nokogiri 1.1.0 → 1.1.1

data/lib/nokogiri/decorators/slop.rb

@@ -12,8 +12,8 @@ module Nokogiri
           if hash[:css]
             list = css("#{name}#{hash[:css]}")
           elsif hash[:xpath]
-            conds = Array(hash[:xpath]).collect{|j| "[#{j}]"}
-            list = xpath("./#{name}#{conds}")
+            conds = Array(hash[:xpath]).join(' and ')
+            list = xpath("./#{name}[#{conds}]")
           end
         else
           CSS::Parser.without_cache do

data/lib/nokogiri/version.rb

@@ -1,3 +1,3 @@
 module Nokogiri
-  VERSION = '1.1.0'
+  VERSION = '1.1.1'
 end

data/lib/nokogiri/xml.rb

@@ -1,8 +1,8 @@
 require 'nokogiri/xml/sax'
-require 'nokogiri/xml/xpath_handler'
 require 'nokogiri/xml/before_handler'
 require 'nokogiri/xml/after_handler'
 require 'nokogiri/xml/node'
+require 'nokogiri/xml/attr'
 require 'nokogiri/xml/dtd'
 require 'nokogiri/xml/text'
 require 'nokogiri/xml/cdata'

data/lib/nokogiri/xml/attr.rb

@@ -0,0 +1,10 @@
+module Nokogiri
+  module XML
+    class Attr < Node
+      def value
+        children.first.to_s
+      end
+      alias :to_s :value
+    end
+  end
+end

data/lib/nokogiri/xml/document.rb

@@ -16,7 +16,6 @@ module Nokogiri
 
       ###
       # Explore a document with shortcut methods.
-
       def slop!
         unless decorators(XML::Node).include? Nokogiri::Decorators::Slop
           decorators(XML::Node) << Nokogiri::Decorators::Slop

data/lib/nokogiri/xml/node.rb

@@ -40,9 +40,9 @@ module Nokogiri
         first = self.child
         return list unless first # Empty list
 
-        list << first unless first.blank?
+        list << first
         while first = first.next
-          list << first unless first.blank?
+          list << first
         end
         list
       end
@@ -54,6 +54,7 @@ module Nokogiri
       def search *paths
         ns = paths.last.is_a?(Hash) ? paths.pop : {}
         xpath(*(paths.map { |path|
+          path = path.to_s
           path =~ /^(\.\/|\/)/ ? path : CSS.xpath_for(path, :prefix => ".//")
         }.flatten.uniq) + [ns])
       end
@@ -68,17 +69,21 @@ module Nokogiri
       #   node.xpath('.//xmlns:name', node.root.namespaces)
       #
       # Custom XPath functions may also be defined.  To define custom functions
-      # create a class which subclasses XPathHandler and implement the
-      # function you want to define.  For example:
+      # create a class and implement the # function you want to define.
+      # For example:
       #
-      #   node.xpath('.//title[regex(., "\w+")]', Class.new(XPathHandler) {
+      #   node.xpath('.//title[regex(., "\w+")]', Class.new {
       #     def regex node_set, regex
       #       node_set.find_all { |node| node['some_attribute'] =~ /#{regex}/ }
       #     end
       #   })
       #
       def xpath *paths
-        handler = paths.last.is_a?(XPathHandler) ? paths.pop : nil
+        # Pop off our custom function handler if it exists
+        handler = ![
+          Hash, String, Symbol
+        ].include?(paths.last.class) ? paths.pop : nil
+
         ns = paths.last.is_a?(Hash) ? paths.pop : {}
 
         return NodeSet.new(document) unless document.root
@@ -112,19 +117,23 @@ module Nokogiri
       #   node.css('div + p.green', 'div#one')
       #
       # Custom CSS pseudo classes may also be defined.  To define custom pseudo
-      # classes, create a class which subclasses SelectorHandler and implement
-      # the the custom pseudo class you want defined.  The first argument to
-      # the method will be the current matching NodeSet.  Any other arguments
-      # are ones that you pass in.  For example:
+      # classes, create a class and implement the custom pseudo class you
+      # want defined.  The first argument to the method will be the current
+      # matching NodeSet.  Any other arguments are ones that you pass in.
+      # For example:
       #
-      #   node.css('title:regex("\w+")', Class.new(SelectorHandler) {
+      #   node.css('title:regex("\w+")', Class.new {
       #     def regex node_set, regex
       #       node_set.find_all { |node| node['some_attribute'] =~ /#{regex}/ }
       #     end
       #   })
       #
       def css *rules
-        handler = rules.last.is_a?(XPathHandler) ? rules.pop : nil
+        # Pop off our custom function handler if it exists
+        handler = ![
+          Hash, String, Symbol
+        ].include?(rules.last.class) ? rules.pop : nil
+
         ns = rules.last.is_a?(Hash) ? rules.pop : {}
 
         rules = rules.map { |rule|
@@ -135,7 +144,7 @@ module Nokogiri
       end
 
       def at path, ns = {}
-        search("#{path}", ns).first
+        search(path, ns).first
       end
 
       def [](property)
@@ -147,10 +156,30 @@ module Nokogiri
         next_sibling
       end
 
+      def previous
+        previous_sibling
+      end
+
       def remove
         unlink
       end
 
+      ####
+      # Returns a hash containing the node's attributes.  The key is the
+      # attribute name, the value is the string value of the attribute.
+      def attributes
+        Hash[*(attribute_nodes.map { |node|
+          [node.name, node]
+        }.flatten)]
+      end
+
+      ###
+      # Remove the attribute named +name+
+      def remove_attribute name
+        attributes[name].remove if key? name
+      end
+      alias :delete :remove_attribute
+
       ####
       # Create nodes from +data+ and insert them before this node
       # (as a sibling).
@@ -225,10 +254,23 @@ module Nokogiri
         type == HTML_DOCUMENT_NODE
       end
 
-      def to_html
-        to_xml
+      def text?
+        type == TEXT_NODE
+      end
+
+      def read_only?
+        # According to gdome2, these are read-only node types
+        [NOTATION_NODE, ENTITY_NODE, ENTITY_DECL].include?(type)
+      end
+
+      def element?
+        type == ELEMENT_NODE
+      end
+      alias :elem? :element?
+
+      def to_s
+        document.xml? ? to_xml : to_html
       end
-      alias :to_s :to_html
 
       def inner_html
         children.map { |x| x.to_html }.join
@@ -280,6 +322,12 @@ Node.replace requires a Node argument, and cannot accept a Document.
         end
         replace_with_node new_node
       end
+
+      def == other
+        return false unless other
+        return false unless other.respond_to?(:pointer_id)
+        pointer_id == other.pointer_id
+      end
     end
   end
 end

data/lib/nokogiri/xml/node_set.rb

@@ -152,6 +152,12 @@ module Nokogiri
       end
       alias :text :inner_text
 
+      ###
+      # Get the inner html of all contained Node objects
+      def inner_html
+        collect{|j| j.inner_html}.join('')
+      end
+
       ###
       # Wrap this NodeSet with +html+ or the results of the builder in +blk+
       def wrap(html, &blk)
@@ -175,6 +181,10 @@ module Nokogiri
         map { |x| x.to_html }.join('')
       end
 
+      def to_xml *args
+        map { |x| x.to_xml(*args) }.join('')
+      end
+
       def size
         length
       end

data/lib/nokogiri/xml/reader.rb

@@ -3,6 +3,12 @@ module Nokogiri
     class Reader
       include Enumerable
 
+      def attributes
+        Hash[*(attribute_nodes.map { |node|
+          [node.name, node.to_s]
+        }.flatten)].merge(namespaces || {})
+      end
+
       def each(&block)
         while node = self.read
           block.call(node)

data/test/hpricot/test_alter.rb

@@ -11,13 +11,13 @@ class TestAlter < Nokogiri::TestCase
   def test_before
     test0 = "<link rel='stylesheet' href='test0.css' />"
     @basic.at("link").before(test0)
-    assert_equal 'test0.css', @basic.at("link").attributes['href']
+    assert_equal 'test0.css', @basic.at("link").attributes['href'].to_s
   end
 
   def test_after
     test_inf = "<link rel='stylesheet' href='test_inf.css' />"
     @basic.search("link")[-1].after(test_inf)
-    assert_equal 'test_inf.css', @basic.search("link")[-1].attributes['href']
+    assert_equal 'test_inf.css', @basic.search("link")[-1]['href']
   end
 
   def test_wrap
@@ -29,10 +29,10 @@ class TestAlter < Nokogiri::TestCase
   def test_add_class
     first_p = (@basic/"p:first").add_class("testing123")
     assert first_p[0].get_attribute("class").split(" ").include?("testing123")
-    assert((Nokogiri.Hpricot(@basic.to_html)/"p:first")[0].attributes["class"].split(" ").include?("testing123"))
+    assert((Nokogiri.Hpricot(@basic.to_html)/"p:first")[0]["class"].split(" ").include?("testing123"))
     ####
     # Modified.  We do not support OB1 bug.
-    assert !(Nokogiri.Hpricot(@basic.to_html)/"p:gt(1)")[0].attributes["class"].split(" ").include?("testing123")
+    assert !(Nokogiri.Hpricot(@basic.to_html)/"p:gt(1)")[0]["class"].split(" ").include?("testing123")
   end
   
   def test_change_attributes
@@ -40,9 +40,9 @@ class TestAlter < Nokogiri::TestCase
     all_as = (@basic/"a").attr("href", "http://my_new_href.com")
     all_lb = (@basic/"link").attr("href") { |e| e.name }
     GC.start # try to shake out GC bugs with xpath and node sets.
-    assert_changed(@basic, "p", all_ps) {|p| p.attributes["title"] == "Some Title"}
-    assert_changed(@basic, "a", all_as) {|a| a.attributes["href"] == "http://my_new_href.com"}
-    assert_changed(@basic, "link", all_lb) {|a| a.attributes["href"] == "link" }
+    assert_changed(@basic, "p", all_ps) {|p| p.attributes["title"].to_s == "Some Title"}
+    assert_changed(@basic, "a", all_as) {|a| a.attributes["href"].to_s == "http://my_new_href.com"}
+    assert_changed(@basic, "link", all_lb) {|a| a.attributes["href"].to_s == "link" }
   end
   
   def test_remove_attr

data/test/hpricot/test_parser.rb

@@ -33,7 +33,7 @@ class TestParser < Nokogiri::TestCase
   
   def test_filter_contains
     @basic = Hpricot.parse(TestFiles::BASIC)
-    assert_equal '<title>Sample XHTML</title>', @basic.search("title:contains('Sample')").to_s
+    assert_equal '<title>Sample XHTML</title>', @basic.search("title:contains('Sample')").to_s.chomp
   end
 
   def test_get_element_by_id
@@ -98,9 +98,9 @@ class TestParser < Nokogiri::TestCase
 
   def test_positional
     h = Nokogiri.Hpricot( "<div><br/><p>one</p><p>two</p></div>" )
-    assert_equal "<p>one</p>", h.search("div/p:eq(1)").to_s # MODIFIED: eq(0) -> eq(1), and removed initial '//'
-    assert_equal "<p>one</p>", h.search("div/p:first").to_s # MODIFIED: removed initial '//'
-    assert_equal "<p>one</p>", h.search("div/p:first()").to_s # MODIFIED: removed initial '//'
+    assert_equal "<p>one</p>", h.search("div/p:eq(1)").to_s.chomp # MODIFIED: eq(0) -> eq(1), and removed initial '//'
+    assert_equal "<p>one</p>", h.search("div/p:first").to_s.chomp # MODIFIED: removed initial '//'
+    assert_equal "<p>one</p>", h.search("div/p:first()").to_s.chomp # MODIFIED: removed initial '//'
   end
 
   def test_pace
@@ -209,7 +209,7 @@ class TestParser < Nokogiri::TestCase
         @basic.search('p:eq(3)').to_html # under Hpricot this was eq(2)
     ##
     #  MODIFIED: to be blank-agnostic, because Nokogiri's to_html is slightly different compared to Hpricot.
-    assert_equal '<p class="last final"> <b>THE FINAL PARAGRAPH</b> </p>',
+    assert_equal '<p class="last final"><b>THE FINAL PARAGRAPH</b></p>',
         @basic.search('p:last').to_html.gsub(/\s+/,' ')
     assert_equal 'last final', @basic.search('p:last-of-type').first.get_attribute('class') # MODIFIED to not have '//' prefix
   end
@@ -409,7 +409,8 @@ class TestParser < Nokogiri::TestCase
   def test_keep_cdata
     str = %{<script> /*<![CDATA[*/
     /*]]>*/ </script>}
-    assert_match str, Nokogiri.Hpricot(str).to_html
+    # MODIFIED: if you want the cdata, to_xml it
+    assert_match str, Nokogiri.Hpricot(str).to_xml
   end
 
   def test_namespace

data/test/hpricot/test_preserved.rb

@@ -66,7 +66,6 @@ class TestPreserved < Nokogiri::TestCase
     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.attributes['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

data/test/html/test_node.rb

@@ -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/xml/test_attr.rb

@@ -0,0 +1,15 @@
+require File.expand_path(File.join(File.dirname(__FILE__), '..', "helper"))
+
+module Nokogiri
+  module XML
+    class TestAttr < Nokogiri::TestCase
+      def test_unlink
+        xml = Nokogiri::XML.parse(File.read(XML_FILE), XML_FILE)
+        address = xml.xpath('/staff/employee/address').first
+        assert_equal 'Yes', address['domestic']
+        address.attribute_nodes.first.unlink
+        assert_nil address['domestic']
+      end
+    end
+  end
+end

data/test/xml/test_node.rb

@@ -11,6 +11,44 @@ module Nokogiri
           address.ancestors.map { |x| x.name }
       end
 
+      def test_read_only?
+        xml = Nokogiri::XML.parse(File.read(XML_FILE), XML_FILE)
+        assert entity_decl = xml.internal_subset.children.find { |x|
+          x.type == Node::ENTITY_DECL
+        }
+        assert entity_decl.read_only?
+      end
+
+      def test_remove_attribute
+        xml = Nokogiri::XML.parse(File.read(XML_FILE), XML_FILE)
+        address = xml.xpath('/staff/employee/address').first
+        assert_equal 'Yes', address['domestic']
+        address.remove_attribute 'domestic'
+        assert_nil address['domestic']
+      end
+
+      def test_delete
+        xml = Nokogiri::XML.parse(File.read(XML_FILE), XML_FILE)
+        address = xml.xpath('/staff/employee/address').first
+        assert_equal 'Yes', address['domestic']
+        address.delete 'domestic'
+        assert_nil address['domestic']
+      end
+
+      def test_angry_add_child
+        xml = Nokogiri::XML.parse(File.read(XML_FILE), XML_FILE)
+        child = xml.css('employee').first
+
+        assert new_child = child.children.first
+
+        last = child.children.last
+
+        # Magic!  Don't try this at home folks
+        child.add_child(new_child)
+        assert_equal new_child, child.children.last
+        assert_equal last, child.children.last
+      end
+
       def test_add_child
         xml = Nokogiri::XML(<<-eoxml)
         <root>
@@ -18,6 +56,7 @@ module Nokogiri
         </root>
         eoxml
         text_node = Nokogiri::XML::Text.new('hello', xml)
+        assert_equal Nokogiri::XML::Node::TEXT_NODE, text_node.type
         xml.root.add_child text_node
         assert_match 'hello', xml.to_s
       end
@@ -40,12 +79,49 @@ module Nokogiri
         </root>
         eoxml
         b_node = Nokogiri::XML::Node.new('a', xml)
+        assert_equal Nokogiri::XML::Node::ELEMENT_NODE, b_node.type
         b_node.content = 'first'
         a_node = xml.xpath('//a').first
         a_node.add_previous_sibling(b_node)
         assert_equal('first', xml.xpath('//a').first.text)
       end
 
+      def test_add_previous_sibling_merge
+        xml = Nokogiri::XML(<<-eoxml)
+        <root>
+          <a>Hello world</a>
+        </root>
+        eoxml
+
+        assert a_tag = xml.css('a').first
+
+        left_space = a_tag.previous
+        right_space = a_tag.next
+        assert left_space.text?
+        assert right_space.text?
+
+        left_space.add_previous_sibling(right_space)
+        assert_equal left_space, right_space
+      end
+
+      def test_add_next_sibling_merge
+        xml = Nokogiri::XML(<<-eoxml)
+        <root>
+          <a>Hello world</a>
+        </root>
+        eoxml
+
+        assert a_tag = xml.css('a').first
+
+        left_space = a_tag.previous
+        right_space = a_tag.next
+        assert left_space.text?
+        assert right_space.text?
+
+        right_space.add_next_sibling(left_space)
+        assert_equal left_space, right_space
+      end
+
       def test_find_by_css_with_tilde_eql
         xml = Nokogiri::XML.parse(<<-eoxml)
         <root>
@@ -177,8 +253,9 @@ module Nokogiri
         assert node = xml.search('//address')[2]
         attr = node.attributes
         assert_equal 2, attr.size
-        assert_equal 'Yes', attr['domestic']
-        assert_equal 'No', attr['street']
+        assert_equal 'Yes', attr['domestic'].value
+        assert_equal 'Yes', attr['domestic'].to_s
+        assert_equal 'No', attr['street'].value
       end
 
       def test_path
@@ -188,6 +265,15 @@ module Nokogiri
         assert_equal('/staff/employee[1]', node.path)
       end
 
+      def test_search_by_symbol
+        xml = Nokogiri::XML.parse(File.read(XML_FILE), XML_FILE)
+        assert set = xml.search(:employee)
+        assert 5, set.length
+
+        assert node = xml.at(:employee)
+        assert node.text =~ /EMP0001/
+      end
+
       def test_new_node
         xml = Nokogiri::XML.parse(File.read(XML_FILE), XML_FILE)
         node = Nokogiri::XML::Node.new('form', xml)
@@ -204,6 +290,13 @@ module Nokogiri
         assert_equal('hello world!', node.content)
       end
 
+      def test_whitespace_nodes
+        doc = Nokogiri::XML.parse("<root><b>Foo</b>\n<i>Bar</i> <p>Bazz</p></root>")
+        children = doc.at('//root').children.collect{|j| j.to_s}
+        assert_equal "\n", children[1]
+        assert_equal " ", children[3]
+      end
+
       def test_replace
         xml = Nokogiri::XML.parse(File.read(XML_FILE))
         set = xml.search('//employee')
@@ -278,6 +371,24 @@ EOF
         assert_equal "hello c", xml.search("//c:div", xml.namespaces).first.inner_text
       end
 
+      def test_namespace
+        xml = Nokogiri::XML.parse(<<-EOF)
+<x xmlns:a='http://foo.com/' xmlns:b='http://bar.com/'>
+  <y xmlns:c='http://bazz.com/'>
+    <a:div>hello a</a:div>
+    <b:div>hello b</b:div>
+    <c:div>hello c</c:div>
+    <div>hello moon</div>
+  </y>  
+</x>
+EOF
+        set = xml.search("//y/*")
+        assert_equal "a", set[0].namespace
+        assert_equal "b", set[1].namespace
+        assert_equal "c", set[2].namespace
+        assert_equal nil, set[3].namespace
+      end
+
     end
   end
 end