rexml 3.2.5 → 3.4.4

data/lib/rexml/attribute.rb

@@ -1,4 +1,4 @@
-# frozen_string_literal: false
+# frozen_string_literal: true
 require_relative "namespace"
 require_relative 'text'
 
@@ -13,9 +13,6 @@ module REXML
 
     # The element to which this attribute belongs
     attr_reader :element
-    # The normalized value of this attribute.  That is, the attribute with
-    # entities intact.
-    attr_writer :normalized
     PATTERN = /\s*(#{NAME_STR})\s*=\s*(["'])(.*?)\2/um
 
     NEEDS_A_SECOND_CHECK = /(<|&((#{Entity::NAME});|(#0*((?:\d+)|(?:x[a-fA-F0-9]+)));)?)/um
@@ -122,18 +119,18 @@ module REXML
     #  b = Attribute.new( "ns:x", "y" )
     #  b.to_string     # -> "ns:x='y'"
     def to_string
+      value = to_s
       if @element and @element.context and @element.context[:attribute_quote] == :quote
-        %Q^#@expanded_name="#{to_s().gsub(/"/, '&quot;')}"^
+        value = value.gsub('"', '&quot;') if value.include?('"')
+        %Q^#@expanded_name="#{value}"^
       else
-        "#@expanded_name='#{to_s().gsub(/'/, '&apos;')}'"
+        value = value.gsub("'", '&apos;') if value.include?("'")
+        "#@expanded_name='#{value}'"
       end
     end
 
     def doctype
-      if @element
-        doc = @element.document
-        doc.doctype if doc
-      end
+      @element&.document&.doctype
     end
 
     # Returns the attribute value, with entities replaced
@@ -141,7 +138,6 @@ module REXML
       return @normalized if @normalized
 
       @normalized = Text::normalize( @unnormalized, doctype )
-      @unnormalized = nil
       @normalized
     end
 
@@ -149,9 +145,16 @@ module REXML
     # have been expanded to their values
     def value
       return @unnormalized if @unnormalized
-      @unnormalized = Text::unnormalize( @normalized, doctype )
-      @normalized = nil
-      @unnormalized
+
+      @unnormalized = Text::unnormalize(@normalized, doctype,
+                                        entity_expansion_text_limit: @element&.document&.entity_expansion_text_limit)
+    end
+
+    # The normalized value of this attribute.  That is, the attribute with
+    # entities intact.
+    def normalized=(new_normalized)
+      @normalized = new_normalized
+      @unnormalized = nil
     end
 
     # Returns a copy of this attribute
@@ -167,7 +170,7 @@ module REXML
       @element = element
 
       if @normalized
-        Text.check( @normalized, NEEDS_A_SECOND_CHECK, doctype )
+        Text.check( @normalized, NEEDS_A_SECOND_CHECK )
       end
 
       self
@@ -190,15 +193,17 @@ module REXML
     end
 
     def inspect
-      rv = ""
+      rv = +""
       write( rv )
       rv
     end
 
     def xpath
-      path = @element.xpath
-      path += "/@#{self.expanded_name}"
-      return path
+      @element.xpath + "/@#{self.expanded_name}"
+    end
+
+    def document
+      @element&.document
     end
   end
 end

data/lib/rexml/cdata.rb

@@ -58,7 +58,7 @@ module REXML
     #  c = CData.new( " Some text " )
     #  c.write( $stdout )     #->  <![CDATA[ Some text ]]>
     def write( output=$stdout, indent=-1, transitive=false, ie_hack=false )
-      Kernel.warn( "#{self.class.name}.write is deprecated", uplevel: 1)
+      Kernel.warn( "#{self.class.name}#write is deprecated", uplevel: 1)
       indent( output, indent )
       output << START
       output << @string

data/lib/rexml/child.rb

@@ -83,13 +83,12 @@ module REXML
     # Returns:: the document this child belongs to, or nil if this child
     # belongs to no document
     def document
-      return parent.document unless parent.nil?
-      nil
+      parent&.document
     end
 
     # This doesn't yet handle encodings
     def bytes
-      document.encoding
+      document&.encoding
 
       to_s
     end

data/lib/rexml/comment.rb

@@ -48,7 +48,7 @@ module REXML
     # ie_hack::
     #    Needed for conformity to the child API, but not used by this class.
     def write( output, indent=-1, transitive=false, ie_hack=false )
-      Kernel.warn("Comment.write is deprecated.  See REXML::Formatters", uplevel: 1)
+      Kernel.warn("#{self.class.name}#write is deprecated.  See REXML::Formatters", uplevel: 1)
       indent( output, indent )
       output << START
       output << @string

data/lib/rexml/doctype.rb

@@ -171,15 +171,11 @@ module REXML
     end
 
     def context
-      if @parent
-        @parent.context
-      else
-        nil
-      end
+      @parent&.context
     end
 
     def entity( name )
-      @entities[name].unnormalized if @entities[name]
+      @entities[name]&.unnormalized
     end
 
     def add child
@@ -288,8 +284,7 @@ module REXML
     end
 
     def to_s
-      context = nil
-      context = parent.context if parent
+      context = parent&.context
       notation = "<!NOTATION #{@name}"
       reference_writer = ReferenceWriter.new(@middle, @public, @system, context)
       reference_writer.write(notation)

data/lib/rexml/document.rb

@@ -69,7 +69,7 @@ module REXML
     #   d.to_s # => "<root><foo>Foo</foo><bar>Bar</bar></root>"
     #
     # When argument +document+ is given, it must be an existing
-    # document object, whose context and attributes (but not chidren)
+    # document object, whose context and attributes (but not children)
     # are cloned into the new document:
     #
     #   d = REXML::Document.new(xml_string)
@@ -91,9 +91,13 @@ module REXML
     #
     def initialize( source = nil, context = {} )
       @entity_expansion_count = 0
+      @entity_expansion_limit = Security.entity_expansion_limit
+      @entity_expansion_text_limit = Security.entity_expansion_text_limit
       super()
       @context = context
-      return if source.nil?
+      # `source = ""` is an invalid usage because no root element XML is an invalid XML.
+      # But we accept `""` for backward compatibility.
+      return if source.nil? or source == ""
       if source.kind_of? Document
         @context = source.context
         super source
@@ -307,8 +311,8 @@ module REXML
     end
 
     # :call-seq:
-    #    doc.write(output=$stdout, indent=-1, transtive=false, ie_hack=false, encoding=nil)
-    #    doc.write(options={:output => $stdout, :indent => -1, :transtive => false, :ie_hack => false, :encoding => nil})
+    #    doc.write(output=$stdout, indent=-1, transitive=false, ie_hack=false, encoding=nil)
+    #    doc.write(options={:output => $stdout, :indent => -1, :transitive => false, :ie_hack => false, :encoding => nil})
     #
     # Write the XML tree out, optionally with indent.  This writes out the
     # entire XML document, including XML declarations, doctype declarations,
@@ -413,7 +417,7 @@ module REXML
     #
     # Deprecated. Use REXML::Security.entity_expansion_limit= instead.
     def Document::entity_expansion_limit
-      return Security.entity_expansion_limit
+      Security.entity_expansion_limit
     end
 
     # Set the entity expansion limit. By default the limit is set to 10240.
@@ -427,14 +431,16 @@ module REXML
     #
     # Deprecated. Use REXML::Security.entity_expansion_text_limit instead.
     def Document::entity_expansion_text_limit
-      return Security.entity_expansion_text_limit
+      Security.entity_expansion_text_limit
     end
 
     attr_reader :entity_expansion_count
+    attr_writer :entity_expansion_limit
+    attr_accessor :entity_expansion_text_limit
 
     def record_entity_expansion
       @entity_expansion_count += 1
-      if @entity_expansion_count > Security.entity_expansion_limit
+      if @entity_expansion_count > @entity_expansion_limit
         raise "number of entity expansions exceeded, processing aborted."
       end
     end
@@ -444,6 +450,20 @@ module REXML
     end
 
     private
+
+    attr_accessor :namespaces_cache
+
+    # New document level cache is created and available in this block.
+    # This API is thread unsafe. Users can't change this document in this block.
+    def enable_cache
+      @namespaces_cache = {}
+      begin
+        yield
+      ensure
+        @namespaces_cache = nil
+      end
+    end
+
     def build( source )
       Parsers::TreeParser.new( source, self ).parse
     end

data/lib/rexml/element.rb

@@ -7,14 +7,6 @@ require_relative "xpath"
 require_relative "parseexception"
 
 module REXML
-  # An implementation note about namespaces:
-  # As we parse, when we find namespaces we put them in a hash and assign
-  # them a unique ID.  We then convert the namespace prefix for the node
-  # to the unique ID.  This makes namespace lookup much faster for the
-  # cost of extra memory use.  We save the namespace prefix for the
-  # context node and convert it back when we write it.
-  @@namespaces = {}
-
   # An \REXML::Element object represents an XML element.
   #
   # An element:
@@ -449,9 +441,14 @@ module REXML
     # Related: #root_node, #document.
     #
     def root
-      return elements[1] if self.kind_of? Document
-      return self if parent.kind_of? Document or parent.nil?
-      return parent.root
+      target = self
+      while target
+        return target.elements[1] if target.kind_of? Document
+        parent = target.parent
+        return target if parent.kind_of? Document or parent.nil?
+        target = parent
+      end
+      nil
     end
 
     # :call-seq:
@@ -476,8 +473,7 @@ module REXML
     # Related: #root, #root_node.
     #
     def document
-      rt = root
-      rt.parent if rt
+      root&.parent
     end
 
     # :call-seq:
@@ -569,7 +565,7 @@ module REXML
       prefixes = []
       prefixes = parent.prefixes if parent
       prefixes |= attributes.prefixes
-      return prefixes
+      prefixes
     end
 
     # :call-seq:
@@ -592,10 +588,12 @@ module REXML
     #   d.elements['//c'].namespaces # => {"x"=>"1", "y"=>"2", "z"=>"3"}
     #
     def namespaces
-      namespaces = {}
-      namespaces = parent.namespaces if parent
-      namespaces = namespaces.merge( attributes.namespaces )
-      return namespaces
+      namespaces_cache = document&.__send__(:namespaces_cache)
+      if namespaces_cache
+        namespaces_cache[self] ||= calculate_namespaces
+      else
+        calculate_namespaces
+      end
     end
 
     # :call-seq:
@@ -622,15 +620,11 @@ module REXML
       if prefix.nil?
         prefix = prefix()
       end
-      if prefix == ''
-        prefix = "xmlns"
-      else
-        prefix = "xmlns:#{prefix}" unless prefix[0,5] == 'xmlns'
-      end
-      ns = attributes[ prefix ]
-      ns = parent.namespace(prefix) if ns.nil? and parent
+      prefix = (prefix == '') ? 'xmlns' : prefix.delete_prefix("xmlns:")
+      ns = namespaces[prefix]
+
       ns = '' if ns.nil? and prefix == 'xmlns'
-      return ns
+      ns
     end
 
     # :call-seq:
@@ -962,7 +956,7 @@ module REXML
     def next_element
       element = next_sibling
       element = element.next_sibling until element.nil? or element.kind_of? Element
-      return element
+      element
     end
 
     # :call-seq:
@@ -978,7 +972,7 @@ module REXML
     def previous_element
       element = previous_sibling
       element = element.previous_sibling until element.nil? or element.kind_of? Element
-      return element
+      element
     end
 
 
@@ -989,7 +983,7 @@ module REXML
     # :call-seq:
     #   has_text? -> true or false
     #
-    # Returns +true if the element has one or more text noded,
+    # Returns +true+ if the element has one or more text noded,
     # +false+ otherwise:
     #
     #   d = REXML::Document.new '<a><b/>text<c/></a>'
@@ -1006,7 +1000,7 @@ module REXML
     #   text(xpath = nil) -> text_string or nil
     #
     # Returns the text string from the first text node child
-    # in a specified element, if it exists, # +nil+ otherwise.
+    # in a specified element, if it exists, +nil+ otherwise.
     #
     # With no argument, returns the text from the first text node in +self+:
     #
@@ -1014,7 +1008,7 @@ module REXML
     #   d.root.text.class # => String
     #   d.root.text       # => "some text "
     #
-    # With argument +xpath+, returns text from the the first text node
+    # With argument +xpath+, returns text from the first text node
     # in the element that matches +xpath+:
     #
     #   d.root.text(1) # => "this is bold!"
@@ -1028,8 +1022,7 @@ module REXML
     #
     def text( path = nil )
       rv = get_text(path)
-      return rv.value unless rv.nil?
-      nil
+      rv&.value
     end
 
     # :call-seq:
@@ -1057,7 +1050,7 @@ module REXML
       else
         rv = @children.find { |node| node.kind_of? Text }
       end
-      return rv
+      rv
     end
 
     # :call-seq:
@@ -1101,7 +1094,7 @@ module REXML
           old_text.replace_with( text )
         end
       end
-      return self
+      self
     end
 
     # :call-seq:
@@ -1152,7 +1145,7 @@ module REXML
         text = Text.new( text, whitespace(), nil, raw() )
       end
       self << text unless text.nil?
-      return self
+      self
     end
 
     # :call-seq:
@@ -1196,7 +1189,7 @@ module REXML
         cur = cur.parent
         path_elements << __to_xpath_helper( cur )
       end
-      return path_elements.reverse.join( "/" )
+      path_elements.reverse.join( "/" )
     end
 
     #################################################
@@ -1284,16 +1277,11 @@ module REXML
     #   document.root.attribute("x", "a") # => a:x='a:x'
     #
     def attribute( name, namespace=nil )
-      prefix = nil
-      if namespaces.respond_to? :key
-        prefix = namespaces.key(namespace) if namespace
-      else
-        prefix = namespaces.index(namespace) if namespace
-      end
+      prefix = namespaces.key(namespace) if namespace
       prefix = nil if prefix == 'xmlns'
 
       ret_val =
-        attributes.get_attribute( "#{prefix ? prefix + ':' : ''}#{name}" )
+        attributes.get_attribute( prefix ? "#{prefix}:#{name}" : name )
 
       return ret_val unless ret_val.nil?
       return nil if prefix.nil?
@@ -1303,7 +1291,6 @@ module REXML
       return nil unless ( namespaces[ prefix ] == namespaces[ 'xmlns' ] )
 
       attributes.get_attribute( name )
-
     end
 
     # :call-seq:
@@ -1317,7 +1304,7 @@ module REXML
     #   b.has_attributes? # => false
     #
     def has_attributes?
-      return !@attributes.empty?
+      !@attributes.empty?
     end
 
     # :call-seq:
@@ -1506,7 +1493,7 @@ module REXML
     #  doc.write( out )     #-> doc is written to the string 'out'
     #  doc.write( $stdout ) #-> doc written to the console
     def write(output=$stdout, indent=-1, transitive=false, ie_hack=false)
-      Kernel.warn("#{self.class.name}.write is deprecated.  See REXML::Formatters", uplevel: 1)
+      Kernel.warn("#{self.class.name}#write is deprecated.  See REXML::Formatters", uplevel: 1)
       formatter = if indent > -1
           if transitive
             require_relative "formatters/transitive"
@@ -1520,8 +1507,15 @@ module REXML
       formatter.write( self, output )
     end
 
-
     private
+    def calculate_namespaces
+      if parent
+        parent.namespaces.merge(attributes.namespaces)
+      else
+        attributes.namespaces
+      end
+    end
+
     def __to_xpath_helper node
       rv = node.expanded_name.clone
       if node.parent
@@ -1688,11 +1682,7 @@ module REXML
           (num += 1) == index
         }
       else
-        return XPath::first( @element, index )
-        #{ |element|
-        #       return element if element.kind_of? Element
-        #}
-        #return nil
+        XPath::first( @element, index )
       end
     end
 
@@ -1739,7 +1729,7 @@ module REXML
       else
         previous.replace_with element
       end
-      return previous
+      previous
     end
 
     # :call-seq:
@@ -1778,7 +1768,7 @@ module REXML
         child == element
       end
       return rv if found == element
-      return -1
+      -1
     end
 
     # :call-seq:
@@ -1857,7 +1847,7 @@ module REXML
         @element.delete element
         element.remove
       end
-      return rv
+      rv
     end
 
     # :call-seq:
@@ -2184,8 +2174,7 @@ module REXML
     #
     def [](name)
       attr = get_attribute(name)
-      return attr.value unless attr.nil?
-      return nil
+      attr&.value
     end
 
     # :call-seq:
@@ -2328,11 +2317,11 @@ module REXML
             return attr
           end
         end
-        element_document = @element.document
-        if element_document and element_document.doctype
+        doctype = @element.document&.doctype
+        if doctype
           expn = @element.expanded_name
-          expn = element_document.doctype.name if expn.size == 0
-          attr_val = element_document.doctype.attribute_of(expn, name)
+          expn = doctype.name if expn.size == 0
+          attr_val = doctype.attribute_of(expn, name)
           return Attribute.new( name, attr_val ) if attr_val
         end
         return nil
@@ -2340,7 +2329,7 @@ module REXML
       if attr.kind_of? Hash
         attr = attr[ @element.prefix ]
       end
-      return attr
+      attr
     end
 
     # :call-seq:
@@ -2374,8 +2363,9 @@ module REXML
       end
 
       unless value.kind_of? Attribute
-        if @element.document and @element.document.doctype
-          value = Text::normalize( value, @element.document.doctype )
+        doctype = @element.document&.doctype
+        if doctype
+          value = Text::normalize( value, doctype )
         else
           value = Text::normalize( value, nil )
         end
@@ -2388,23 +2378,12 @@ module REXML
       elsif old_attr.kind_of? Hash
         old_attr[value.prefix] = value
       elsif old_attr.prefix != value.prefix
-        # Check for conflicting namespaces
-        if value.prefix != "xmlns" and old_attr.prefix != "xmlns"
-          old_namespace = old_attr.namespace
-          new_namespace = value.namespace
-          if old_namespace == new_namespace
-            raise ParseException.new(
-                    "Namespace conflict in adding attribute \"#{value.name}\": "+
-                    "Prefix \"#{old_attr.prefix}\" = \"#{old_namespace}\" and "+
-                    "prefix \"#{value.prefix}\" = \"#{new_namespace}\"")
-          end
-        end
         store value.name, {old_attr.prefix => old_attr,
                            value.prefix    => value}
       else
         store value.name, value
       end
-      return @element
+      @element
     end
 
     # :call-seq:
@@ -2423,10 +2402,11 @@ module REXML
       each_attribute do |attribute|
         ns << attribute.name if attribute.prefix == 'xmlns'
       end
-      if @element.document and @element.document.doctype
+      doctype = @element.document&.doctype
+      if doctype
         expn = @element.expanded_name
-        expn = @element.document.doctype.name if expn.size == 0
-        @element.document.doctype.attributes_of(expn).each {
+        expn = doctype.name if expn.size == 0
+        doctype.attributes_of(expn).each {
           |attribute|
           ns << attribute.name if attribute.prefix == 'xmlns'
         }
@@ -2448,10 +2428,11 @@ module REXML
       each_attribute do |attribute|
         namespaces[attribute.name] = attribute.value if attribute.prefix == 'xmlns' or attribute.name == 'xmlns'
       end
-      if @element.document and @element.document.doctype
+      doctype = @element.document&.doctype
+      if doctype
         expn = @element.expanded_name
-        expn = @element.document.doctype.name if expn.size == 0
-        @element.document.doctype.attributes_of(expn).each {
+        expn = doctype.name if expn.size == 0
+        doctype.attributes_of(expn).each {
           |attribute|
           namespaces[attribute.name] = attribute.value if attribute.prefix == 'xmlns' or attribute.name == 'xmlns'
         }
@@ -2506,9 +2487,7 @@ module REXML
           old.each_value{|v| repl = v}
           store name, repl
         end
-      elsif old.nil?
-        return @element
-      else # the supplied attribute is a top-level one
+      elsif old # the supplied attribute is a top-level one
         super(name)
       end
       @element
@@ -2562,7 +2541,7 @@ module REXML
         rv << attribute if attribute.expanded_name == name
       }
       rv.each{ |attr| attr.remove }
-      return rv
+      rv
     end
 
     # :call-seq:

data/lib/rexml/encoding.rb

@@ -5,7 +5,7 @@ module REXML
     # ID ---> Encoding name
     attr_reader :encoding
     def encoding=(encoding)
-      encoding = encoding.name if encoding.is_a?(Encoding)
+      encoding = encoding.name if encoding.is_a?(::Encoding)
       if encoding.is_a?(String)
         original_encoding = encoding
         encoding = find_encoding(encoding)
@@ -13,12 +13,9 @@ module REXML
           raise ArgumentError, "Bad encoding name #{original_encoding}"
         end
       end
+      encoding = encoding.upcase if encoding
       return false if defined?(@encoding) and encoding == @encoding
-      if encoding
-        @encoding = encoding.upcase
-      else
-        @encoding = 'UTF-8'
-      end
+      @encoding = encoding || "UTF-8"
       true
     end