse("<bar/>") # should not raise an exception
+        assert_equal x.first.name, "bar"
+      end
+
+      def test_parse_with_unparented_html_text_context_node
+        doc = HTML::Document.new
+        elem = XML::Text.new("div", doc)
+        x = elem.parse("<div/>") # should not raise an exception
+        assert_equal x.first.name, "div"
+      end
+
+      def test_parse_with_unparented_fragment_text_context_node
+        doc = XML::DocumentFragment.parse "<div><span>foo</span></div>"
+        elem = doc.at_css "span"
+        x = elem.parse("<span/>") # should not raise an exception
+        assert_equal x.first.name, "span"
+      end
+
+      def test_parse_with_unparented_html_fragment_text_context_node
+        doc = HTML::DocumentFragment.parse "<div><span>foo</span></div>"
+        elem = doc.at_css "span"
+        x = elem.parse("<span/>") # should not raise an exception
+        assert_equal x.first.name, "span"
       end
 
       def test_subclass_dup
@@ -836,7 +858,7 @@ b"></div>
         ne = d1.root.xpath('//a').first.dup(1)
         ne.content += "& < & > \" &"
         d2.root << ne
-        assert_match /<a>&amp;&amp; &lt; &amp; &gt; " &amp;<\/a>/, d2.to_s
+        assert_match /<a>&amp;&amp; &lt; &amp; &gt; \" &amp;<\/a>/, d2.to_s
       end
 
       def test_content_after_appending_text
@@ -1150,6 +1172,26 @@ eoxml
         root << "<a>hello:with_colon</a>"
         assert_match(/hello:with_colon/, document.to_xml)
       end
+
+      def test_document_eh
+        html_doc = Nokogiri::HTML "<div>foo</div>"
+        xml_doc = Nokogiri::XML "<div>foo</div>"
+        html_node = html_doc.at_css "div"
+        xml_node = xml_doc.at_css "div"
+
+        assert html_doc.document?
+        assert xml_doc.document?
+        assert ! html_node.document?
+        assert ! xml_node.document?
+      end
+
+      def test_processing_instruction_eh
+        xml_doc = Nokogiri::XML %Q{<?xml version="1.0"?>\n<?xml-stylesheet type="text/xsl" href="foo.xsl"?>\n<?xml-stylesheet type="text/xsl" href="foo2.xsl"?>\n<root><div>foo</div></root>}
+        pi_node = xml_doc.children.first
+        div_node = xml_doc.at_css "div"
+        assert pi_node.processing_instruction?
+        assert ! div_node.processing_instruction?
+      end
     end
   end
 end

data/test/xml/test_unparented_node.rb

@@ -248,12 +248,17 @@ module Nokogiri
         end
       end
 
-      def test_add_pi_as_previous_sibling_to_root_is_ok
+      def test_add_pi_as_previous_and_next_sibling_to_root_is_ok
         doc = Nokogiri::XML "<root>foo</root>"
         pi = Nokogiri::XML::ProcessingInstruction.new(doc, "xml-stylesheet", %q{type="text/xsl" href="foo.xsl"})
         doc.root.add_previous_sibling pi
         expected_doc = %Q{<?xml version="1.0"?>\n<?xml-stylesheet type="text/xsl" href="foo.xsl"?>\n<root>foo</root>}
         assert_includes doc.to_xml, expected_doc
+
+        pi2 = Nokogiri::XML::ProcessingInstruction.new(doc, "xml-stylesheet", %q{type="text/xsl" href="foo2.xsl"})
+        pi.add_next_sibling pi2
+        expected_doc = %Q{<?xml version="1.0"?>\n<?xml-stylesheet type="text/xsl" href="foo.xsl"?>\n<?xml-stylesheet type="text/xsl" href="foo2.xsl"?>\n<root>foo</root>}
+        assert_includes doc.to_xml, expected_doc
       end
 
       def test_find_by_css_with_tilde_eql

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: nokogiri
 version: !ruby/object:Gem::Version
-  version: 1.6.2.1
+  version: 1.6.3
 platform: java
 authors:
 - Aaron Patterson
@@ -12,7 +12,7 @@ authors:
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2014-05-14 00:00:00.000000000 Z
+date: 2014-07-20 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rdoc