RubyGems - nokogiri - Versions diffs - 1.2.3-x86-mswin32-60 → 1.4.5-x86-mswin32-60 - Mend

nokogiri 1.2.3-x86-mswin32-60 → 1.4.5-x86-mswin32-60

Potentially problematic release.

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

Files changed (319) hide show

data/.autotest +18 -7
data/.gemtest +0 -0
data/CHANGELOG.ja.rdoc +297 -3
data/CHANGELOG.rdoc +289 -0
data/Manifest.txt +148 -37
data/README.ja.rdoc +20 -20
data/README.rdoc +53 -22
data/Rakefile +127 -211
data/bin/nokogiri +54 -0
data/ext/nokogiri/depend +358 -0
data/ext/nokogiri/extconf.rb +89 -54
data/ext/nokogiri/html_document.c +34 -27
data/ext/nokogiri/html_document.h +1 -1
data/ext/nokogiri/html_element_description.c +276 -0
data/ext/nokogiri/html_element_description.h +10 -0
data/ext/nokogiri/html_entity_lookup.c +7 -5
data/ext/nokogiri/html_entity_lookup.h +1 -1
data/ext/nokogiri/html_sax_parser_context.c +94 -0
data/ext/nokogiri/html_sax_parser_context.h +11 -0
data/ext/nokogiri/{native.c → nokogiri.c} +31 -7
data/ext/nokogiri/{native.h → nokogiri.h} +68 -41
data/ext/nokogiri/xml_attr.c +20 -9
data/ext/nokogiri/xml_attr.h +1 -1
data/ext/nokogiri/xml_attribute_decl.c +70 -0
data/ext/nokogiri/xml_attribute_decl.h +9 -0
data/ext/nokogiri/xml_cdata.c +21 -9
data/ext/nokogiri/xml_cdata.h +1 -1
data/ext/nokogiri/xml_comment.c +18 -6
data/ext/nokogiri/xml_comment.h +1 -1
data/ext/nokogiri/xml_document.c +247 -68
data/ext/nokogiri/xml_document.h +5 -3
data/ext/nokogiri/xml_document_fragment.c +15 -7
data/ext/nokogiri/xml_document_fragment.h +1 -1
data/ext/nokogiri/xml_dtd.c +110 -10
data/ext/nokogiri/xml_dtd.h +3 -1
data/ext/nokogiri/xml_element_content.c +123 -0
data/ext/nokogiri/xml_element_content.h +10 -0
data/ext/nokogiri/xml_element_decl.c +69 -0
data/ext/nokogiri/xml_element_decl.h +9 -0
data/ext/nokogiri/xml_encoding_handler.c +79 -0
data/ext/nokogiri/xml_encoding_handler.h +8 -0
data/ext/nokogiri/xml_entity_decl.c +110 -0
data/ext/nokogiri/xml_entity_decl.h +10 -0
data/ext/nokogiri/xml_entity_reference.c +16 -5
data/ext/nokogiri/xml_entity_reference.h +1 -1
data/ext/nokogiri/xml_io.c +40 -8
data/ext/nokogiri/xml_io.h +2 -1
data/ext/nokogiri/xml_libxml2_hacks.c +112 -0
data/ext/nokogiri/xml_libxml2_hacks.h +12 -0
data/ext/nokogiri/xml_namespace.c +84 -0
data/ext/nokogiri/xml_namespace.h +13 -0
data/ext/nokogiri/xml_node.c +782 -225
data/ext/nokogiri/xml_node.h +2 -4
data/ext/nokogiri/xml_node_set.c +253 -34
data/ext/nokogiri/xml_node_set.h +2 -2
data/ext/nokogiri/xml_processing_instruction.c +17 -5
data/ext/nokogiri/xml_processing_instruction.h +1 -1
data/ext/nokogiri/xml_reader.c +277 -85
data/ext/nokogiri/xml_reader.h +1 -1
data/ext/nokogiri/xml_relax_ng.c +168 -0
data/ext/nokogiri/xml_relax_ng.h +9 -0
data/ext/nokogiri/xml_sax_parser.c +183 -111
data/ext/nokogiri/xml_sax_parser.h +30 -1
data/ext/nokogiri/xml_sax_parser_context.c +199 -0
data/ext/nokogiri/xml_sax_parser_context.h +10 -0
data/ext/nokogiri/xml_sax_push_parser.c +42 -12
data/ext/nokogiri/xml_sax_push_parser.h +1 -1
data/ext/nokogiri/xml_schema.c +205 -0
data/ext/nokogiri/xml_schema.h +9 -0
data/ext/nokogiri/xml_syntax_error.c +28 -173
data/ext/nokogiri/xml_syntax_error.h +2 -1
data/ext/nokogiri/xml_text.c +16 -6
data/ext/nokogiri/xml_text.h +1 -1
data/ext/nokogiri/xml_xpath_context.c +104 -47
data/ext/nokogiri/xml_xpath_context.h +1 -1
data/ext/nokogiri/xslt_stylesheet.c +161 -19
data/ext/nokogiri/xslt_stylesheet.h +1 -1
data/lib/nokogiri.rb +47 -8
data/lib/nokogiri/1.8/nokogiri.so +0 -0
data/lib/nokogiri/1.9/nokogiri.so +0 -0
data/lib/nokogiri/css.rb +6 -3
data/lib/nokogiri/css/node.rb +14 -12
data/lib/nokogiri/css/parser.rb +665 -62
data/lib/nokogiri/css/parser.y +20 -10
data/lib/nokogiri/css/parser_extras.rb +91 -0
data/lib/nokogiri/css/tokenizer.rb +148 -5
data/lib/nokogiri/css/tokenizer.rex +10 -9
data/lib/nokogiri/css/xpath_visitor.rb +47 -44
data/lib/nokogiri/decorators/slop.rb +8 -4
data/lib/nokogiri/ffi/encoding_handler.rb +42 -0
data/lib/nokogiri/ffi/html/document.rb +28 -0
data/lib/nokogiri/ffi/html/element_description.rb +81 -0
data/lib/nokogiri/ffi/html/entity_lookup.rb +16 -0
data/lib/nokogiri/ffi/html/sax/parser_context.rb +38 -0
data/lib/nokogiri/ffi/io_callbacks.rb +42 -0
data/lib/nokogiri/ffi/libxml.rb +420 -0
data/lib/nokogiri/ffi/structs/common_node.rb +38 -0
data/lib/nokogiri/ffi/structs/html_elem_desc.rb +24 -0
data/lib/nokogiri/ffi/structs/html_entity_desc.rb +13 -0
data/lib/nokogiri/ffi/structs/xml_alloc.rb +16 -0
data/lib/nokogiri/ffi/structs/xml_attr.rb +20 -0
data/lib/nokogiri/ffi/structs/xml_attribute.rb +27 -0
data/lib/nokogiri/ffi/structs/xml_buffer.rb +16 -0
data/lib/nokogiri/ffi/structs/xml_char_encoding_handler.rb +11 -0
data/lib/nokogiri/ffi/structs/xml_document.rb +117 -0
data/lib/nokogiri/ffi/structs/xml_dtd.rb +28 -0
data/lib/nokogiri/ffi/structs/xml_element.rb +26 -0
data/lib/nokogiri/ffi/structs/xml_element_content.rb +17 -0
data/lib/nokogiri/ffi/structs/xml_entity.rb +32 -0
data/lib/nokogiri/ffi/structs/xml_enumeration.rb +12 -0
data/lib/nokogiri/ffi/structs/xml_node.rb +28 -0
data/lib/nokogiri/ffi/structs/xml_node_set.rb +53 -0
data/lib/nokogiri/ffi/structs/xml_notation.rb +11 -0
data/lib/nokogiri/ffi/structs/xml_ns.rb +15 -0
data/lib/nokogiri/ffi/structs/xml_parser_context.rb +20 -0
data/lib/nokogiri/ffi/structs/xml_parser_input.rb +19 -0
data/lib/nokogiri/ffi/structs/xml_relax_ng.rb +14 -0
data/lib/nokogiri/ffi/structs/xml_sax_handler.rb +51 -0
data/lib/nokogiri/ffi/structs/xml_sax_push_parser_context.rb +124 -0
data/lib/nokogiri/ffi/structs/xml_schema.rb +13 -0
data/lib/nokogiri/ffi/structs/xml_syntax_error.rb +31 -0
data/lib/nokogiri/ffi/structs/xml_text_reader.rb +12 -0
data/lib/nokogiri/ffi/structs/xml_xpath_context.rb +38 -0
data/lib/nokogiri/ffi/structs/xml_xpath_object.rb +35 -0
data/lib/nokogiri/ffi/structs/xml_xpath_parser_context.rb +20 -0
data/lib/nokogiri/ffi/structs/xslt_stylesheet.rb +13 -0
data/lib/nokogiri/ffi/weak_bucket.rb +40 -0
data/lib/nokogiri/ffi/xml/attr.rb +41 -0
data/lib/nokogiri/ffi/xml/attribute_decl.rb +27 -0
data/lib/nokogiri/ffi/xml/cdata.rb +19 -0
data/lib/nokogiri/ffi/xml/comment.rb +18 -0
data/lib/nokogiri/ffi/xml/document.rb +174 -0
data/lib/nokogiri/ffi/xml/document_fragment.rb +21 -0
data/lib/nokogiri/ffi/xml/dtd.rb +67 -0
data/lib/nokogiri/ffi/xml/element_content.rb +43 -0
data/lib/nokogiri/ffi/xml/element_decl.rb +19 -0
data/lib/nokogiri/ffi/xml/entity_decl.rb +36 -0
data/lib/nokogiri/ffi/xml/entity_reference.rb +19 -0
data/lib/nokogiri/ffi/xml/namespace.rb +44 -0
data/lib/nokogiri/ffi/xml/node.rb +559 -0
data/lib/nokogiri/ffi/xml/node_set.rb +150 -0
data/lib/nokogiri/ffi/xml/processing_instruction.rb +20 -0
data/lib/nokogiri/ffi/xml/reader.rb +236 -0
data/lib/nokogiri/ffi/xml/relax_ng.rb +85 -0
data/lib/nokogiri/ffi/xml/sax/parser.rb +143 -0
data/lib/nokogiri/ffi/xml/sax/parser_context.rb +79 -0
data/lib/nokogiri/ffi/xml/sax/push_parser.rb +51 -0
data/lib/nokogiri/ffi/xml/schema.rb +109 -0
data/lib/nokogiri/ffi/xml/syntax_error.rb +98 -0
data/lib/nokogiri/ffi/xml/text.rb +18 -0
data/lib/nokogiri/ffi/xml/xpath.rb +9 -0
data/lib/nokogiri/ffi/xml/xpath_context.rb +153 -0
data/lib/nokogiri/ffi/xslt/stylesheet.rb +77 -0
data/lib/nokogiri/html.rb +13 -47
data/lib/nokogiri/html/builder.rb +27 -1
data/lib/nokogiri/html/document.rb +201 -7
data/lib/nokogiri/html/document_fragment.rb +41 -0
data/lib/nokogiri/html/element_description.rb +23 -0
data/lib/nokogiri/html/entity_lookup.rb +2 -0
data/lib/nokogiri/html/sax/parser.rb +34 -3
data/lib/nokogiri/html/sax/parser_context.rb +16 -0
data/lib/nokogiri/nokogiri.rb +1 -0
data/lib/nokogiri/version.rb +40 -1
data/lib/nokogiri/version_warning.rb +14 -0
data/lib/nokogiri/xml.rb +32 -53
data/lib/nokogiri/xml/attr.rb +5 -0
data/lib/nokogiri/xml/attribute_decl.rb +18 -0
data/lib/nokogiri/xml/builder.rb +349 -29
data/lib/nokogiri/xml/cdata.rb +3 -1
data/lib/nokogiri/xml/character_data.rb +7 -0
data/lib/nokogiri/xml/document.rb +166 -14
data/lib/nokogiri/xml/document_fragment.rb +76 -1
data/lib/nokogiri/xml/dtd.rb +16 -3
data/lib/nokogiri/xml/element_content.rb +36 -0
data/lib/nokogiri/xml/element_decl.rb +13 -0
data/lib/nokogiri/xml/entity_decl.rb +19 -0
data/lib/nokogiri/xml/namespace.rb +13 -0
data/lib/nokogiri/xml/node.rb +561 -166
data/lib/nokogiri/xml/node/save_options.rb +22 -2
data/lib/nokogiri/xml/node_set.rb +202 -40
data/lib/nokogiri/xml/parse_options.rb +93 -0
data/lib/nokogiri/xml/pp.rb +2 -0
data/lib/nokogiri/xml/pp/character_data.rb +18 -0
data/lib/nokogiri/xml/pp/node.rb +56 -0
data/lib/nokogiri/xml/processing_instruction.rb +2 -0
data/lib/nokogiri/xml/reader.rb +93 -8
data/lib/nokogiri/xml/relax_ng.rb +32 -0
data/lib/nokogiri/xml/sax.rb +1 -7
data/lib/nokogiri/xml/sax/document.rb +107 -2
data/lib/nokogiri/xml/sax/parser.rb +57 -7
data/lib/nokogiri/xml/sax/parser_context.rb +16 -0
data/lib/nokogiri/xml/sax/push_parser.rb +13 -1
data/lib/nokogiri/xml/schema.rb +63 -0
data/lib/nokogiri/xml/syntax_error.rb +25 -1
data/lib/nokogiri/xml/text.rb +4 -1
data/lib/nokogiri/xml/xpath.rb +1 -1
data/lib/nokogiri/xml/xpath/syntax_error.rb +3 -0
data/lib/nokogiri/xml/xpath_context.rb +2 -0
data/lib/nokogiri/xslt.rb +26 -2
data/lib/nokogiri/xslt/stylesheet.rb +19 -0
data/lib/xsd/xmlparser/nokogiri.rb +45 -9
data/tasks/cross_compile.rb +173 -0
data/tasks/test.rb +25 -69
data/test/css/test_nthiness.rb +3 -4
data/test/css/test_parser.rb +75 -20
data/test/css/test_tokenizer.rb +23 -1
data/test/css/test_xpath_visitor.rb +10 -1
data/test/decorators/test_slop.rb +16 -0
data/test/ffi/test_document.rb +35 -0
data/test/files/2ch.html +108 -0
data/test/files/address_book.rlx +12 -0
data/test/files/address_book.xml +10 -0
data/test/files/bar/bar.xsd +4 -0
data/test/files/encoding.html +82 -0
data/test/files/encoding.xhtml +84 -0
data/test/files/foo/foo.xsd +4 -0
data/test/files/po.xml +32 -0
data/test/files/po.xsd +66 -0
data/test/files/shift_jis.html +10 -0
data/test/files/shift_jis.xml +5 -0
data/test/files/snuggles.xml +3 -0
data/test/files/staff.dtd +10 -0
data/test/files/valid_bar.xml +2 -0
data/test/helper.rb +101 -23
data/test/html/sax/test_parser.rb +81 -2
data/test/html/sax/test_parser_context.rb +48 -0
data/test/html/test_builder.rb +39 -8
data/test/html/test_document.rb +186 -23
data/test/html/test_document_encoding.rb +78 -1
data/test/html/test_document_fragment.rb +253 -0
data/test/html/test_element_description.rb +98 -0
data/test/html/test_named_characters.rb +1 -1
data/test/html/test_node.rb +124 -36
data/test/html/test_node_encoding.rb +27 -0
data/test/test_convert_xpath.rb +1 -52
data/test/test_css_cache.rb +2 -13
data/test/test_encoding_handler.rb +46 -0
data/test/test_memory_leak.rb +88 -19
data/test/test_nokogiri.rb +38 -5
data/test/test_reader.rb +188 -6
data/test/test_soap4r_sax.rb +52 -0
data/test/test_xslt_transforms.rb +183 -83
data/test/xml/node/test_save_options.rb +1 -1
data/test/xml/node/test_subclass.rb +44 -0
data/test/xml/sax/test_parser.rb +175 -4
data/test/xml/sax/test_parser_context.rb +113 -0
data/test/xml/sax/test_push_parser.rb +90 -2
data/test/xml/test_attr.rb +35 -1
data/test/xml/test_attribute_decl.rb +82 -0
data/test/xml/test_builder.rb +186 -1
data/test/xml/test_cdata.rb +32 -1
data/test/xml/test_comment.rb +13 -1
data/test/xml/test_document.rb +415 -43
data/test/xml/test_document_encoding.rb +1 -1
data/test/xml/test_document_fragment.rb +173 -5
data/test/xml/test_dtd.rb +61 -6
data/test/xml/test_dtd_encoding.rb +3 -1
data/test/xml/test_element_content.rb +56 -0
data/test/xml/test_element_decl.rb +73 -0
data/test/xml/test_entity_decl.rb +120 -0
data/test/xml/test_entity_reference.rb +5 -1
data/test/xml/test_namespace.rb +68 -0
data/test/xml/test_node.rb +546 -201
data/test/xml/test_node_attributes.rb +34 -0
data/test/xml/test_node_encoding.rb +33 -3
data/test/xml/test_node_reparenting.rb +321 -0
data/test/xml/test_node_set.rb +538 -2
data/test/xml/test_parse_options.rb +52 -0
data/test/xml/test_processing_instruction.rb +6 -1
data/test/xml/test_reader_encoding.rb +1 -1
data/test/xml/test_relax_ng.rb +60 -0
data/test/xml/test_schema.rb +94 -0
data/test/xml/test_syntax_error.rb +12 -0
data/test/xml/test_text.rb +35 -1
data/test/xml/test_unparented_node.rb +5 -5
data/test/xml/test_xpath.rb +142 -11
data/test/xslt/test_custom_functions.rb +94 -0
metadata +328 -92
data/ext/nokogiri/html_sax_parser.c +0 -57
data/ext/nokogiri/html_sax_parser.h +0 -11
data/ext/nokogiri/iconv.dll +0 -0
data/ext/nokogiri/libexslt.dll +0 -0
data/ext/nokogiri/libxml2.dll +0 -0
data/ext/nokogiri/libxslt.dll +0 -0
data/ext/nokogiri/native.so +0 -0
data/ext/nokogiri/xml_xpath.c +0 -53
data/ext/nokogiri/xml_xpath.h +0 -11
data/ext/nokogiri/zlib1.dll +0 -0
data/lib/action-nokogiri.rb +0 -30
data/lib/nokogiri/css/generated_parser.rb +0 -713
data/lib/nokogiri/css/generated_tokenizer.rb +0 -144
data/lib/nokogiri/decorators.rb +0 -2
data/lib/nokogiri/decorators/hpricot.rb +0 -3
data/lib/nokogiri/decorators/hpricot/node.rb +0 -56
data/lib/nokogiri/decorators/hpricot/node_set.rb +0 -54
data/lib/nokogiri/decorators/hpricot/xpath_visitor.rb +0 -28
data/lib/nokogiri/hpricot.rb +0 -51
data/lib/nokogiri/xml/comment.rb +0 -6
data/lib/nokogiri/xml/element.rb +0 -6
data/lib/nokogiri/xml/entity_declaration.rb +0 -9
data/lib/nokogiri/xml/fragment_handler.rb +0 -34
data/test/hpricot/files/basic.xhtml +0 -17
data/test/hpricot/files/boingboing.html +0 -2266
data/test/hpricot/files/cy0.html +0 -3653
data/test/hpricot/files/immob.html +0 -400
data/test/hpricot/files/pace_application.html +0 -1320
data/test/hpricot/files/tenderlove.html +0 -16
data/test/hpricot/files/uswebgen.html +0 -220
data/test/hpricot/files/utf8.html +0 -1054
data/test/hpricot/files/week9.html +0 -1723
data/test/hpricot/files/why.xml +0 -19
data/test/hpricot/load_files.rb +0 -11
data/test/hpricot/test_alter.rb +0 -68
data/test/hpricot/test_builder.rb +0 -20
data/test/hpricot/test_parser.rb +0 -426
data/test/hpricot/test_paths.rb +0 -15
data/test/hpricot/test_preserved.rb +0 -77
data/test/hpricot/test_xml.rb +0 -30
data/test/test_gc.rb +0 -15

data/lib/nokogiri/html/document_fragment.rb ADDED Viewed

@@ -0,0 +1,41 @@
+module Nokogiri
+  module HTML
+    class DocumentFragment < Nokogiri::XML::DocumentFragment
+      attr_accessor :errors
+      ####
+      # Create a Nokogiri::XML::DocumentFragment from +tags+, using +encoding+
+      def self.parse tags, encoding = nil
+        doc = HTML::Document.new
+        encoding ||= tags.respond_to?(:encoding) ? tags.encoding.name : 'UTF-8'
+        doc.encoding = encoding
+        new(doc, tags)
+      end
+      def initialize document, tags = nil, ctx = nil
+        return self unless tags
+        if ctx
+          preexisting_errors = document.errors.dup
+          node_set = ctx.parse("<div>#{tags}</div>")
+          node_set.first.children.each { |child| child.parent = self } unless node_set.empty?
+          self.errors = document.errors - preexisting_errors
+        else
+          # This is a horrible hack, but I don't care
+          if tags.strip =~ /^<body/i
+            path = "/html/body"
+          else
+            path = "/html/body/node()"
+          end
+          temp_doc = HTML::Document.parse "<html><body>#{tags}", nil, document.encoding
+          temp_doc.xpath(path).each { |child| child.parent = self }
+          self.errors = temp_doc.errors
+        end
+        children
+      end
+    end
+  end
+end

data/lib/nokogiri/html/element_description.rb ADDED Viewed

@@ -0,0 +1,23 @@
+module Nokogiri
+  module HTML
+    class ElementDescription
+      ###
+      # Is this element a block element?
+      def block?
+        !inline?
+      end
+      ###
+      # Convert this description to a string
+      def to_s
+        "#{name}: #{description}"
+      end
+      ###
+      # Inspection information
+      def inspect
+        "#<#{self.class.name}: #{name} #{description}>"
+      end
+    end
+  end
+end

data/lib/nokogiri/html/entity_lookup.rb CHANGED Viewed

@@ -3,6 +3,8 @@ module Nokogiri
     class EntityDescription < Struct.new(:value, :name, :description); end
     class EntityLookup
+      ###
+      # Look up entity with +name+
       def [] name
         (val = get(name)) && val.value
       end

data/lib/nokogiri/html/sax/parser.rb CHANGED Viewed

@@ -1,19 +1,50 @@
 module Nokogiri
   module HTML
+    ###
+    # Nokogiri lets you write a SAX parser to process HTML but get HTML
+    # correction features.
+    #
+    # See Nokogiri::HTML::SAX::Parser for a basic example of using a
+    # SAX parser with HTML.
+    #
+    # For more information on SAX parsers, see Nokogiri::XML::SAX
     module SAX
-      class Parser < XML::SAX::Parser
+      ###
+      # This class lets you perform SAX style parsing on HTML with HTML
+      # error correction.
+      #
+      # Here is a basic usage example:
+      #
+      #   class MyDoc < Nokogiri::XML::SAX::Document
+      #     def start_element name, attributes = []
+      #       puts "found a #{name}"
+      #     end
+      #   end
+      #
+      #   parser = Nokogiri::HTML::SAX::Parser.new(MyDoc.new)
+      #   parser.parse(File.read(ARGV[0], 'rb'))
+      #
+      # For more information on SAX parsers, see Nokogiri::XML::SAX
+      class Parser < Nokogiri::XML::SAX::Parser
         ###
         # Parse html stored in +data+ using +encoding+
         def parse_memory data, encoding = 'UTF-8'
-          native_parse_memory(data, encoding)
+          raise ArgumentError unless data
+          return unless data.length > 0
+          ctx = ParserContext.memory(data, encoding)
+          yield ctx if block_given?
+          ctx.parse_with self
         end
         ###
         # Parse a file with +filename+
         def parse_file filename, encoding = 'UTF-8'
+          raise ArgumentError unless filename
           raise Errno::ENOENT unless File.exists?(filename)
           raise Errno::EISDIR if File.directory?(filename)
-          native_parse_file filename, encoding
+          ctx = ParserContext.file(filename, encoding)
+          yield ctx if block_given?
+          ctx.parse_with self
         end
       end
     end

data/lib/nokogiri/html/sax/parser_context.rb ADDED Viewed

@@ -0,0 +1,16 @@
+module Nokogiri
+  module HTML
+    module SAX
+      ###
+      # Context for HTML SAX parsers.  This class is usually not instantiated
+      # by the user.  Instead, you should be looking at
+      # Nokogiri::HTML::SAX::Parser
+      class ParserContext < Nokogiri::XML::SAX::ParserContext
+        def self.new thing, encoding = 'UTF-8'
+          [:read, :close].all? { |x| thing.respond_to?(x) } ?  super :
+            memory(thing, encoding)
+        end
+      end
+    end
+  end
+end

data/lib/nokogiri/nokogiri.rb ADDED Viewed

	@@ -0,0 +1 @@
1	+ require "nokogiri/#{RUBY_VERSION.sub(/\.\d+$/, '')}/nokogiri"

data/lib/nokogiri/version.rb CHANGED Viewed

@@ -1,4 +1,43 @@
 module Nokogiri
   # The version of Nokogiri you are using
-  VERSION = '1.2.3'
+  VERSION = '1.4.5'
+  # More complete version information about libxml
+  VERSION_INFO = {}
+  VERSION_INFO['warnings']              = []
+  VERSION_INFO['nokogiri']              = Nokogiri::VERSION
+  VERSION_INFO['ruby']                  = {}
+  VERSION_INFO['ruby']['version']       = ::RUBY_VERSION
+  VERSION_INFO['ruby']['platform']      = ::RUBY_PLATFORM
+  VERSION_INFO['ruby']['description']   = ::RUBY_DESCRIPTION
+  VERSION_INFO['ruby']['engine']        = defined?(RUBY_ENGINE) ? RUBY_ENGINE : 'mri'
+  VERSION_INFO['ruby']['jruby']         = ::JRUBY_VERSION if RUBY_PLATFORM == "java"
+  if defined?(LIBXML_VERSION)
+    VERSION_INFO['libxml']              = {}
+    VERSION_INFO['libxml']['binding']   = 'extension'
+    VERSION_INFO['libxml']['compiled']  = LIBXML_VERSION
+    VERSION_INFO['libxml']['loaded']    = LIBXML_PARSER_VERSION.scan(/^(.*)(..)(..)$/).first.collect{|j|j.to_i}.join(".")
+    if VERSION_INFO['libxml']['compiled'] != VERSION_INFO['libxml']['loaded']
+      warning = "Nokogiri was built against LibXML version #{VERSION_INFO['libxml']['compiled']}, but has dynamically loaded #{VERSION_INFO['libxml']['loaded']}"
+      VERSION_INFO['warnings'] << warning
+      warn "WARNING: #{warning}"
+    end
+  end
+  def self.uses_libxml? # :nodoc:
+    !Nokogiri::VERSION_INFO['libxml'].nil?
+  end
+  def self.ffi? # :nodoc:
+    uses_libxml? && Nokogiri::VERSION_INFO['libxml']['binding'] == 'ffi'
+  end
+  def self.jruby?
+    false # for 1.5 compatibility
+  end
+  def self.is_2_6_16? # :nodoc:
+    Nokogiri::VERSION_INFO['libxml']['loaded'] <= '2.6.16'
+  end
 end

data/lib/nokogiri/version_warning.rb ADDED Viewed

@@ -0,0 +1,14 @@
+module Nokogiri
+  if self.is_2_6_16?
+    VERSION_INFO['warnings'] << "libxml 2.6.16 is old and buggy."
+    if !defined?(I_KNOW_I_AM_USING_AN_OLD_AND_BUGGY_VERSION_OF_LIBXML2)
+      warn <<-eom
+HI.  You're using libxml2 version 2.6.16 which is over 4 years old and has
+plenty of bugs.  We suggest that for maximum HTML/XML parsing pleasure, you
+upgrade your version of libxml2 and re-install nokogiri.  If you like using
+libxml2 version 2.6.16, but don't like this warning, please define the constant
+I_KNOW_I_AM_USING_AN_OLD_AND_BUGGY_VERSION_OF_LIBXML2 before requring nokogiri.
+      eom
+    end
+  end
+end

data/lib/nokogiri/xml.rb CHANGED Viewed

@@ -1,14 +1,19 @@
+require 'nokogiri/xml/pp'
+require 'nokogiri/xml/parse_options'
 require 'nokogiri/xml/sax'
-require 'nokogiri/xml/fragment_handler'
 require 'nokogiri/xml/node'
+require 'nokogiri/xml/attribute_decl'
+require 'nokogiri/xml/element_decl'
+require 'nokogiri/xml/element_content'
+require 'nokogiri/xml/character_data'
+require 'nokogiri/xml/namespace'
 require 'nokogiri/xml/attr'
 require 'nokogiri/xml/dtd'
-require 'nokogiri/xml/text'
 require 'nokogiri/xml/cdata'
-require 'nokogiri/xml/processing_instruction'
-require 'nokogiri/xml/comment'
+require 'nokogiri/xml/text'
 require 'nokogiri/xml/document'
 require 'nokogiri/xml/document_fragment'
+require 'nokogiri/xml/processing_instruction'
 require 'nokogiri/xml/node_set'
 require 'nokogiri/xml/syntax_error'
 require 'nokogiri/xml/xpath'
@@ -16,72 +21,46 @@ require 'nokogiri/xml/xpath_context'
 require 'nokogiri/xml/builder'
 require 'nokogiri/xml/reader'
 require 'nokogiri/xml/notation'
-require 'nokogiri/xml/element'
-require 'nokogiri/xml/entity_declaration'
+require 'nokogiri/xml/entity_decl'
+require 'nokogiri/xml/schema'
+require 'nokogiri/xml/relax_ng'
 module Nokogiri
   class << self
     ###
-    # Parse an XML file.  +thing+ may be a String, or any object that
-    # responds to _read_ and _close_ such as an IO, or StringIO.
-    # +url+ is resource where this document is located.  +encoding+ is the
-    # encoding that should be used when processing the document. +options+
-    # is a number that sets options in the parser, such as
-    # Nokogiri::XML::PARSE_RECOVER.  See the constants in
-    # Nokogiri::XML.
-    def XML thing, url = nil, encoding = nil, options = 1
-      Nokogiri::XML.parse(thing, url, encoding, options)
+    # Parse XML.  Convenience method for Nokogiri::XML::Document.parse
+    def XML thing, url = nil, encoding = nil, options = XML::ParseOptions::DEFAULT_XML, &block
+      Nokogiri::XML::Document.parse(thing, url, encoding, options, &block)
     end
   end
   module XML
-    # Parser options
-    PARSE_RECOVER     = 1 << 0  # Recover from errors
-    PARSE_NOENT       = 1 << 1  # Substitute entities
-    PARSE_DTDLOAD     = 1 << 2  # Load external subsets
-    PARSE_DTDATTR     = 1 << 3  # Default DTD attributes
-    PARSE_DTDVALID    = 1 << 4  # validate with the DTD
-    PARSE_NOERROR     = 1 << 5  # suppress error reports
-    PARSE_NOWARNING   = 1 << 6  # suppress warning reports
-    PARSE_PEDANTIC    = 1 << 7  # pedantic error reporting
-    PARSE_NOBLANKS    = 1 << 8  # remove blank nodes
-    PARSE_SAX1        = 1 << 9  # use the SAX1 interface internally
-    PARSE_XINCLUDE    = 1 << 10 # Implement XInclude substitition
-    PARSE_NONET       = 1 << 11 # Forbid network access
-    PARSE_NODICT      = 1 << 12 # Do not reuse the context dictionnary
-    PARSE_NSCLEAN     = 1 << 13 # remove redundant namespaces declarations
-    PARSE_NOCDATA     = 1 << 14 # merge CDATA as text nodes
-    PARSE_NOXINCNODE  = 1 << 15 # do not generate XINCLUDE START/END nodes
     class << self
-      def Reader string, url = nil, encoding = nil, options = 0
-        Reader.from_memory(string, url, encoding, options)
-      end
       ###
-      # Parse an XML document.  See Nokogiri.XML.
-      def parse string_or_io, url = nil, encoding = nil, options = 2159
-        if string_or_io.respond_to?(:read)
-          url ||= string_or_io.respond_to?(:path) ? string_or_io.path : nil
-          return Document.read_io(string_or_io, url, encoding, options)
-        end
+      # Parse an XML document using the Nokogiri::XML::Reader API.  See
+      # Nokogiri::XML::Reader for mor information
+      def Reader string_or_io, url = nil, encoding = nil, options = ParseOptions::STRICT
-        # read_memory pukes on empty docs
-        return Document.new if string_or_io.nil? or string_or_io.empty?
+        options = Nokogiri::XML::ParseOptions.new(options) if Fixnum === options
+        # Give the options to the user
+        yield options if block_given?
-        Document.read_memory(string_or_io, url, encoding, options)
+        if string_or_io.respond_to? :read
+          return Reader.from_io(string_or_io, url, encoding, options.to_i)
+        end
+        Reader.from_memory(string_or_io, url, encoding, options.to_i)
       end
       ###
-      # Sets whether or not entities should be substituted.
-      def substitute_entities=(value = true)
-        Document.substitute_entities = value
+      # Parse XML.  Convenience method for Nokogiri::XML::Document.parse
+      def parse thing, url = nil, encoding = nil, options = ParseOptions::DEFAULT_XML, &block
+        Document.parse(thing, url, encoding, options, &block)
       end
-      ###
-      # Sets whether or not external subsets should be loaded
-      def load_external_subsets=(value = true)
-        Document.load_external_subsets = value
+      ####
+      # Parse a fragment from +string+ in to a NodeSet.
+      def fragment string
+        XML::DocumentFragment.parse(string)
       end
     end
   end

data/lib/nokogiri/xml/attr.rb CHANGED Viewed

@@ -4,6 +4,11 @@ module Nokogiri
       alias :value :content
       alias :to_s :content
       alias :content= :value=
+      private
+      def inspect_attributes
+        [:name, :namespace, :value]
+      end
     end
   end
 end

data/lib/nokogiri/xml/attribute_decl.rb ADDED Viewed

@@ -0,0 +1,18 @@
+module Nokogiri
+  module XML
+    ###
+    # Represents an attribute declaration in a DTD
+    class AttributeDecl < Nokogiri::XML::Node
+      undef_method :attribute_nodes
+      undef_method :attributes
+      undef_method :content
+      undef_method :namespace
+      undef_method :namespace_definitions
+      undef_method :line
+      def inspect
+        "#<#{self.class.name}:#{sprintf("0x%x", object_id)} #{to_s.inspect}>"
+      end
+    end
+  end
+end

data/lib/nokogiri/xml/builder.rb CHANGED Viewed

@@ -1,46 +1,350 @@
 module Nokogiri
   module XML
+    ###
+    # Nokogiri builder can be used for building XML and HTML documents.
+    #
+    # == Synopsis:
+    #
+    #   builder = Nokogiri::XML::Builder.new do |xml|
+    #     xml.root {
+    #       xml.products {
+    #         xml.widget {
+    #           xml.id_ "10"
+    #           xml.name "Awesome widget"
+    #         }
+    #       }
+    #     }
+    #   end
+    #   puts builder.to_xml
+    #
+    # Will output:
+    #
+    #   <?xml version="1.0"?>
+    #   <root>
+    #     <products>
+    #       <widget>
+    #         <id>10</id>
+    #         <name>Awesome widget</name>
+    #       </widget>
+    #     </products>
+    #   </root>
+    #
+    #
+    # === Builder scope
+    #
+    # The builder allows two forms.  When the builder is supplied with a block
+    # that has a parameter, the outside scope is maintained.  This means you
+    # can access variables that are outside your builder.  If you don't need
+    # outside scope, you can use the builder without the "xml" prefix like
+    # this:
+    #
+    #   builder = Nokogiri::XML::Builder.new do
+    #     root {
+    #       products {
+    #         widget {
+    #           id_ "10"
+    #           name "Awesome widget"
+    #         }
+    #       }
+    #     }
+    #   end
+    #
+    # == Special Tags
+    #
+    # The builder works by taking advantage of method_missing.  Unfortunately
+    # some methods are defined in ruby that are difficult or dangerous to
+    # remove.  You may want to create tags with the name "type", "class", and
+    # "id" for example.  In that case, you can use an underscore to
+    # disambiguate your tag name from the method call.
+    #
+    # Here is an example of using the underscore to disambiguate tag names from
+    # ruby methods:
+    #
+    #   @objects = [Object.new, Object.new, Object.new]
+    #
+    #   builder = Nokogiri::XML::Builder.new do |xml|
+    #     xml.root {
+    #       xml.objects {
+    #         @objects.each do |o|
+    #           xml.object {
+    #             xml.type_   o.type
+    #             xml.class_  o.class.name
+    #             xml.id_     o.id
+    #           }
+    #         end
+    #       }
+    #     }
+    #   end
+    #   puts builder.to_xml
+    #
+    # The underscore may be used with any tag name, and the last underscore
+    # will just be removed.  This code will output the following XML:
+    #
+    #   <?xml version="1.0"?>
+    #   <root>
+    #     <objects>
+    #       <object>
+    #         <type>Object</type>
+    #         <class>Object</class>
+    #         <id>48390</id>
+    #       </object>
+    #       <object>
+    #         <type>Object</type>
+    #         <class>Object</class>
+    #         <id>48380</id>
+    #       </object>
+    #       <object>
+    #         <type>Object</type>
+    #         <class>Object</class>
+    #         <id>48370</id>
+    #       </object>
+    #     </objects>
+    #   </root>
+    #
+    # == Tag Attributes
+    #
+    # Tag attributes may be supplied as method arguments.  Here is our
+    # previous example, but using attributes rather than tags:
+    #
+    #   @objects = [Object.new, Object.new, Object.new]
+    #
+    #   builder = Nokogiri::XML::Builder.new do |xml|
+    #     xml.root {
+    #       xml.objects {
+    #         @objects.each do |o|
+    #           xml.object(:type => o.type, :class => o.class, :id => o.id)
+    #         end
+    #       }
+    #     }
+    #   end
+    #   puts builder.to_xml
+    #
+    # === Tag Attribute Short Cuts
+    #
+    # A couple attribute short cuts are available when building tags.  The
+    # short cuts are available by special method calls when building a tag.
+    #
+    # This example builds an "object" tag with the class attribute "classy"
+    # and the id of "thing":
+    #
+    #   builder = Nokogiri::XML::Builder.new do |xml|
+    #     xml.root {
+    #       xml.objects {
+    #         xml.object.classy.thing!
+    #       }
+    #     }
+    #   end
+    #   puts builder.to_xml
+    #
+    # Which will output:
+    #
+    #   <?xml version="1.0"?>
+    #   <root>
+    #     <objects>
+    #       <object class="classy" id="thing"/>
+    #     </objects>
+    #   </root>
+    #
+    # All other options are still supported with this syntax, including
+    # blocks and extra tag attributes.
+    #
+    # == Namespaces
+    #
+    # Namespaces are added similarly to attributes.  Nokogiri::XML::Builder
+    # assumes that when an attribute starts with "xmlns", it is meant to be
+    # a namespace:
+    #
+    #   builder = Nokogiri::XML::Builder.new { |xml|
+    #     xml.root('xmlns' => 'default', 'xmlns:foo' => 'bar') do
+    #       xml.tenderlove
+    #     end
+    #   }
+    #   puts builder.to_xml
+    #
+    # Will output XML like this:
+    #
+    #   <?xml version="1.0"?>
+    #   <root xmlns:foo="bar" xmlns="default">
+    #     <tenderlove/>
+    #   </root>
+    #
+    # === Referencing declared namespaces
+    #
+    # Tags that reference non-default namespaces (i.e. a tag "foo:bar") can be
+    # built by using the Nokogiri::XML::Builder#[] method.
+    #
+    # For example:
+    #
+    #   builder = Nokogiri::XML::Builder.new do |xml|
+    #     xml.root('xmlns:foo' => 'bar') {
+    #       xml.objects {
+    #         xml['foo'].object.classy.thing!
+    #       }
+    #     }
+    #   end
+    #   puts builder.to_xml
+    #
+    # Will output this XML:
+    #
+    #   <?xml version="1.0"?>
+    #   <root xmlns:foo="bar">
+    #     <objects>
+    #       <foo:object class="classy" id="thing"/>
+    #     </objects>
+    #   </root>
+    #
+    # Note the "foo:object" tag.
+    #
+    # == Document Types
+    #
+    # To create a document type (DTD), access use the Builder#doc method to get
+    # the current context document.  Then call Node#create_internal_subset to
+    # create the DTD node.
+    #
+    # For example, this Ruby:
+    #
+    #   builder = Nokogiri::XML::Builder.new do |xml|
+    #     xml.doc.create_internal_subset(
+    #       'html',
+    #       "-//W3C//DTD HTML 4.01 Transitional//EN",
+    #       "http://www.w3.org/TR/html4/loose.dtd"
+    #     )
+    #     xml.root do
+    #       xml.foo
+    #     end
+    #   end
+    #
+    #   puts builder.to_xml
+    #
+    # Will output this xml:
+    #
+    #   <?xml version="1.0"?>
+    #   <!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
+    #   <root>
+    #     <foo/>
+    #   </root>
+    #
     class Builder
-      attr_accessor :doc, :parent, :context
-      def initialize &block
-        namespace = self.class.name.split('::')
-        namespace[-1] = 'Document'
-        @doc = eval(namespace.join('::')).new
-        @parent = @doc
-        @context = nil
-        if block_given?
+      # The current Document object being built
+      attr_accessor :doc
+      # The parent of the current node being built
+      attr_accessor :parent
+      # A context object for use when the block has no arguments
+      attr_accessor :context
+      attr_accessor :arity # :nodoc:
+      ###
+      # Create a builder with an existing root object.  This is for use when
+      # you have an existing document that you would like to augment with
+      # builder methods.  The builder context created will start with the
+      # given +root+ node.
+      #
+      # For example:
+      #
+      #   doc = Nokogiri::XML(open('somedoc.xml'))
+      #   Nokogiri::XML::Builder.with(doc.at('some_tag')) do |xml|
+      #     # ... Use normal builder methods here ...
+      #     xml.awesome # add the "awesome" tag below "some_tag"
+      #   end
+      #
+      def self.with root, &block
+        new({}, root, &block)
+      end
+      ###
+      # Create a new Builder object.  +options+ are sent to the top level
+      # Document that is being built.
+      #
+      # Building a document with a particular encoding for example:
+      #
+      #   Nokogiri::XML::Builder.new(:encoding => 'UTF-8') do |xml|
+      #     ...
+      #   end
+      def initialize options = {}, root = nil, &block
+        if root
+          @doc    = root.document
+          @parent = root
+        else
+          namespace     = self.class.name.split('::')
+          namespace[-1] = 'Document'
+          @doc          = eval(namespace.join('::')).new
+          @parent       = @doc
+        end
+        @context  = nil
+        @arity    = nil
+        @ns       = nil
+        options.each do |k,v|
+          @doc.send(:"#{k}=", v)
+        end
+        return unless block_given?
+        @arity = block.arity
+        if @arity <= 0
           @context = eval('self', block.binding)
+          instance_eval(&block)
+        else
+          yield self
         end
-        instance_eval(&block) if block_given?
         @parent = @doc
       end
+      ###
+      # Create a Text Node with content of +string+
       def text string
-        node = Nokogiri::XML::Text.new(string.to_s, @doc)
-        insert(node)
+        insert @doc.create_text_node(string)
       end
+      ###
+      # Create a CDATA Node with content of +string+
       def cdata string
-        node = Nokogiri::XML::CDATA.new(@doc, string.to_s)
-        insert(node)
+        insert(doc.create_cdata(string))
+      end
+      ###
+      # Build a tag that is associated with namespace +ns+.  Raises an
+      # ArgumentError if +ns+ has not been defined higher in the tree.
+      def [] ns
+        @ns = @parent.namespace_definitions.find { |x| x.prefix == ns.to_s }
+        return self if @ns
+        @parent.ancestors.each do |a|
+          next if a == doc
+          @ns = a.namespace_definitions.find { |x| x.prefix == ns.to_s }
+          return self if @ns
+        end
+        raise ArgumentError, "Namespace #{ns} has not been defined"
+      end
+      ###
+      # Convert this Builder object to XML
+      def to_xml(*args)
+        @doc.to_xml(*args)
       end
-      def to_xml
-        @doc.to_xml
+      ###
+      # Append the given raw XML +string+ to the document
+      def << string
+        @doc.fragment(string).children.each { |x| insert(x) }
       end
-      def method_missing method, *args, &block
+      def method_missing method, *args, &block # :nodoc:
         if @context && @context.respond_to?(method)
           @context.send(method, *args, &block)
         else
-          node = Nokogiri::XML::Node.new(method.to_s, @doc) { |n|
-            args.each do |arg|
-              case arg
-              when Hash
-                arg.each { |k,v| n[k.to_s] = v.to_s }
-              else
-                n.content = arg
-              end
+          node = @doc.create_element(method.to_s.sub(/[_!]$/, ''),*args) { |n|
+            # Set up the namespace
+            if @ns
+              n.namespace = @ns
+              @ns = nil
             end
           }
           insert(node, &block)
@@ -48,22 +352,38 @@ module Nokogiri
       end
       private
+      ###
+      # Insert +node+ as a child of the current Node
       def insert(node, &block)
         node.parent = @parent
         if block_given?
-          @parent = node
-          instance_eval(&block)
-          @parent = node.parent
+          old_parent = @parent
+          @parent    = node
+          @arity ||= block.arity
+          if @arity <= 0
+            instance_eval(&block)
+          else
+            block.call(self)
+          end
+          @parent = old_parent
         end
         NodeBuilder.new(node, self)
       end
       class NodeBuilder # :nodoc:
-        def initialize(node, doc_builder)
+        def initialize node, doc_builder
           @node = node
           @doc_builder = doc_builder
         end
+        def []= k, v
+          @node[k] = v
+        end
+        def [] k
+          @node[k]
+        end
         def method_missing(method, *args, &block)
           opts = args.last.is_a?(Hash) ? args.pop : {}
           case method.to_s
@@ -73,7 +393,7 @@ module Nokogiri
           when /^(.*)=/
             @node[$1] = args.first
           else
-            @node['class'] =
+            @node['class'] =
               ((@node['class'] || '').split(/\s/) + [method.to_s]).join(' ')
             @node.content = args.first if args.first
           end