nokogiri 1.3.2-java → 1.3.3-java

data/test/hpricot/test_paths.rb

@@ -5,11 +5,11 @@ class TestParser < Nokogiri::TestCase
   include Nokogiri
 
   def test_roundtrip
-    @basic = Hpricot.parse(TestFiles::BASIC)
+    @basic = Nokogiri.parse(TestFiles::BASIC)
     %w[link link[2] body #link1 a p.ohmy].each do |css_sel|
       ele = @basic.at(css_sel)
       assert_equal ele, @basic.at(ele.css_path), ele.css_path
-      assert_equal ele, @basic.at(ele.xpath), ele.xpath
+      assert_equal ele, @basic.at(ele.path), ele.xpath
     end
   end
 end

data/test/hpricot/test_preserved.rb

@@ -64,10 +64,10 @@ class TestPreserved < Nokogiri::TestCase
   def test_escaping_of_attrs
     # ampersands in URLs
     str = %{<a href="http://google.com/search?q=nokogiri&amp;l=en">Google</a>}
-    link = (doc = Nokogiri.Hpricot(str)).at(:a)
+    link = (doc = Nokogiri(str)).at(:a)
     assert_equal "http://google.com/search?q=nokogiri&l=en", link['href']
     assert_equal "http://google.com/search?q=nokogiri&l=en", link.get_attribute('href')
-    assert_equal "http://google.com/search?q=nokogiri&l=en", link.raw_attributes['href']
+    assert_equal "http://google.com/search?q=nokogiri&l=en", link['href']
     assert_equal str, link.to_html
 
     # alter the url

data/test/hpricot/test_xml.rb

@@ -6,14 +6,14 @@ class TestParser < Nokogiri::TestCase
   # normally, the link tags are empty HTML tags.
   # contributed by laudney.
   def test_normally_empty
-    doc = Hpricot::XML("<rss><channel><title>this is title</title><link>http://fake.com</link></channel></rss>")
+    doc = Nokogiri::XML("<rss><channel><title>this is title</title><link>http://fake.com</link></channel></rss>")
     assert_equal "this is title", (doc/:rss/:channel/:title).text
     assert_equal "http://fake.com", (doc/:rss/:channel/:link).text
   end
 
   # make sure XML doesn't get downcased
   def test_casing
-    doc = Hpricot::XML(TestFiles::WHY)
+    doc = Nokogiri::XML(TestFiles::WHY)
 
     ### Modified.
     # I don't want to differentiate pseudo classes from namespaces.  If
@@ -24,7 +24,7 @@ class TestParser < Nokogiri::TestCase
 
   # be sure tags named "text" are ok
   def test_text_tags
-    doc = Hpricot::XML("<feed><title>City Poisoned</title><text>Rita Lee has poisoned Brazil.</text></feed>")
+    doc = Nokogiri::XML("<feed><title>City Poisoned</title><text>Rita Lee has poisoned Brazil.</text></feed>")
     assert_equal "City Poisoned", (doc/"title").text
   end
 end

data/test/html/sax/test_parser.rb

@@ -9,6 +9,18 @@ module Nokogiri
           @parser = HTML::SAX::Parser.new(Doc.new)
         end
 
+        def test_parse_empty_document
+          # This caused a segfault in libxml 2.6.x
+          assert_nothing_raised { @parser.parse '' }
+        end
+
+        def test_parse_empty_file
+          # Make sure empty files don't break stuff
+          empty_file_name =  File.join(Dir.tmpdir, 'bogus.xml')
+          FileUtils.touch empty_file_name
+          assert_nothing_raised { @parser.parse_file empty_file_name }
+        end
+
         def test_parse_file
           @parser.parse_file(HTML_FILE)
           assert_equal 1110, @parser.document.end_elements.length

data/test/html/test_builder.rb

@@ -47,11 +47,13 @@ module Nokogiri
 
       def test_tag_nesting
         builder = Nokogiri::HTML::Builder.new do
-          span.left ''
-          span.middle {
-            div.icon ''
+          body {
+            span.left ''
+            span.middle {
+              div.icon ''
+            }
+            span.right ''
           }
-          span.right ''
         end
         assert node = builder.doc.css('span.right').first
         assert_equal 'middle', node.previous_sibling['class']

data/test/html/test_document.rb

@@ -115,6 +115,13 @@ module Nokogiri
         assert_instance_of Nokogiri::HTML::Document, parent
       end
 
+      def test_parse_handles_nil_gracefully
+        assert_nothing_raised do
+          @doc = Nokogiri::HTML::Document.parse(nil)
+        end
+        assert_instance_of Nokogiri::HTML::Document, @doc
+      end
+
       def test_parse_empty_document
         doc = Nokogiri::HTML("\n")
         assert_equal 0, doc.css('a').length

data/test/html/test_document_encoding.rb

@@ -24,6 +24,23 @@ module Nokogiri
           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)
+<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
+        end
+
         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">

data/test/html/test_document_fragment.rb

@@ -37,6 +37,12 @@ module Nokogiri
         assert_instance_of klass, fragment
       end
 
+      def test_subclass_parse
+        klass = Class.new(Nokogiri::HTML::DocumentFragment)
+        doc = klass.parse("<div>a</div>")
+        assert_instance_of klass, doc
+      end
+
       def test_html_fragment
         fragment = Nokogiri::HTML.fragment("<div>a</div>")
         assert_equal "<div>a</div>", fragment.to_s
@@ -64,6 +70,12 @@ module Nokogiri
         assert_equal "<div>b</div>", fragment.to_s
       end
 
+      def test_html_fragment_with_leading_whitespace_and_newline
+        doc = "     \n<div>b</div>  "
+        fragment = Nokogiri::HTML::Document.new.fragment(doc)
+        assert_equal "<div>b</div>", fragment.to_s
+      end
+
       def test_to_s
         doc = "<span>foo<br></span><span>bar</span>"
         fragment = Nokogiri::HTML::Document.new.fragment(doc)

data/test/html/test_node.rb

@@ -88,8 +88,11 @@ module Nokogiri
 
       def test_inner_html=
         assert div = @html.at('//div')
-        div.inner_html = '<span>testing</span>'
-        assert_equal 'span', div.children.first.name
+        div.inner_html = '1<span>2</span>3'
+        assert_equal '1',    div.children[0].to_s
+        assert_equal 'span', div.children[1].name
+        assert_equal '2',    div.children[1].inner_text
+        assert_equal '3',    div.children[2].to_s
 
         div.inner_html = 'testing'
         assert_equal 'testing', div.content

data/test/test_convert_xpath.rb

@@ -1,36 +1,17 @@
 require File.expand_path(File.join(File.dirname(__FILE__), "helper"))
 
-begin
-  require 'rubygems'
-  require 'hpricot'
-  HAS_HPRICOT = true
-rescue LoadError
-  HAS_HPRICOT = false
-end
+require 'rubygems'
 
 class TestConvertXPath < Nokogiri::TestCase
 
   def setup
     super
     @N = Nokogiri(File.read(HTML_FILE))
-    @NH = Nokogiri.Hpricot(File.read(HTML_FILE)) # decorated document
-    @H = Hpricot(File.read(HTML_FILE)) if HAS_HPRICOT
   end
 
   def assert_syntactical_equivalence(hpath, xpath, match, &blk)
     blk ||= lambda {|j| j.first}
     assert_equal match, blk.call(@N.search(xpath)), "xpath result did not match"
-    if HAS_HPRICOT
-      assert_equal match, blk.call(@H.search(hpath)).chomp, "hpath result did not match"
-    end
-    assert_equal [xpath], @NH.convert_to_xpath(hpath), "converted hpath did not match xpath"
-  end
-
-  def test_ordinary_xpath_conversions
-    assert_equal(".//p", @NH.convert_to_xpath("p").first)
-    assert_equal(".//p", @NH.convert_to_xpath(:p).first)
-    assert_equal(".//p", @NH.convert_to_xpath("//p").first)
-    assert_equal(".//p", @NH.convert_to_xpath(".//p").first)
   end
 
   def test_child_tag
@@ -117,21 +98,6 @@ class TestConvertXPath < Nokogiri::TestCase
   end
 
   def test_positional
-##
-#  we are intentionally NOT staying compatible with nth-and-friends, as Hpricot has an OB1 bug.
-#
-#     assert_syntactical_equivalence("div > div:eq(0)", ".//div/div[position() = 1]", "\r\nTender Lovemaking\r\nThe act of making love, tenderly.\r\n") do |j|
-#       j.first.inner_text
-#     end
-#     assert_syntactical_equivalence("div/div:eq(0)", ".//div/div[position() = 1]", "\r\nTender Lovemaking\r\nThe act of making love, tenderly.\r\n") do |j|
-#       j.first.inner_text
-#     end
-#     assert_syntactical_equivalence("div/div:nth(0)", ".//div/div[position() = 1]", "\r\nTender Lovemaking\r\nThe act of making love, tenderly.\r\n") do |j|
-#       j.first.inner_text
-#     end
-#     assert_syntactical_equivalence("div/div:nth-of-type(0)", ".//div/div[position() = 1]", "\r\nTender Lovemaking\r\nThe act of making love, tenderly.\r\n") do |j|
-#       j.first.inner_text
-#     end
     assert_syntactical_equivalence("div/div:first()", ".//div/div[position() = 1]", "\r\nTender Lovemaking\r\nThe act of making love, tenderly.\r\n".gsub(/[\r\n]/, '')) do |j|
       j.first.inner_text.gsub(/[\r\n]/, '')
     end
@@ -152,21 +118,6 @@ class TestConvertXPath < Nokogiri::TestCase
     end
   end
 
-  def test_compat_mode_namespaces
-    assert_equal(".//*[name()='t:sam']", @NH.convert_to_xpath("//t:sam").first)
-    assert_equal(".//*[name()='t:sam'][@rel='bookmark'][1]", @NH.convert_to_xpath("//t:sam[@rel='bookmark'][1]").first)
-  end
-
-  ##
-  #  'and' is not supported by hpricot
-#   def test_and
-#     assert_syntactical_equivalence("div[h1 and small]", ".//div[h1 and small]", "\r\nTender Lovemaking\r\nThe act of making love, tenderly.\r\n") do |j|
-#       j.inner_text
-#     end
-#   end
-
-
-
 # TODO:
 #       doc/'title ~ link' -> links that are siblings of title
 #       doc/'p[@class~="final"]' -> class includes string (whitespacy)

data/test/test_css_cache.rb

@@ -31,7 +31,7 @@ class TestCssCache < Nokogiri::TestCase
       times = cache_setting ? 4 : nil
 
       Nokogiri::CSS::Parser.set_cache cache_setting
-      
+
       Nokogiri::CSS.xpath_for(@css)
       Nokogiri::CSS.xpath_for(@css)
       Nokogiri::CSS::Parser.new.xpath_for(@css)
@@ -39,17 +39,6 @@ class TestCssCache < Nokogiri::TestCase
 
       assert_equal(times, Nokogiri::CSS::Parser.class_eval { @cache.count })
     end
-
-    define_method "test_hpricot_cache_#{cache_setting ? "true" : "false"}" do
-      times = cache_setting ? 2 : nil
-      Nokogiri::CSS::Parser.set_cache cache_setting
-
-      nh = Nokogiri.Hpricot("<html></html>")
-
-      nh.convert_to_xpath(@css)
-      nh.convert_to_xpath(@css)
-      assert_equal(times, Nokogiri::CSS::Parser.class_eval { @cache.count })
-    end
   end
 
 

data/test/test_nokogiri.rb

@@ -30,6 +30,13 @@ class TestNokogiri < Nokogiri::TestCase
     assert_equal "#{major}.#{minor}.#{bug}", Nokogiri::VERSION_INFO['libxml']['loaded']
   end
 
+  def test_parse_with_io
+    doc = Nokogiri.parse(
+      StringIO.new("<html><head><title></title><body></body></html>")
+    )
+    assert_instance_of Nokogiri::HTML::Document, doc
+  end
+
   def test_xml?
     doc = Nokogiri.parse(File.read(XML_FILE))
     assert doc.xml?

data/test/test_reader.rb

@@ -313,4 +313,18 @@ class TestReader < Nokogiri::TestCase
     end
     assert called
   end
+
+  def test_large_document_smoke_test
+    #  simply run on a large document to verify that there no GC issues
+    xml = []
+    xml << "<elements>"
+    10000.times { |j| xml << "<element id=\"#{j}\"/>" }
+    xml << "</elements>"
+    xml = xml.join("\n")
+
+    Nokogiri::XML::Reader.from_memory(xml).each do |e|
+      e.attributes
+    end
+  end
+
 end

data/test/xml/test_document.rb

@@ -10,6 +10,37 @@ module Nokogiri
         @xml = Nokogiri::XML.parse(File.read(XML_FILE), XML_FILE)
       end
 
+      def test_empty_node_converted_to_html_is_not_self_closing
+        doc = Nokogiri::XML('<a></a>')
+        assert_equal "<a></a>", doc.inner_html
+      end
+
+      def test_add_child_with_fragment
+        doc = Nokogiri::XML::Document.new
+        fragment = doc.fragment('<hello />')
+        doc.add_child fragment
+        assert_equal '/hello', doc.at('//hello').path
+        assert_equal 'hello', doc.root.name
+      end
+
+      def test_add_child_with_multiple_roots
+        assert_raises(RuntimeError) do
+          @xml << Node.new('foo', @xml)
+        end
+      end
+
+      def test_move_root_to_document_with_no_root
+        sender = Nokogiri::XML('<root>foo</root>')
+        newdoc = Nokogiri::XML::Document.new
+        newdoc.root = sender.root
+      end
+
+      def test_move_root_with_existing_root_gets_gcd
+        doc = Nokogiri::XML('<root>test</root>')
+        doc2 = Nokogiri::XML("<root>#{'x' * 5000000}</root>")
+        doc2.root = doc.root
+      end
+
       def test_validate
         assert_equal 44, @xml.validate.length
       end
@@ -35,6 +66,13 @@ module Nokogiri
         end
       end
 
+      def test_parse_handles_nil_gracefully
+        assert_nothing_raised do
+          @doc = Nokogiri::XML::Document.parse(nil)
+        end
+        assert_instance_of Nokogiri::XML::Document, @doc
+      end
+
       def test_parse_takes_block
         options = nil
         Nokogiri::XML.parse(File.read(XML_FILE), XML_FILE) do |cfg|
@@ -215,6 +253,12 @@ module Nokogiri
         assert_raises(Nokogiri::XML::SyntaxError) {
           Nokogiri::XML('<foo><bar></foo>', nil, nil, 0)
         }
+
+        assert_raises(Nokogiri::XML::SyntaxError) {
+          Nokogiri::XML('<foo><bar></foo>') { |cfg|
+            cfg.strict
+          }
+        }
       end
 
       def test_XML_function

data/test/xml/test_document_fragment.rb

@@ -37,6 +37,12 @@ module Nokogiri
         assert_instance_of klass, fragment
       end
 
+      def test_subclass_parse
+        klass = Class.new(Nokogiri::XML::DocumentFragment)
+        doc = klass.parse("<div>a</div>")
+        assert_instance_of klass, doc
+      end
+
       def test_xml_fragment
         fragment = Nokogiri::XML.fragment("<div>a</div>")
         assert_equal "<div>a</div>", fragment.to_s
@@ -71,6 +77,12 @@ module Nokogiri
         fragment = Nokogiri::XML::Document.new.fragment(doc)
         assert_equal "<div>b</div>", fragment.to_s
       end
+
+      def test_xml_fragment_with_leading_whitespace_and_newline
+        doc = "     \n<div>b</div>  "
+        fragment = Nokogiri::XML::Document.new.fragment(doc)
+        assert_equal "<div>b</div>", fragment.to_s
+      end
     end
   end
 end

data/test/xml/test_node.rb

@@ -114,13 +114,21 @@ module Nokogiri
         end
       end
 
+      def test_fragment_creates_elements
+        apple = @xml.fragment('<Apple/>')
+        apple.children.each do |child|
+          assert_equal Nokogiri::XML::Node::ELEMENT_NODE, child.type
+          assert_instance_of Nokogiri::XML::Element, child
+        end
+      end
+
       def test_node_added_to_root_should_get_namespace
         fruits = Nokogiri::XML(<<-eoxml)
           <Fruit xmlns='http://www.fruits.org'>
           </Fruit>
         eoxml
         apple = fruits.fragment('<Apple/>')
-        fruits << apple
+        fruits.root << apple
         assert_equal 1, fruits.xpath('//xmlns:Apple').length
       end
 
@@ -364,7 +372,7 @@ module Nokogiri
       def test_new
         assert node = Nokogiri::XML::Node.new('input', @xml)
         assert_equal 1, node.node_type
-        assert_instance_of Nokogiri::XML::Node, node
+        assert_instance_of Nokogiri::XML::Element, node
       end
 
       def test_to_str

data/test/xml/test_node_encoding.rb

@@ -14,6 +14,29 @@ module Nokogiri
           assert_equal @html.encoding, node['href'].encoding.name
         end
 
+        def test_text_encoding_is_utf_8
+          @html = Nokogiri::HTML(File.open(NICH_FILE))
+          assert_equal 'UTF-8', @html.text.encoding.name
+        end
+
+        def test_serialize_encoding_html
+          @html = Nokogiri::HTML(File.open(NICH_FILE))
+          assert_equal @html.encoding.downcase,
+            @html.serialize.encoding.name.downcase
+
+          @doc = Nokogiri::HTML(@html.serialize)
+          assert_equal @html.serialize, @doc.serialize
+        end
+
+        def test_serialize_encoding_xml
+          @xml = Nokogiri::XML(File.open(SHIFT_JIS_XML))
+          assert_equal @xml.encoding.downcase,
+            @xml.serialize.encoding.name.downcase
+
+          @doc = Nokogiri::XML(@xml.serialize)
+          assert_equal @xml.serialize, @doc.serialize
+        end
+
         def test_encode_special_chars
           foo = @html.css('a').first.encode_special_chars('foo')
           assert_equal @html.encoding, foo.encoding.name