adding more nokogiri tests and making the main rails tests pass

[#2190 state:resolved]
Signed-off-by: NJeremy Kemper <jeremy@bitsweat.net>

activesupport/lib/active_support/xml_mini/nokogiri.rb

+require 'nokogiri'
+
 # = XmlMini Nokogiri implementation
 module ActiveSupport
   module XmlMini_Nokogiri #:nodoc:
@@ -10,7 +12,9 @@ def parse(string)
       if string.blank?
         {}
       else
-        Nokogiri::XML(string).to_hash
+        doc = Nokogiri::XML(string)
+        raise doc.errors.first if doc.errors.length > 0
+        doc.to_hash
       end
     end
 
@@ -31,8 +35,8 @@ module Node
         def to_hash(hash = {})
           hash[name] ||= attributes_as_hash
 
-          walker = lambda { |child, memo, callback|
-            next if child.blank?
+          walker = lambda { |memo, parent, child, callback|
+            next if child.blank? && 'file' != parent['type']
 
             if child.text?
               (memo[CONTENT_ROOT] ||= '') << child.content
@@ -41,18 +45,21 @@ def to_hash(hash = {})
 
             name = child.name
 
+            child_hash = child.attributes_as_hash
             if memo[name]
               memo[name] = [memo[name]].flatten
-              memo[name] << child.attributes_as_hash
+              memo[name] << child_hash
             else
-              memo[name] = child.attributes_as_hash
+              memo[name] = child_hash
             end
 
             # Recusively walk children
-            child.children.each { |c| callback.call(c, memo[name], callback) }
+            child.children.each { |c|
+              callback.call(child_hash, child, c, callback)
+            }
           }
 
-          children.each { |c| walker.call(c, hash[name], walker) }
+          children.each { |c| walker.call(hash[name], self, c, walker) }
           hash
         end
 

activesupport/test/core_ext/hash_ext_test.rb

@@ -884,7 +884,12 @@ def test_array_values_are_not_sorted
   end
 
   def test_expansion_count_is_limited
-    expected = defined?(LibXML) ? LibXML::XML::Error : RuntimeError
+    expected = {
+      'ActiveSupport::XmlMini_REXML'    => 'RuntimeError',
+      'ActiveSupport::XmlMini_Nokogiri' => 'Nokogiri::XML::SyntaxError',
+      'ActiveSupport::XmlMini_LibXML'   => 'LibXML::XML::Error',
+    }[ActiveSupport::XmlMini.backend.name].constantize
+
     assert_raise expected do
       attack_xml = <<-EOT
       <?xml version="1.0" encoding="UTF-8"?>

activesupport/test/xml_mini/nokogiri_engine_test.rb

@@ -21,6 +21,42 @@ def teardown
     XmlMini.backend = @default_backend
   end
 
+  def test_file_from_xml
+    hash = Hash.from_xml(<<-eoxml)
+      <blog>
+        <logo type="file" name="logo.png" content_type="image/png">
+        </logo>
+      </blog>
+    eoxml
+    assert hash.has_key?('blog')
+    assert hash['blog'].has_key?('logo')
+
+    file = hash['blog']['logo']
+    assert_equal 'logo.png', file.original_filename
+    assert_equal 'image/png', file.content_type
+  end
+
+  def test_exception_thrown_on_expansion_attack
+    assert_raise Nokogiri::XML::SyntaxError do
+      attack_xml = <<-EOT
+      <?xml version="1.0" encoding="UTF-8"?>
+      <!DOCTYPE member [
+        <!ENTITY a "&b;&b;&b;&b;&b;&b;&b;&b;&b;&b;">
+        <!ENTITY b "&c;&c;&c;&c;&c;&c;&c;&c;&c;&c;">
+        <!ENTITY c "&d;&d;&d;&d;&d;&d;&d;&d;&d;&d;">
+        <!ENTITY d "&e;&e;&e;&e;&e;&e;&e;&e;&e;&e;">
+        <!ENTITY e "&f;&f;&f;&f;&f;&f;&f;&f;&f;&f;">
+        <!ENTITY f "&g;&g;&g;&g;&g;&g;&g;&g;&g;&g;">
+        <!ENTITY g "xxxxxxxxxxxxxxxxxxxxxxxxxxxxxx">
+      ]>
+      <member>
+      &a;
+      </member>
+      EOT
+      Hash.from_xml(attack_xml)
+    end
+  end
+
   def test_setting_nokogiri_as_backend
     XmlMini.backend = 'Nokogiri'
     assert_equal XmlMini_Nokogiri, XmlMini.backend
@@ -31,6 +67,17 @@ def test_blank_returns_empty_hash
     assert_equal({}, XmlMini.parse(''))
   end
 
+  def test_array_type_makes_an_array
+    assert_equal_rexml(<<-eoxml)
+      <blog>
+        <posts type="array">
+          <post>a post</post>
+          <post>another post</post>
+        </posts>
+      </blog>
+    eoxml
+  end
+
   def test_one_node_document_as_hash
     assert_equal_rexml(<<-eoxml)
     <products/>