nokogiri 1.4.3.1-java → 1.5.0-java

data/lib/nokogiri/xml/node.rb

@@ -95,13 +95,17 @@ module Nokogiri
       # optional hash of namespaces may be appended.
       # See Node#xpath and Node#css.
       def search *paths
+        # TODO use         paths, handler, ns, binds = extract_params(paths)
         ns = paths.last.is_a?(Hash) ? paths.pop :
           (document.root ? document.root.namespaces : {})
+
+        prefix = "#{implied_xpath_context}/"
+
         xpath(*(paths.map { |path|
           path = path.to_s
-          path =~ /^(\.\/|\/)/ ? path : CSS.xpath_for(
+          path =~ /^(\.\/|\/|\.\.)/ ? path : CSS.xpath_for(
             path,
-            :prefix => ".//",
+            :prefix => prefix,
             :ns     => ns
           )
         }.flatten.uniq) + [ns])
@@ -109,16 +113,28 @@ module Nokogiri
       alias :/ :search
 
       ###
+      # call-seq: xpath *paths, [namespace-bindings, variable-bindings, custom-handler-class]
+      #
       # Search this node for XPath +paths+. +paths+ must be one or more XPath
-      # queries.  A hash of namespaces may be appended.  For example:
+      # queries.
       #
       #   node.xpath('.//title')
-      #   node.xpath('.//foo:name', { 'foo' => 'http://example.org/' })
+      #
+      # A hash of namespace bindings may be appended. For example:
+      #
+      #   node.xpath('.//foo:name', {'foo' => 'http://example.org/'})
       #   node.xpath('.//xmlns:name', node.root.namespaces)
       #
-      # Custom XPath functions may also be defined.  To define custom functions
-      # create a class and implement the # function you want to define.
-      # For example:
+      # A hash of variable bindings may also be appended to the namespace bindings. For example:
+      #
+      #   node.xpath('.//address[@domestic=$value]', nil, {:value => 'Yes'})
+      #
+      # Custom XPath functions may also be defined.  To define custom
+      # functions create a class and implement the function you want
+      # to define.  The first argument to the method will be the
+      # current matching NodeSet.  Any other arguments are ones that
+      # you pass in.  Note that this class may appear anywhere in the
+      # argument list.  For example:
       #
       #   node.xpath('.//title[regex(., "\w+")]', Class.new {
       #     def regex node_set, regex
@@ -127,19 +143,19 @@ module Nokogiri
       #   }.new)
       #
       def xpath *paths
-        # 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 :
-          (document.root ? document.root.namespaces : {})
-
         return NodeSet.new(document) unless document
 
+        paths, handler, ns, binds = extract_params(paths)
+
         sets = paths.map { |path|
           ctx = XPathContext.new(self)
           ctx.register_namespaces(ns)
+          path = path.gsub(/\/xmlns:/,'/:') unless Nokogiri.uses_libxml?
+
+          binds.each do |key,value|
+            ctx.register_variable key.to_s, value
+          end if binds
+
           ctx.evaluate(path, handler)
         }
         return sets.first if sets.length == 1
@@ -154,18 +170,24 @@ module Nokogiri
       end
 
       ###
+      # call-seq: css *rules, [namespace-bindings, custom-pseudo-class]
+      #
       # Search this node for CSS +rules+. +rules+ must be one or more CSS
-      # selectors.  For example:
+      # selectors. For example:
       #
       #   node.css('title')
       #   node.css('body h1.bold')
       #   node.css('div + p.green', 'div#one')
       #
-      # Custom CSS pseudo classes may also be defined.  To define custom pseudo
-      # 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:
+      # A hash of namespace bindings may be appended. For example:
+      #
+      #   node.css('bike|tire', {'bike' => 'http://schwinn.com/'})
+      #
+      # Custom CSS pseudo classes may also be defined.  To define
+      # custom pseudo 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 {
       #     def regex node_set, regex
@@ -173,18 +195,21 @@ module Nokogiri
       #     end
       #   }.new)
       #
+      # Note that the CSS query string is case-sensitive with regards
+      # to your document type. That is, if you're looking for "H1" in
+      # an HTML document, you'll never find anything, since HTML tags
+      # will match only lowercase CSS queries. However, "H1" might be
+      # found in an XML document, where tags names are case-sensitive
+      # (e.g., "H1" is distinct from "h1").
+      #
       def css *rules
-        # Pop off our custom function handler if it exists
-        handler = ![
-          Hash, String, Symbol
-        ].include?(rules.last.class) ? rules.pop : nil
+        rules, handler, ns, binds = extract_params(rules)
 
-        ns = rules.last.is_a?(Hash) ? rules.pop :
-          (document.root ? document.root.namespaces : {})
+        prefix = "#{implied_xpath_context}/"
 
         rules = rules.map { |rule|
-          CSS.xpath_for(rule, :prefix => ".//", :ns => ns)
-        }.flatten.uniq + [ns, handler].compact
+          CSS.xpath_for(rule, :prefix => prefix, :ns => ns)
+        }.flatten.uniq + [ns, handler, binds].compact
 
         xpath(*rules)
       end
@@ -234,7 +259,9 @@ module Nokogiri
       # Add +node_or_tags+ as a child of this Node.
       # +node_or_tags+ can be a Nokogiri::XML::Node, a ::DocumentFragment, a ::NodeSet, or a string containing markup.
       #
-      # Returns the new child node.
+      # Returns the reparented node (if +node_or_tags+ is a Node), or NodeSet (if +node_or_tags+ is a DocumentFragment, NodeSet, or string).
+      #
+      # Also see related method +<<+.
       def add_child node_or_tags
         node_or_tags = coerce(node_or_tags)
         if node_or_tags.is_a?(XML::NodeSet)
@@ -242,42 +269,66 @@ module Nokogiri
         else
           add_child_node node_or_tags
         end
+        node_or_tags
       end
 
+      ###
+      # Add +node_or_tags+ as a child of this Node.
+      # +node_or_tags+ can be a Nokogiri::XML::Node, a ::DocumentFragment, a ::NodeSet, or a string containing markup.
+      #
+      # Returns self, to support chaining of calls (e.g., root << child1 << child2)
+      #
+      # Also see related method +add_child+.
+      def << node_or_tags
+        add_child node_or_tags
+        self
+      end
       ###
       # Insert +node_or_tags+ before this Node (as a sibling).
       # +node_or_tags+ can be a Nokogiri::XML::Node, a ::DocumentFragment, a ::NodeSet, or a string containing markup.
       #
-      # Returns the new sibling node.
+      # Returns the reparented node (if +node_or_tags+ is a Node), or NodeSet (if +node_or_tags+ is a DocumentFragment, NodeSet, or string).
       #
       # Also see related method +before+.
       def add_previous_sibling node_or_tags
         node_or_tags = coerce(node_or_tags)
         if node_or_tags.is_a?(XML::NodeSet)
-          node_or_tags.each { |n| add_previous_sibling_node n }
+          if text?
+            pivot = Nokogiri::XML::Node.new 'dummy', document
+            add_previous_sibling_node pivot
+          else
+            pivot = self
+          end
+          node_or_tags.each { |n| pivot.send :add_previous_sibling_node, n }
+          pivot.unlink if text?
         else
           add_previous_sibling_node node_or_tags
         end
+        node_or_tags
       end
 
       ###
       # Insert +node_or_tags+ after this Node (as a sibling).
       # +node_or_tags+ can be a Nokogiri::XML::Node, a ::DocumentFragment, a ::NodeSet, or a string containing markup.
       #
-      # Returns the new sibling node.
+      # Returns the reparented node (if +node_or_tags+ is a Node), or NodeSet (if +node_or_tags+ is a DocumentFragment, NodeSet, or string).
       #
       # Also see related method +after+.
       def add_next_sibling node_or_tags
         node_or_tags = coerce(node_or_tags)
         if node_or_tags.is_a?(XML::NodeSet)
-          if '1.8.6' == RUBY_VERSION
-            node_or_tags.reverse.each { |n| add_next_sibling_node n }
+          if text?
+            pivot = Nokogiri::XML::Node.new 'dummy', document
+            add_next_sibling_node pivot
           else
-            node_or_tags.reverse_each { |n| add_next_sibling_node n }
+            pivot = self
           end
+          node_or_tags.reverse_each { |n| pivot.send :add_next_sibling_node, n }
+          pivot.unlink if text?
         else
           add_next_sibling_node node_or_tags
         end
+        node_or_tags
       end
 
       ####
@@ -305,36 +356,58 @@ module Nokogiri
       end
 
       ####
-      # Set the inner_html for this Node to +node_or_tags+
+      # Set the inner html for this Node to +node_or_tags+
       # +node_or_tags+ can be a Nokogiri::XML::Node, a Nokogiri::XML::DocumentFragment, or a string containing markup.
       #
       # Returns self.
+      #
+      # Also see related method +children=+
       def inner_html= node_or_tags
+        self.children = node_or_tags
+        self
+      end
+
+      ####
+      # Set the inner html for this Node +node_or_tags+
+      # +node_or_tags+ can be a Nokogiri::XML::Node, a Nokogiri::XML::DocumentFragment, or a string containing markup.
+      #
+      # Returns the reparented node (if +node_or_tags+ is a Node), or NodeSet (if +node_or_tags+ is a DocumentFragment, NodeSet, or string).
+      #
+      # Also see related method +inner_html=+
+      def children= node_or_tags
         node_or_tags = coerce(node_or_tags)
         children.unlink
         if node_or_tags.is_a?(XML::NodeSet)
           node_or_tags.each { |n| add_child_node n }
         else
-          add_child node_or_tags
+          add_child_node node_or_tags
         end
-        self
+        node_or_tags
       end
 
       ####
       # Replace this Node with +node_or_tags+.
       # +node_or_tags+ can be a Nokogiri::XML::Node, a ::DocumentFragment, a ::NodeSet, or a string containing markup.
       #
-      # Returns the new child node.
+      # Returns the reparented node (if +node_or_tags+ is a Node), or NodeSet (if +node_or_tags+ is a DocumentFragment, NodeSet, or string).
       #
       # Also see related method +swap+.
       def replace node_or_tags
         node_or_tags = coerce(node_or_tags)
         if node_or_tags.is_a?(XML::NodeSet)
-          node_or_tags.each { |n| add_previous_sibling n }
-          unlink
+          if text?
+            replacee = Nokogiri::XML::Node.new 'dummy', document
+            add_previous_sibling_node replacee
+            unlink
+          else
+            replacee = self
+          end
+          node_or_tags.each { |n| replacee.add_previous_sibling n }
+          replacee.unlink
         else
           replace_node node_or_tags
         end
+        node_or_tags
       end
 
       ####
@@ -365,7 +438,6 @@ module Nokogiri
       alias :text           :content
       alias :inner_text     :content
       alias :has_attribute? :key?
-      alias :<<             :add_child
       alias :name           :node_name
       alias :name=          :node_name=
       alias :type           :node_type
@@ -430,7 +502,8 @@ module Nokogiri
       # Parse +string_or_io+ as a document fragment within the context of
       # *this* node.  Returns a XML::NodeSet containing the nodes parsed from
       # +string_or_io+.
-      def parse string_or_io, options = ParseOptions::DEFAULT_XML
+      def parse string_or_io, options = nil
+        options ||= (document.html? ? ParseOptions::DEFAULT_HTML : ParseOptions::DEFAULT_XML)
         if Fixnum === options
           options = Nokogiri::XML::ParseOptions.new(options)
         end
@@ -442,7 +515,16 @@ module Nokogiri
           string_or_io
 
         return Nokogiri::XML::NodeSet.new(document) if contents.empty?
-        in_context(contents, options.to_i)
+
+        ##
+        # This is a horrible hack, but I don't care. See #313 for background.
+        error_count = document.errors.length
+        node_set = in_context(contents, options.to_i)
+        if node_set.empty? and document.errors.length > error_count and options.recover?
+          fragment = Nokogiri::HTML::DocumentFragment.parse contents
+          node_set = fragment.children
+        end
+        node_set
       end
 
       ####
@@ -459,7 +541,19 @@ module Nokogiri
       end
 
       ###
-      # Get a hash containing the Namespace definitions for this Node
+      # Returns a Hash of {prefix => value} for all namespaces on this
+      # node and its ancestors.
+      #
+      # This method returns the same namespaces as #namespace_scopes.
+      #
+      # Returns namespaces in scope for self -- those defined on self
+      # element directly or any ancestor node -- as a Hash of
+      # attribute-name/value pairs. Note that the keys in this hash
+      # XML attributes that would be used to define this namespace,
+      # such as "xmlns:prefix", not just the prefix. Default namespace
+      # set on self will be included with key "xmlns". However,
+      # default namespaces set on ancestor will NOT be, even if self
+      # has no explicit default namespace.
       def namespaces
         Hash[*namespace_scopes.map { |nd|
           key = ['xmlns', nd.prefix].compact.join(':')
@@ -567,14 +661,22 @@ module Nokogiri
       end
 
       ###
-      # Set the default namespace for this node to +url+
+      # Adds a default namespace supplied as a string +url+ href, to self.
+      # The consequence is as an xmlns attribute with supplied argument were
+      # present in parsed XML.  A default namespace set with this method will
+      # now show up in #attributes, but when this node is serialized to XML an
+      # "xmlns" attribute will appear. See also #namespace and #namespace=
       def default_namespace= url
         add_namespace_definition(nil, url)
       end
       alias :add_namespace :add_namespace_definition
 
       ###
-      # Set the namespace for this node to +ns+
+      # Set the default namespace on this node (as would be defined with an
+      # "xmlns=" attribute in XML source), as a Namespace object +ns+. Note that
+      # a Namespace added this way will NOT be serialized as an xmlns attribute
+      # for this node. You probably want #default_namespace= instead, or perhaps
+      # #add_namespace_definition with a nil prefix argument.
       def namespace= ns
         return set_namespace(ns) unless ns
 
@@ -626,10 +728,11 @@ module Nokogiri
       def serialize *args, &block
         options = args.first.is_a?(Hash) ? args.shift : {
           :encoding   => args[0],
-          :save_with  => args[1] || SaveOptions::FORMAT
+          :save_with  => args[1]
         }
 
         encoding = options[:encoding] || document.encoding
+        options[:encoding] = encoding
 
         outstring = ""
         if encoding && outstring.respond_to?(:force_encoding)
@@ -649,13 +752,10 @@ module Nokogiri
       # use Node#to_xhtml instead.
       def to_html options = {}
         # FIXME: this is a hack around broken libxml versions
-        return dump_html if %w[2 6] === LIBXML_VERSION.split('.')[0..1]
-
-        options[:save_with] ||= SaveOptions::FORMAT |
-                                SaveOptions::NO_DECLARATION |
-                                SaveOptions::NO_EMPTY_TAGS |
-                                SaveOptions::AS_HTML
+        return dump_html if Nokogiri.uses_libxml? && %w[2 6] === LIBXML_VERSION.split('.')[0..1]
 
+        options[:save_with] |= SaveOptions::DEFAULT_HTML if options[:save_with]
+        options[:save_with] = SaveOptions::DEFAULT_HTML unless options[:save_with]
         serialize(options)
       end
 
@@ -666,8 +766,8 @@ module Nokogiri
       #
       # See Node#write_to for a list of +options+
       def to_xml options = {}
-        options[:save_with] ||= SaveOptions::FORMAT | SaveOptions::AS_XML
-
+        options[:save_with] |= SaveOptions::DEFAULT_XML if options[:save_with]
+        options[:save_with] = SaveOptions::DEFAULT_XML unless options[:save_with]
         serialize(options)
       end
 
@@ -679,13 +779,10 @@ module Nokogiri
       # See Node#write_to for a list of +options+
       def to_xhtml options = {}
         # FIXME: this is a hack around broken libxml versions
-        return dump_html if %w[2 6] === LIBXML_VERSION.split('.')[0..1]
-
-        options[:save_with] ||= SaveOptions::FORMAT |
-                                SaveOptions::NO_DECLARATION |
-                                SaveOptions::NO_EMPTY_TAGS |
-                                SaveOptions::AS_XHTML
+        return dump_html if Nokogiri.uses_libxml? && %w[2 6] === LIBXML_VERSION.split('.')[0..1]
 
+        options[:save_with] |= SaveOptions::DEFAULT_XHTML if options[:save_with]
+        options[:save_with] = SaveOptions::DEFAULT_XHTML unless options[:save_with]
         serialize(options)
       end
 
@@ -709,12 +806,16 @@ module Nokogiri
       def write_to io, *options
         options       = options.first.is_a?(Hash) ? options.shift : {}
         encoding      = options[:encoding] || options[0]
-        save_options  = options[:save_with] || options[1] || SaveOptions::FORMAT
+        if Nokogiri.jruby?
+          save_options  = options[:save_with] || options[1]
+          indent_times  = options[:indent] || 0
+        else
+          save_options  = options[:save_with] || options[1] || SaveOptions::FORMAT
+          indent_times  = options[:indent] || 2
+        end
         indent_text   = options[:indent_text] || ' '
-        indent_times  = options[:indent] || 2
-
 
-        config = SaveOptions.new(save_options)
+        config = SaveOptions.new(save_options.to_i)
         yield config if block_given?
 
         native_write_to(io, encoding, indent_text * indent_times, config.options)
@@ -726,12 +827,9 @@ module Nokogiri
       # See Node#write_to for a list of +options+
       def write_html_to io, options = {}
         # FIXME: this is a hack around broken libxml versions
-        return (io << dump_html) if %w[2 6] === LIBXML_VERSION.split('.')[0..1]
+        return (io << dump_html) if Nokogiri.uses_libxml? && %w[2 6] === LIBXML_VERSION.split('.')[0..1]
 
-        options[:save_with] ||= SaveOptions::FORMAT |
-          SaveOptions::NO_DECLARATION |
-          SaveOptions::NO_EMPTY_TAGS |
-          SaveOptions::AS_HTML
+        options[:save_with] ||= SaveOptions::DEFAULT_HTML
         write_to io, options
       end
 
@@ -741,12 +839,9 @@ module Nokogiri
       # See Node#write_to for a list of +options+
       def write_xhtml_to io, options = {}
         # FIXME: this is a hack around broken libxml versions
-        return (io << dump_html) if %w[2 6] === LIBXML_VERSION.split('.')[0..1]
+        return (io << dump_html) if Nokogiri.uses_libxml? && %w[2 6] === LIBXML_VERSION.split('.')[0..1]
 
-        options[:save_with] ||= SaveOptions::FORMAT |
-          SaveOptions::NO_DECLARATION |
-          SaveOptions::NO_EMPTY_TAGS |
-          SaveOptions::AS_XHTML
+        options[:save_with] ||= SaveOptions::DEFAULT_XHTML
         write_to io, options
       end
 
@@ -757,7 +852,7 @@ module Nokogiri
       #
       # See Node#write_to for a list of options
       def write_xml_to io, options = {}
-        options[:save_with] ||= SaveOptions::FORMAT | SaveOptions::AS_XML
+        options[:save_with] ||= SaveOptions::DEFAULT_XML
         write_to io, options
       end
 
@@ -772,6 +867,27 @@ module Nokogiri
 
       private
 
+      def extract_params params # :nodoc:
+        # Pop off our custom function handler if it exists
+        handler = params.find { |param|
+          ![Hash, String, Symbol].include?(param.class)
+        }
+
+        params -= [handler] if handler
+
+        hashes = []
+        while Hash === params.last || params.last.nil?
+          hashes << params.pop
+          break if params.empty?
+        end
+
+        ns, binds = hashes.reverse
+
+        ns ||= document.root ? document.root.namespaces : {}
+
+        [params, handler, ns, binds]
+      end
+
       def coerce data # :nodoc:
         return data                    if data.is_a?(XML::NodeSet)
         return data.children           if data.is_a?(XML::DocumentFragment)
@@ -787,6 +903,10 @@ Requires a Node, NodeSet or String argument, and cannot accept a #{data.class}.
         data
       end
 
+      def implied_xpath_context
+        "./"
+      end
+
       def inspect_attributes
         [:name, :namespace, :attribute_nodes, :children]
       end

data/lib/nokogiri/xml/node_set.rb

@@ -42,7 +42,7 @@ module Nokogiri
       end
 
       ###
-      # Returns the index of the first node in self that is == to +node+. Returns nil if no match is found. 
+      # Returns the index of the first node in self that is == to +node+. Returns nil if no match is found.
       def index(node)
         each_with_index { |member, j| return j if member == node }
         nil
@@ -104,7 +104,7 @@ module Nokogiri
 
         each do |node|
           doc = node.document
-          search_ns = ns || doc.root ? doc.root.namespaces : {}
+          search_ns = ns || (doc.root ? doc.root.namespaces : {})
 
           xpaths = paths.map { |rule|
             [
@@ -257,7 +257,7 @@ module Nokogiri
       # Wrap this NodeSet with +html+ or the results of the builder in +blk+
       def wrap(html, &blk)
         each do |j|
-          new_parent = document.root.parse(html).first
+          new_parent = document.parse(html).first
           j.add_next_sibling(new_parent)
           new_parent.add_child(j)
         end
@@ -273,6 +273,13 @@ module Nokogiri
       ###
       # Convert this NodeSet to HTML
       def to_html *args
+        if Nokogiri.jruby?
+          options = args.first.is_a?(Hash) ? args.shift : {}
+          if !options[:save_with]
+            options[:save_with] = Node::SaveOptions::NO_DECLARATION | Node::SaveOptions::NO_EMPTY_TAGS | Node::SaveOptions::AS_HTML
+          end
+          args.insert(0, options)
+        end
         map { |x| x.to_html(*args) }.join
       end
 

data/lib/nokogiri/xml/parse_options.rb

@@ -37,6 +37,14 @@ module Nokogiri
       NOCDATA     = 1 << 14
       # do not generate XINCLUDE START/END nodes
       NOXINCNODE  = 1 << 15
+      # compact small text nodes; no modification of the tree allowed afterwards (will possibly crash if you try to modify the tree)
+      COMPACT     = 1 << 16
+      # parse using XML-1.0 before update 5
+      OLD10       = 1 << 17
+      # do not fixup XINCLUDE xml:base uris
+      NOBASEFIX   = 1 << 18
+      # relax any hardcoded limit from the parser
+      HUGE        = 1 << 19
 
       # the default options used for parsing XML documents
       DEFAULT_XML  = RECOVER

data/lib/nokogiri/xml/reader.rb

@@ -30,6 +30,42 @@ module Nokogiri
     class Reader
       include Enumerable
 
+      TYPE_NONE = 0
+      # Element node type
+      TYPE_ELEMENT = 1
+      # Attribute node type
+      TYPE_ATTRIBUTE = 2
+      # Text node type
+      TYPE_TEXT = 3
+      # CDATA node type
+      TYPE_CDATA = 4
+      # Entity Reference node type
+      TYPE_ENTITY_REFERENCE = 5
+      # Entity node type
+      TYPE_ENTITY = 6
+      # PI node type
+      TYPE_PROCESSING_INSTRUCTION = 7
+      # Comment node type
+      TYPE_COMMENT = 8
+      # Document node type
+      TYPE_DOCUMENT = 9
+      # Document Type node type
+      TYPE_DOCUMENT_TYPE = 10
+      # Document Fragment node type
+      TYPE_DOCUMENT_FRAGMENT = 11
+      # Notation node type
+      TYPE_NOTATION = 12
+      # Whitespace node type
+      TYPE_WHITESPACE = 13
+      # Significant Whitespace node type
+      TYPE_SIGNIFICANT_WHITESPACE = 14
+      # Element end node type
+      TYPE_END_ELEMENT = 15
+      # Entity end node type
+      TYPE_END_ENTITY = 16
+      # XML Declaration node type
+      TYPE_XML_DECLARATION = 17
+
       # A list of errors encountered while parsing
       attr_accessor :errors
 
@@ -51,9 +87,9 @@ module Nokogiri
       ###
       # Get a list of attributes for the current node.
       def attributes
-        Hash[*(attribute_nodes.map { |node|
+        Hash[attribute_nodes.map { |node|
           [node.name, node.to_s]
-        }.flatten)].merge(namespaces || {})
+        }].merge(namespaces || {})
       end
 
       ###
@@ -65,10 +101,10 @@ module Nokogiri
       end
 
       ###
-      # Move the cursor through the document yielding each node to the block
-      def each(&block)
-        while node = self.read
-          block.call(node)
+      # Move the cursor through the document yielding the cursor to the block
+      def each
+        while cursor = self.read
+          yield cursor
         end
       end
     end

data/lib/nokogiri/xml/sax/document.rb

@@ -85,7 +85,9 @@ module Nokogiri
 
         ###
         # Called at the beginning of an element
-        # +name+ is the name of the tag with +attrs+ as attributes
+        # * +name+ is the name of the tag
+        # * +attrs+ are an assoc list of namespaces and attributes, e.g.:
+        #     [ ["xmlns:foo", "http://sample.net"], ["size", "large"] ]
         def start_element name, attrs = []
         end
 
@@ -110,7 +112,7 @@ module Nokogiri
             [['xmlns', ns_prefix].compact.join(':'), ns_uri]
           } + attrs.map { |attr|
             [[attr.prefix, attr.localname].compact.join(':'), attr.value]
-          }.flatten
+          }
           start_element name, attributes
         end
 

data/lib/nokogiri/xml/schema.rb

@@ -43,7 +43,13 @@ module Nokogiri
       # Nokogiri::XML::SyntaxError objects found while validating the
       # +thing+ is returned.
       def validate thing
-        thing.is_a?(Nokogiri::XML::Document) ? validate_document(thing) : validate_file(thing)
+        if thing.is_a?(Nokogiri::XML::Document) 
+          validate_document(thing) 
+        elsif File.file?(thing)
+          validate_file(thing)
+        else
+          raise ArgumentError, "Must provide Nokogiri::Xml::Document or the name of an existing file"
+        end
       end
 
       ###