nokogiri 1.0.7 → 1.1.0

data/lib/nokogiri/css/parser.rb

@@ -1,35 +1,37 @@
+require 'thread'
+
 module Nokogiri
   module CSS
     class Parser < GeneratedTokenizer
+      @cache_on = true
+      @cache    = {}
+      @mutex    = Mutex.new
+
       class << self
+        attr_accessor :cache_on
+        alias :cache_on? :cache_on
+        alias :set_cache :cache_on=
+
         def parse string
           new.parse(string)
         end
+
         def xpath_for string, options={}
           new.xpath_for(string, options)
         end
 
-        def set_cache setting
-          @cache_on = setting ? true : false
-        end
-
-        def cache_on?
-          @cache ||= {}
-          instance_variable_defined?('@cache_on') ? @cache_on : true
-        end
-
-        def check_cache string
-          return unless cache_on?
-          @cache[string]
+        def [] string
+          return unless @cache_on
+          @mutex.synchronize { @cache[string] }
         end
 
-        def add_cache string, value
-          return value unless cache_on?
-          @cache[string] = value
+        def []= string, value
+          return value unless @cache_on
+          @mutex.synchronize { @cache[string] = value }
         end
 
         def clear_cache
-          @cache = {}
+          @mutex.synchronize { @cache = {} }
         end
 
         def without_cache &block
@@ -42,13 +44,15 @@ module Nokogiri
       alias :parse :scan_str
 
       def xpath_for string, options={}
-        v = self.class.check_cache(string)
-        return v unless v.nil?
+        v = self.class[string]
+        return v if v
 
         prefix = options[:prefix] || nil
         visitor = options[:visitor] || nil
         args = [prefix, visitor]
-        self.class.add_cache(string, parse(string).map {|ast| ast.to_xpath(prefix, visitor)})
+        self.class[string] = parse(string).map { |ast|
+          ast.to_xpath(prefix, visitor)
+        }
       end
 
       def on_error error_token_id, error_value, value_stack

data/lib/nokogiri/css/parser.y

@@ -104,7 +104,9 @@ rule
       }
     ;
   expr
-    : NUMBER
+    : NUMBER COMMA s_0toN expr { result = [val.first, val.last] }
+    | STRING COMMA s_0toN expr { result = [val.first, val.last] }
+    | NUMBER
     | STRING
     ;
   an_plus_b

data/lib/nokogiri/css/selector_handler.rb

@@ -0,0 +1,6 @@
+module Nokogiri
+  module CSS
+    class SelectorHandler < XML::XPathHandler
+    end
+  end
+end

data/lib/nokogiri/css/xpath_visitor.rb

@@ -29,8 +29,11 @@ module Nokogiri
           "position() > #{node.value[1]}"
         when /^only-child\(/
           "last() = 1"
+        when /^comment\(/
+          "comment()"
         else
-          node.value.first + ')'
+          args = ['.'] + node.value[1..-1]
+          "#{node.value.first}#{args.join(', ')})"
         end
       end
 
@@ -106,7 +109,7 @@ module Nokogiri
           when "parent" then "node()"
           when "root" then "not(parent::*)"
           else
-            '1 = 1'
+            node.value.first + "(.)"
           end
         end
       end

data/lib/nokogiri/html.rb

@@ -3,6 +3,14 @@ require 'nokogiri/html/sax/parser'
 
 module Nokogiri
   class << self
+    ###
+    # Parse HTML.  +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 HTML thing, url = nil, encoding = nil, options = 2145
       Nokogiri::HTML.parse(thing, url, encoding, options)
     end
@@ -17,6 +25,8 @@ module Nokogiri
     PARSE_NONET     = 1 << 11 # No network access
 
     class << self
+      ###
+      # Parse HTML.  See Nokogiri.HTML.
       def parse string_or_io, url = nil, encoding = nil, options = 2145
         if string_or_io.respond_to?(:read)
           url ||= string_or_io.respond_to?(:path) ? string_or_io.path : nil

data/lib/nokogiri/version.rb

@@ -1,3 +1,3 @@
 module Nokogiri
-  VERSION = '1.0.7'
+  VERSION = '1.1.0'
 end

data/lib/nokogiri/xml.rb

@@ -1,4 +1,5 @@
 require 'nokogiri/xml/sax'
+require 'nokogiri/xml/xpath_handler'
 require 'nokogiri/xml/before_handler'
 require 'nokogiri/xml/after_handler'
 require 'nokogiri/xml/node'

data/lib/nokogiri/xml/node.rb

@@ -1,17 +1,26 @@
 module Nokogiri
   module XML
     class Node
+      ELEMENT_NODE =       1
+      ATTRIBUTE_NODE =     2
+      TEXT_NODE =          3
       CDATA_SECTION_NODE = 4
-      COMMENT_NODE = 8
-      DOCUMENT_NODE = 9
+      ENTITY_REF_NODE =    5
+      ENTITY_NODE =        6
+      PI_NODE =            7
+      COMMENT_NODE =       8
+      DOCUMENT_NODE =      9
+      DOCUMENT_TYPE_NODE = 10
+      DOCUMENT_FRAG_NODE = 11
+      NOTATION_NODE =      12
       HTML_DOCUMENT_NODE = 13
-      DTD_NODE = 14
-      ELEMENT_DECL = 15
-      ATTRIBUTE_DECL = 16
-      ENTITY_DECL = 17
-      NAMESPACE_DECL = 18
-      XINCLUDE_START = 19
-      XINCLUDE_END = 20
+      DTD_NODE =           14
+      ELEMENT_DECL =       15
+      ATTRIBUTE_DECL =     16
+      ENTITY_DECL =        17
+      NAMESPACE_DECL =     18
+      XINCLUDE_START =     19
+      XINCLUDE_END =       20
       DOCB_DOCUMENT_NODE = 21
 
       attr_accessor :document
@@ -50,7 +59,26 @@ module Nokogiri
       end
       alias :/ :search
 
+      ###
+      # Search this node for XPath +paths+. +paths+ must be one or more XPath
+      # queries.  A hash of namespaces may be appended.  For example:
+      #
+      #   node.xpath('.//title')
+      #   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 which subclasses XPathHandler and implement the
+      # function you want to define.  For example:
+      #
+      #   node.xpath('.//title[regex(., "\w+")]', Class.new(XPathHandler) {
+      #     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
         ns = paths.last.is_a?(Hash) ? paths.pop : {}
 
         return NodeSet.new(document) unless document.root
@@ -58,7 +86,7 @@ module Nokogiri
         sets = paths.map { |path|
           ctx = XPathContext.new(self)
           ctx.register_namespaces(ns)
-          set = ctx.evaluate(path).node_set
+          set = ctx.evaluate(path, handler).node_set
           set.document = document
           document.decorate(set)
           set
@@ -75,8 +103,35 @@ module Nokogiri
         end
       end
 
+      ###
+      # Search this node for CSS +rules+. +rules+ must be one or more CSS
+      # 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 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:
+      #
+      #   node.css('title:regex("\w+")', Class.new(SelectorHandler) {
+      #     def regex node_set, regex
+      #       node_set.find_all { |node| node['some_attribute'] =~ /#{regex}/ }
+      #     end
+      #   })
+      #
       def css *rules
-        xpath(*(rules.map { |rule| CSS.xpath_for(rule, :prefix => ".//") }.flatten.uniq))
+        handler = rules.last.is_a?(XPathHandler) ? rules.pop : nil
+        ns = rules.last.is_a?(Hash) ? rules.pop : {}
+
+        rules = rules.map { |rule|
+          CSS.xpath_for(rule, :prefix => ".//")
+        }.flatten.uniq + [ns, handler].compact
+
+        xpath(*rules)
       end
 
       def at path, ns = {}
@@ -143,6 +198,17 @@ module Nokogiri
         self.native_content = encode_special_chars(string)
       end
 
+      ###
+      # Set the parent Node for this Node
+      def parent= parent_node
+        parent_node.add_child(self)
+        parent_node
+      end
+
+      def << child
+        add_child child
+      end
+
       def comment?
         type == COMMENT_NODE
       end
@@ -182,12 +248,38 @@ module Nokogiri
         ns
       end
 
+      ###
+      # Get a list of ancestor Node for this Node
+      def ancestors
+        parents = []
+
+        this_parent = self.parent
+
+        while this_parent != nil
+          parents << this_parent
+          this_parent = this_parent.parent
+        end
+        parents
+      end
+
       ####
       # Yields self and all children to +block+ recursively.
       def traverse(&block)
         children.each{|j| j.traverse(&block) }
         block.call(self)
       end
+
+      ####
+      #  replace node with the new node in the document.
+      def replace(new_node)
+        if new_node.is_a?(Document)
+          raise ArgumentError, <<-EOERR
+Node.replace requires a Node argument, and cannot accept a Document.
+(You probably want to select a node from the Document with at() or search(), or create a new Node via Node.new().)
+          EOERR
+        end
+        replace_with_node new_node
+      end
     end
   end
 end

data/lib/nokogiri/xml/node_set.rb

@@ -5,8 +5,9 @@ module Nokogiri
 
       attr_accessor :document
 
-      def initialize document
+      def initialize document, list = []
         @document = document
+        list.each { |x| self << x }
         yield self if block_given?
       end
 

data/lib/nokogiri/xml/xpath_handler.rb

@@ -0,0 +1,6 @@
+module Nokogiri
+  module XML
+    class XPathHandler
+    end
+  end
+end

data/lib/nokogiri/xslt.rb

@@ -6,6 +6,23 @@ module Nokogiri
       def parse(string)
         Stylesheet.parse_stylesheet_doc(XML.parse(string))
       end
+      
+      def quote_params params
+        parray = (params.instance_of?(Hash) ? params.to_a.flatten : params).dup
+        parray.each_with_index do |v,i|
+          if i % 2 > 0
+            parray[i]=
+              if v =~ /'/
+                "concat('#{ v.gsub(/'/, %q{', "'", '}) }')"
+              else
+                "'#{v}'";
+              end
+          else
+            parray[i] = v.to_s
+          end
+        end
+        parray.flatten
+      end
     end
   end
 end

data/test/css/test_parser.rb

@@ -158,15 +158,15 @@ module Nokogiri
       end
 
       def test_pseudo_class_no_ident
-        assert_xpath "//*[1 = 1]", @parser.parse(':link')
+        assert_xpath "//*[link(.)]", @parser.parse(':link')
       end
 
       def test_pseudo_class
-        assert_xpath "//a[1 = 1]", @parser.parse('a:link')
-        assert_xpath "//a[1 = 1]", @parser.parse('a:visited')
-        assert_xpath "//a[1 = 1]", @parser.parse('a:hover')
-        assert_xpath "//a[1 = 1]", @parser.parse('a:active')
-        assert_xpath  "//a[1 = 1 and contains(concat(' ', @class, ' '), ' foo ')]",
+        assert_xpath "//a[link(.)]", @parser.parse('a:link')
+        assert_xpath "//a[visited(.)]", @parser.parse('a:visited')
+        assert_xpath "//a[hover(.)]", @parser.parse('a:hover')
+        assert_xpath "//a[active(.)]", @parser.parse('a:active')
+        assert_xpath  "//a[active(.) and contains(concat(' ', @class, ' '), ' foo ')]",
                       @parser.parse('a:active.foo')
       end
 

data/test/css/test_xpath_visitor.rb

@@ -7,6 +7,16 @@ module Nokogiri
         @parser = Nokogiri::CSS::Parser.new
       end
 
+      def test_unknown_psuedo_classes_get_pushed_down
+        assert_xpath("//a[aaron(.)]", @parser.parse('a:aaron'))
+      end
+
+      def test_unknown_functions_get_dot_plus_args
+        assert_xpath("//a[aaron(.)]", @parser.parse('a:aaron()'))
+        assert_xpath("//a[aaron(., 12)]", @parser.parse('a:aaron(12)'))
+        assert_xpath("//a[aaron(., 12, 1)]", @parser.parse('a:aaron(12, 1)'))
+      end
+
       def test_class_selectors
         assert_xpath  "//*[contains(concat(' ', @class, ' '), ' red ')]",
                       @parser.parse(".red")

data/test/files/exslt.xml

@@ -0,0 +1,8 @@
+<?xml version="1.0"?>
+<root>
+   <max>
+      <value>1</value>
+      <value>3</value>
+      <value>2</value>
+   </max>
+</root>

data/test/files/exslt.xslt

@@ -0,0 +1,35 @@
+<?xml version="1.0" encoding="ISO-8859-1"?>
+
+<xsl:stylesheet version="1.0"
+				xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
+				xmlns:func="http://exslt.org/functions"
+				xmlns:my="urn:my-functions"
+                xmlns:date="http://exslt.org/dates-and-times"
+                xmlns:math="http://exslt.org/math"
+				extension-element-prefixes="func date"
+                >
+
+  <xsl:param name="p1"/>
+  <xsl:param name="p2"/>
+  <xsl:param name="p3"/>
+  <xsl:param name="p4"/>
+
+  <xsl:template match="/">
+     <root>
+        <function><xsl:value-of select="my:func()"/></function>
+        <date><xsl:value-of select="date:date()"/></date>
+        <max><xsl:value-of select="math:max(//max/value)"/></max>
+        <params>
+           <p1><xsl:value-of select="$p1"/></p1>
+           <p2><xsl:value-of select="$p2"/></p2>
+           <p3><xsl:value-of select="$p3"/></p3>
+           <p4><xsl:value-of select="$p4"/></p4>
+        </params>
+     </root>
+  </xsl:template>
+
+  <func:function name="my:func">
+	<func:result select="'func-result'"/>
+  </func:function>
+
+</xsl:stylesheet>

data/test/helper.rb

@@ -12,6 +12,8 @@ module Nokogiri
     ASSETS_DIR = File.join(File.dirname(__FILE__), 'files')
     XML_FILE = File.join(ASSETS_DIR, 'staff.xml')
     XSLT_FILE = File.join(ASSETS_DIR, 'staff.xslt')
+    EXSLT_FILE = File.join(ASSETS_DIR, 'exslt.xslt')
+    EXML_FILE = File.join(ASSETS_DIR, 'exslt.xml')
     HTML_FILE = File.join(ASSETS_DIR, 'tlm.html')
 
     unless RUBY_VERSION >= '1.9'