RubyGems - nokogiri - Versions diffs - 1.1.1-java - Mend

nokogiri 1.1.1-java

Potentially problematic release.

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

Files changed (142) hide show

data/History.ja.txt +99 -0
data/History.txt +99 -0
data/Manifest.txt +141 -0
data/README.ja.txt +100 -0
data/README.txt +109 -0
data/Rakefile +354 -0
data/ext/nokogiri/extconf.rb +93 -0
data/ext/nokogiri/html_document.c +86 -0
data/ext/nokogiri/html_document.h +10 -0
data/ext/nokogiri/html_sax_parser.c +36 -0
data/ext/nokogiri/html_sax_parser.h +11 -0
data/ext/nokogiri/native.c +41 -0
data/ext/nokogiri/native.h +50 -0
data/ext/nokogiri/xml_cdata.c +44 -0
data/ext/nokogiri/xml_cdata.h +9 -0
data/ext/nokogiri/xml_comment.c +42 -0
data/ext/nokogiri/xml_comment.h +9 -0
data/ext/nokogiri/xml_document.c +206 -0
data/ext/nokogiri/xml_document.h +10 -0
data/ext/nokogiri/xml_dtd.c +121 -0
data/ext/nokogiri/xml_dtd.h +8 -0
data/ext/nokogiri/xml_io.c +17 -0
data/ext/nokogiri/xml_io.h +9 -0
data/ext/nokogiri/xml_node.c +727 -0
data/ext/nokogiri/xml_node.h +13 -0
data/ext/nokogiri/xml_node_set.c +118 -0
data/ext/nokogiri/xml_node_set.h +9 -0
data/ext/nokogiri/xml_reader.c +465 -0
data/ext/nokogiri/xml_reader.h +10 -0
data/ext/nokogiri/xml_sax_parser.c +201 -0
data/ext/nokogiri/xml_sax_parser.h +10 -0
data/ext/nokogiri/xml_syntax_error.c +199 -0
data/ext/nokogiri/xml_syntax_error.h +11 -0
data/ext/nokogiri/xml_text.c +40 -0
data/ext/nokogiri/xml_text.h +9 -0
data/ext/nokogiri/xml_xpath.c +53 -0
data/ext/nokogiri/xml_xpath.h +11 -0
data/ext/nokogiri/xml_xpath_context.c +214 -0
data/ext/nokogiri/xml_xpath_context.h +9 -0
data/ext/nokogiri/xslt_stylesheet.c +123 -0
data/ext/nokogiri/xslt_stylesheet.h +9 -0
data/lib/action-nokogiri.rb +30 -0
data/lib/nokogiri.rb +72 -0
data/lib/nokogiri/css.rb +25 -0
data/lib/nokogiri/css/generated_parser.rb +721 -0
data/lib/nokogiri/css/generated_tokenizer.rb +159 -0
data/lib/nokogiri/css/node.rb +97 -0
data/lib/nokogiri/css/parser.rb +64 -0
data/lib/nokogiri/css/parser.y +216 -0
data/lib/nokogiri/css/syntax_error.rb +6 -0
data/lib/nokogiri/css/tokenizer.rb +9 -0
data/lib/nokogiri/css/tokenizer.rex +63 -0
data/lib/nokogiri/css/xpath_visitor.rb +168 -0
data/lib/nokogiri/decorators.rb +2 -0
data/lib/nokogiri/decorators/hpricot.rb +3 -0
data/lib/nokogiri/decorators/hpricot/node.rb +56 -0
data/lib/nokogiri/decorators/hpricot/node_set.rb +54 -0
data/lib/nokogiri/decorators/hpricot/xpath_visitor.rb +28 -0
data/lib/nokogiri/decorators/slop.rb +31 -0
data/lib/nokogiri/hpricot.rb +51 -0
data/lib/nokogiri/html.rb +105 -0
data/lib/nokogiri/html/builder.rb +9 -0
data/lib/nokogiri/html/document.rb +9 -0
data/lib/nokogiri/html/sax/parser.rb +21 -0
data/lib/nokogiri/version.rb +3 -0
data/lib/nokogiri/xml.rb +83 -0
data/lib/nokogiri/xml/after_handler.rb +18 -0
data/lib/nokogiri/xml/attr.rb +10 -0
data/lib/nokogiri/xml/before_handler.rb +33 -0
data/lib/nokogiri/xml/builder.rb +84 -0
data/lib/nokogiri/xml/cdata.rb +9 -0
data/lib/nokogiri/xml/comment.rb +6 -0
data/lib/nokogiri/xml/document.rb +55 -0
data/lib/nokogiri/xml/dtd.rb +6 -0
data/lib/nokogiri/xml/element.rb +6 -0
data/lib/nokogiri/xml/entity_declaration.rb +9 -0
data/lib/nokogiri/xml/node.rb +333 -0
data/lib/nokogiri/xml/node_set.rb +197 -0
data/lib/nokogiri/xml/notation.rb +6 -0
data/lib/nokogiri/xml/reader.rb +20 -0
data/lib/nokogiri/xml/sax.rb +9 -0
data/lib/nokogiri/xml/sax/document.rb +59 -0
data/lib/nokogiri/xml/sax/parser.rb +37 -0
data/lib/nokogiri/xml/syntax_error.rb +21 -0
data/lib/nokogiri/xml/text.rb +6 -0
data/lib/nokogiri/xml/xpath.rb +10 -0
data/lib/nokogiri/xml/xpath/syntax_error.rb +8 -0
data/lib/nokogiri/xml/xpath_context.rb +14 -0
data/lib/nokogiri/xslt.rb +28 -0
data/lib/nokogiri/xslt/stylesheet.rb +6 -0
data/test/css/test_nthiness.rb +159 -0
data/test/css/test_parser.rb +237 -0
data/test/css/test_tokenizer.rb +162 -0
data/test/css/test_xpath_visitor.rb +64 -0
data/test/files/dont_hurt_em_why.xml +422 -0
data/test/files/exslt.xml +8 -0
data/test/files/exslt.xslt +35 -0
data/test/files/staff.xml +59 -0
data/test/files/staff.xslt +32 -0
data/test/files/tlm.html +850 -0
data/test/helper.rb +78 -0
data/test/hpricot/files/basic.xhtml +17 -0
data/test/hpricot/files/boingboing.html +2266 -0
data/test/hpricot/files/cy0.html +3653 -0
data/test/hpricot/files/immob.html +400 -0
data/test/hpricot/files/pace_application.html +1320 -0
data/test/hpricot/files/tenderlove.html +16 -0
data/test/hpricot/files/uswebgen.html +220 -0
data/test/hpricot/files/utf8.html +1054 -0
data/test/hpricot/files/week9.html +1723 -0
data/test/hpricot/files/why.xml +19 -0
data/test/hpricot/load_files.rb +11 -0
data/test/hpricot/test_alter.rb +67 -0
data/test/hpricot/test_builder.rb +27 -0
data/test/hpricot/test_parser.rb +426 -0
data/test/hpricot/test_paths.rb +15 -0
data/test/hpricot/test_preserved.rb +77 -0
data/test/hpricot/test_xml.rb +30 -0
data/test/html/sax/test_parser.rb +27 -0
data/test/html/test_builder.rb +89 -0
data/test/html/test_document.rb +150 -0
data/test/html/test_node.rb +21 -0
data/test/test_convert_xpath.rb +185 -0
data/test/test_css_cache.rb +57 -0
data/test/test_gc.rb +15 -0
data/test/test_memory_leak.rb +38 -0
data/test/test_nokogiri.rb +97 -0
data/test/test_reader.rb +222 -0
data/test/test_xslt_transforms.rb +93 -0
data/test/xml/sax/test_parser.rb +95 -0
data/test/xml/test_attr.rb +15 -0
data/test/xml/test_builder.rb +16 -0
data/test/xml/test_cdata.rb +18 -0
data/test/xml/test_comment.rb +16 -0
data/test/xml/test_document.rb +195 -0
data/test/xml/test_dtd.rb +43 -0
data/test/xml/test_node.rb +394 -0
data/test/xml/test_node_set.rb +143 -0
data/test/xml/test_text.rb +13 -0
data/test/xml/test_xpath.rb +105 -0
data/vendor/hoe.rb +1020 -0
metadata +233 -0

@@ -0,0 +1,15 @@
+require File.expand_path(File.join(File.dirname(__FILE__), '..', "helper"))
+require File.join(File.dirname(__FILE__),"load_files")
+class TestParser < Nokogiri::TestCase
+  include Nokogiri
+  def test_roundtrip
+    @basic = Hpricot.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
+    end
+  end
+end

data/test/hpricot/test_preserved.rb ADDED

@@ -0,0 +1,77 @@
+require File.expand_path(File.join(File.dirname(__FILE__), '..', "helper"))
+require File.join(File.dirname(__FILE__),"load_files")
+class TestPreserved < Nokogiri::TestCase
+  def assert_roundtrip str
+    doc = Nokogiri.Hpricot(str)
+    yield doc if block_given?
+    str2 = doc.to_original_html
+    [*str].zip([*str2]).each do |s1, s2|
+      assert_equal s1, s2
+    end
+  end
+  def assert_html str1, str2
+    doc = Nokogiri.Hpricot(str2)
+    yield doc if block_given?
+    assert_equal str1, doc.to_original_html
+  end
+  ####
+  # Not supporting to_original_html
+  #def test_simple
+  #  str = "<p>Hpricot is a <b>you know <i>uh</b> fine thing.</p>"
+  #  assert_html str, str
+  #  assert_html "<p class=\"new\">Hpricot is a <b>you know <i>uh</b> fine thing.</p>", str do |doc|
+  #    (doc/:p).set('class', 'new')
+  #  end
+  #end
+  ####
+  # Not supporting to_original_html
+  #def test_parent
+  #  str = "<html><base href='/'><head><title>Test</title></head><body><div id='wrap'><p>Paragraph one.</p><p>Paragraph two.</p></div></body></html>"
+  #  assert_html str, str
+  #  assert_html "<html><base href='/'><body><div id=\"all\"><div><p>Paragraph one.</p></div><div><p>Paragraph two.</p></div></div></body></html>", str do |doc|
+  #    (doc/:head).remove
+  #    (doc/:div).set('id', 'all')
+  #    (doc/:p).wrap('<div></div>')
+  #  end
+  #end
+  # Not really a valid test.  If libxml can figure out the encoding of the file,
+  # it will use that encoding, otherwise it uses the &#xwhatever so that no data
+  # is lost.
+  #
+  # libxml on OSX can't figure out the encoding, so this tests passes.  linux
+  # can figure out the encoding, so it fails.
+  #def test_escaping_of_contents
+  #  doc = Nokogiri.Hpricot(TestFiles::BOINGBOING)
+  #  assert_equal "Fukuda&#x2019;s Automatic Door opens around your body as you pass through it. The idea is to save energy and keep the room clean.", doc.at("img[@alt='200606131240']").next.to_s.strip
+  #end
+  ####
+  # Modified.  No.
+  #def test_files
+  #  assert_roundtrip TestFiles::BASIC
+  #  assert_roundtrip TestFiles::BOINGBOING
+  #  assert_roundtrip TestFiles::CY0
+  #end
+  ####
+  # Modified..  When calling "to_html" on the document, proper html/doc tags
+  # are produced too.
+  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)
+    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 str, link.to_html
+    # alter the url
+    link['href'] = "javascript:alert(\"AGGA-KA-BOO!\")"
+    assert_equal %{<a href="javascript:alert(&quot;AGGA-KA-BOO!&quot;)">Google</a>}, link.to_html.gsub(/%22/, '&quot;')
+  end
+end

data/test/hpricot/test_xml.rb ADDED

@@ -0,0 +1,30 @@
+require File.expand_path(File.join(File.dirname(__FILE__), '..', "helper"))
+require File.join(File.dirname(__FILE__),"load_files")
+class TestParser < Nokogiri::TestCase
+  include Nokogiri
+  # 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>")
+    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)
+    ### Modified.
+    # I don't want to differentiate pseudo classes from namespaces.  If
+    # you're parsing xml, use XPath.  That's what its for.  :-P
+    assert_equal "hourly", (doc.at "//sy:updatePeriod").content
+    assert_equal 1, (doc/"guid[@isPermaLink]").length
+  end
+  # 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>")
+    assert_equal "City Poisoned", (doc/"title").text
+  end
+end

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

@@ -0,0 +1,27 @@
+require File.expand_path(File.join(File.dirname(__FILE__), '..', '..', "helper"))
+module Nokogiri
+  module HTML
+    module SAX
+      class TestParser < Nokogiri::SAX::TestCase
+        def setup
+          @parser = HTML::SAX::Parser.new(Doc.new)
+        end
+        def test_parse_file
+          @parser.parse_file(HTML_FILE)
+          assert_equal 1110, @parser.document.end_elements.length
+        end
+        def test_parse_document
+          @parser.parse_memory(<<-eoxml)
+            <p>Paragraph 1</p>
+            <p>Paragraph 2</p>
+          eoxml
+          assert_equal([["html", []], ["body", []], ["p", []], ["p", []]],
+                       @parser.document.start_elements)
+        end
+      end
+    end
+  end
+end

data/test/html/test_builder.rb ADDED

@@ -0,0 +1,89 @@
+require File.expand_path(File.join(File.dirname(__FILE__), '..', "helper"))
+module Nokogiri
+  module HTML
+    class TestBuilder < Nokogiri::TestCase
+      def test_hash_as_attributes
+        builder = Nokogiri::HTML::Builder.new do
+          div(:id => 'awesome') {
+            h1 "america"
+          }
+        end
+        assert_equal('<div id="awesome"><h1>america</h1></div>',
+                     builder.to_html.gsub(/\n/, ''))
+      end
+      def test_has_ampersand
+        builder = Nokogiri::HTML::Builder.new do
+          div.rad.thing! {
+            text "<awe&some>"
+            b "hello & world"
+          }
+        end
+        assert_equal(
+          '<div class="rad" id="thing">&lt;awe&amp;some&gt;<b>hello &amp; world</b></div>',
+                     builder.to_html.gsub(/\n/, ''))
+      end
+      def test_multi_tags
+        builder = Nokogiri::HTML::Builder.new do
+          div.rad.thing! {
+            text "<awesome>"
+            b "hello"
+          }
+        end
+        assert_equal(
+          '<div class="rad" id="thing">&lt;awesome&gt;<b>hello</b></div>',
+                     builder.doc.to_html.gsub(/\n/, ''))
+      end
+      def test_attributes_plus_block
+        builder = Nokogiri::HTML::Builder.new do
+          div.rad.thing! {
+            text "<awesome>"
+          }
+        end
+        assert_equal('<div class="rad" id="thing">&lt;awesome&gt;</div>',
+                     builder.doc.to_html.chomp)
+      end
+      def test_builder_adds_attributes
+        builder = Nokogiri::HTML::Builder.new do
+          div.rad.thing! "tender div"
+        end
+        assert_equal('<div class="rad" id="thing">tender div</div>',
+                     builder.doc.to_html.chomp)
+      end
+      def test_bold_tag
+        builder = Nokogiri::HTML::Builder.new do
+          b "bold tag"
+        end
+        assert_equal('<b>bold tag</b>', builder.doc.to_html.chomp)
+      end
+      def test_html_then_body_tag
+        builder = Nokogiri::HTML::Builder.new do
+          html {
+            body {
+              b "bold tag"
+            }
+          }
+        end
+        assert_equal('<html><body><b>bold tag</b></body></html>',
+                     builder.doc.to_html.chomp)
+      end
+      def test_instance_eval_with_delegation_to_block_context
+        class << self
+          def foo
+            "foo!"
+          end
+        end
+        builder = Nokogiri::HTML::Builder.new { text foo }
+        assert builder.to_html.include?("foo!")
+      end
+    end
+  end
+end

data/test/html/test_document.rb ADDED

@@ -0,0 +1,150 @@
+require File.expand_path(File.join(File.dirname(__FILE__), '..', "helper"))
+module Nokogiri
+  module HTML
+    class TestDocument < Nokogiri::TestCase
+      def setup
+        @html = Nokogiri::HTML.parse(File.read(HTML_FILE))
+      end
+      def test_HTML_function
+        html = Nokogiri::HTML(File.read(HTML_FILE))
+        assert html.html?
+      end
+      def test_relative_css
+        html = Nokogiri::HTML(<<-eohtml)
+        <html>
+          <body>
+            <div>
+              <p>inside div tag</p>
+            </div>
+            <p>outside div tag</p>
+          </body>
+        </html>
+        eohtml
+        set = html.search('div').search('p')
+        assert_equal(1, set.length)
+        assert_equal('inside div tag', set.first.inner_text)
+      end
+      def test_multi_css
+        html = Nokogiri::HTML(<<-eohtml)
+        <html>
+          <body>
+            <div>
+              <p>p tag</p>
+              <a>a tag</a>
+            </div>
+          </body>
+        </html>
+        eohtml
+        set = html.css('p, a')
+        assert_equal(2, set.length)
+        assert_equal ['a tag', 'p tag'].sort, set.map { |x| x.content }.sort
+      end
+      def test_inner_text
+        html = Nokogiri::HTML(<<-eohtml)
+        <html>
+          <body>
+            <div>
+              <p>
+                Hello world!
+              </p>
+            </div>
+          </body>
+        </html>
+        eohtml
+        node = html.xpath('//div').first
+        assert_equal('Hello world!', node.inner_text.strip)
+      end
+      def test_inner_html
+        html = Nokogiri::HTML(<<-eohtml)
+        <html>
+          <body>
+            <div>
+              <p>
+                Hello world!
+              </p>
+            </div>
+          </body>
+        </html>
+        eohtml
+        node = html.xpath('//div').first
+        assert_equal('<p>Helloworld!</p>', node.inner_html.gsub(/\s/, ''))
+      end
+      def test_fragment
+        node_set = Nokogiri::HTML.fragment(<<-eohtml)
+          <div>
+            <b>Hello World</b>
+          </div>
+        eohtml
+        assert_equal 1, node_set.length
+        assert_equal 'div', node_set.first.name
+        assert_match(/Hello World/, node_set.to_html)
+      end
+      def test_relative_css_finder
+        doc = Nokogiri::HTML(<<-eohtml)
+          <html>
+            <body>
+              <div class="red">
+                <p>
+                  inside red
+                </p>
+              </div>
+              <div class="green">
+                <p>
+                  inside green
+                </p>
+              </div>
+            </body>
+          </html>
+        eohtml
+        red_divs = doc.css('div.red')
+        assert_equal 1, red_divs.length
+        p_tags = red_divs.first.css('p')
+        assert_equal 1, p_tags.length
+        assert_equal 'inside red', p_tags.first.text.strip
+      end
+      def test_find_classes
+        doc = Nokogiri::HTML(<<-eohtml)
+          <html>
+            <body>
+              <p class="red">RED</p>
+              <p class="awesome red">RED</p>
+              <p class="notred">GREEN</p>
+              <p class="green notred">GREEN</p>
+            </body>
+          </html>
+        eohtml
+        list = doc.css('.red')
+        assert_equal 2, list.length
+        assert_equal %w{ RED RED }, list.map { |x| x.text }
+      end
+      def test_parse_can_take_io
+        html = nil
+        File.open(HTML_FILE, 'rb') { |f|
+          html = Nokogiri::HTML(f)
+        }
+        assert html.html?
+      end
+      def test_html?
+        assert !@html.xml?
+        assert @html.html?
+      end
+      def test_serialize
+        assert @html.serialize
+        assert @html.to_html
+      end
+    end
+  end
+end

data/test/html/test_node.rb ADDED

@@ -0,0 +1,21 @@
+require File.expand_path(File.join(File.dirname(__FILE__), '..', "helper"))
+require 'nkf'
+module Nokogiri
+  module HTML
+    class TestNode < Nokogiri::TestCase
+      def test_to_html_does_not_contain_entities
+        html = NKF.nkf("-e --msdos", <<-EOH)
+        <html><body>
+        <p> test paragraph
+        foo bar </p>
+        </body></html>
+        EOH
+        nokogiri = Nokogiri::HTML.parse(html)
+        assert_equal "<p>testparagraph\r\nfoobar</p>",
+          nokogiri.at("p").to_html.gsub(/ /, '')
+      end
+    end
+  end
+end

data/test/test_convert_xpath.rb ADDED

@@ -0,0 +1,185 @@
+require File.expand_path(File.join(File.dirname(__FILE__), "helper"))
+begin
+  require 'rubygems'
+  require 'hpricot'
+  HAS_HPRICOT = true
+rescue LoadError
+  HAS_HPRICOT = false
+end
+class TestConvertXPath < Nokogiri::TestCase
+  def setup
+    @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)), "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
+    assert_syntactical_equivalence("h1[a]", ".//h1[child::a]", "Tender Lovemaking") do |j|
+      j.inner_text
+    end
+  end
+  def test_child_tag_equals
+    assert_syntactical_equivalence("h1[a='Tender Lovemaking']", ".//h1[child::a = 'Tender Lovemaking']", "Tender Lovemaking") do |j|
+      j.inner_text
+    end
+  end
+  def test_filter_contains
+    assert_syntactical_equivalence("title:contains('Tender')", ".//title[contains(., 'Tender')]",
+                                   "Tender Lovemaking  ") do |j|
+      j.inner_text
+    end
+  end
+  def test_filter_comment
+    assert_syntactical_equivalence("div comment()[2]", ".//div//comment()[position() = 2]", "<!-- end of header -->") do |j|
+      j.first.to_s
+    end
+  end
+  def test_filter_text
+    assert_syntactical_equivalence("a[text()]", ".//a[normalize-space(child::text())]", "<a href=\"http://tenderlovemaking.com\">Tender Lovemaking</a>") do |j|
+      j.first.to_s
+    end
+    assert_syntactical_equivalence("a[text()='Tender Lovemaking']", ".//a[normalize-space(child::text()) = 'Tender Lovemaking']", "<a href=\"http://tenderlovemaking.com\">Tender Lovemaking</a>") do |j|
+      j.first.to_s
+    end
+    assert_syntactical_equivalence("a/text()", ".//a/child::text()", "Tender Lovemaking") do |j|
+      j.first.to_s
+    end
+    assert_syntactical_equivalence("h2//a[text()!='Back Home!']", ".//h2//a[normalize-space(child::text()) != 'Back Home!']", "Meow meow meow meow meow") do |j|
+      j.first.inner_text
+    end
+  end
+  def test_filter_by_attr
+    assert_syntactical_equivalence("a[@href='http://blog.geminigeek.com/wordpress-theme']",
+                                   ".//a[@href = 'http://blog.geminigeek.com/wordpress-theme']",
+                                   "http://blog.geminigeek.com/wordpress-theme") do |j|
+      j.first["href"]
+    end
+  end
+  def test_css_id
+    assert_syntactical_equivalence("#linkcat-7", ".//*[@id = 'linkcat-7']", "linkcat-7") do |j|
+      j.first["id"]
+    end
+    assert_syntactical_equivalence("li#linkcat-7", ".//li[@id = 'linkcat-7']", "linkcat-7") do |j|
+      j.first["id"]
+    end
+  end
+  def test_css_class
+    assert_syntactical_equivalence(".cat-item-15", ".//*[contains(concat(' ', @class, ' '), ' cat-item-15 ')]",
+                                   "cat-item cat-item-15") do |j|
+      j.first["class"]
+    end
+    assert_syntactical_equivalence("li.cat-item-15", ".//li[contains(concat(' ', @class, ' '), ' cat-item-15 ')]",
+                                   "cat-item cat-item-15") do |j|
+      j.first["class"]
+    end
+  end
+  def test_css_tags
+    assert_syntactical_equivalence("div li a", ".//div//li//a", "http://brobinius.org/") do |j|
+      j.first.inner_text
+    end
+    assert_syntactical_equivalence("div li > a", ".//div//li/a", "http://brobinius.org/") do |j|
+      j.first.inner_text
+    end
+    assert_syntactical_equivalence("h1 ~ small", ".//small[preceding-sibling::h1]", "The act of making love, tenderly.") do |j|
+      j.first.inner_text
+    end
+    assert_syntactical_equivalence("h1 ~ small", ".//small[preceding-sibling::h1]", "The act of making love, tenderly.") do |j|
+      j.first.inner_text
+    end
+  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
+    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
+    assert_syntactical_equivalence("div//a:last()", ".//div//a[position() = last()]", "Wordpress") do |j|
+      j.last.inner_text
+    end
+    assert_syntactical_equivalence("div//a:last", ".//div//a[position() = last()]", "Wordpress") do |j|
+      j.last.inner_text
+    end
+  end
+  def test_multiple_filters
+    assert_syntactical_equivalence("a[@rel='bookmark'][1]", ".//a[@rel = 'bookmark' and position() = 1]", "Back Home!") do |j|
+      j.first.inner_text
+    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)
+#       doc/'p[text()*="final"]' -> class includes string (index) (broken: always returns true?)
+#       doc/'p[text()$="final"]' -> /final$/
+#       doc/'p[text()|="final"]' -> /^final$/
+#       doc/'p[text()^="final"]' -> string starts with 'final
+#       nth_first
+#       nth_last
+#       even
+#       odd
+#       first-child, nth-child, last-child, nth-last-child, nth-last-of-type
+#       only-of-type, only-child
+#       parent
+#       empty
+#       root
+end