class REXML::Attribute

Defines an Element Attribute; IE, a attribute=value pair, as in: <element attribute=“value”/>. Attributes can be in their own namespaces. General users of REXML will not interact with the Attribute class much.

Constants

NEEDS_A_SECOND_CHECK
PATTERN

Attributes

element[R]

The element to which this attribute belongs

normalized[W]

The normalized value of this attribute. That is, the attribute with entities intact.

Public Class Methods

new( first, second=nil, parent=nil ) click to toggle source

Constructor. FIXME: The parser doesn't catch illegal characters in attributes

first

Either: an Attribute, which this new attribute will become a clone of; or a String, which is the name of this attribute

second

If first is an Attribute, then this may be an Element, or nil. If nil, then the Element parent of this attribute is the parent of the first Attribute. If the first argument is a String, then this must also be a String, and is the content of the attribute. If this is the content, it must be fully normalized (contain no illegal characters).

parent

Ignored unless first is a String; otherwise, may be the Element parent of this attribute, or nil.

Attribute.new( attribute_to_clone )
Attribute.new( attribute_to_clone, parent_element )
Attribute.new( "attr", "attr_value" )
Attribute.new( "attr", "attr_value", parent_element )
# File lib/rexml/attribute.rb, line 44
def initialize( first, second=nil, parent=nil )
  @normalized = @unnormalized = @element = nil
  if first.kind_of? Attribute
    self.name = first.expanded_name
    @unnormalized = first.value
    if second.kind_of? Element
      @element = second
    else
      @element = first.element
    end
  elsif first.kind_of? String
    @element = parent
    self.name = first
    @normalized = second.to_s
  else
    raise "illegal argument #{first.class.name} to Attribute constructor"
  end
end

Public Instance Methods

==( other ) click to toggle source

Returns true if other is an Attribute and has the same name and value, false otherwise.

# File lib/rexml/attribute.rb, line 95
def ==( other )
  other.kind_of?(Attribute) and other.name==name and other.value==value
end
clone() click to toggle source

Returns a copy of this attribute

# File lib/rexml/attribute.rb, line 144
def clone
  Attribute.new self
end
doctype() click to toggle source
# File lib/rexml/attribute.rb, line 118
def doctype
  if @element
    doc = @element.document
    doc.doctype if doc
  end
end
element=( element ) click to toggle source

Sets the element of which this object is an attribute. Normally, this is not directly called.

Returns this attribute

# File lib/rexml/attribute.rb, line 152
def element=( element )
  @element = element

  if @normalized
    Text.check( @normalized, NEEDS_A_SECOND_CHECK, doctype )
  end

  self
end
hash() click to toggle source

Creates (and returns) a hash from both the name and value

# File lib/rexml/attribute.rb, line 100
def hash
  name.hash + value.hash
end
inspect() click to toggle source
# File lib/rexml/attribute.rb, line 178
def inspect
  rv = ""
  write( rv )
  rv
end
namespace(arg=nil) click to toggle source

Returns the namespace URL, if defined, or nil otherwise

e = Element.new("el")
e.add_namespace("ns", "http://url")
e.add_attribute("ns:a", "b")
e.add_attribute("nsx:a", "c")
e.attribute("ns:a").namespace # => "http://url"
e.attribute("nsx:a").namespace # => nil
# File lib/rexml/attribute.rb, line 88
def namespace arg=nil
  arg = prefix if arg.nil?
  @element.namespace arg
end
node_type() click to toggle source
# File lib/rexml/attribute.rb, line 174
def node_type
  :attribute
end
prefix() click to toggle source

Returns the namespace of the attribute.

e = Element.new( "elns:myelement" )
e.add_attribute( "nsa:a", "aval" )
e.add_attribute( "b", "bval" )
e.attributes.get_attribute( "a" ).prefix   # -> "nsa"
e.attributes.get_attribute( "b" ).prefix   # -> "elns"
a = Attribute.new( "x", "y" )
a.prefix                                   # -> ""
Calls superclass method
# File lib/rexml/attribute.rb, line 72
def prefix
  pf = super
  if pf == ""
    pf = @element.prefix if @element
  end
  pf
end
remove() click to toggle source

Removes this Attribute from the tree, and returns true if successfull

This method is usually not called directly.

# File lib/rexml/attribute.rb, line 165
def remove
  @element.attributes.delete self.name unless @element.nil?
end
to_s() click to toggle source

Returns the attribute value, with entities replaced

# File lib/rexml/attribute.rb, line 126
def to_s
  return @normalized if @normalized

  @normalized = Text::normalize( @unnormalized, doctype )
  @unnormalized = nil
  @normalized
end
to_string() click to toggle source

Returns this attribute out as XML source, expanding the name

a = Attribute.new( "x", "y" )
a.to_string     # -> "x='y'"
b = Attribute.new( "ns:x", "y" )
b.to_string     # -> "ns:x='y'"
# File lib/rexml/attribute.rb, line 110
def to_string
  if @element and @element.context and @element.context[:attribute_quote] == :quote
    %Q^#@expanded_name="#{to_s().gsub(/"/, '&quote;')}"^
  else
    "#@expanded_name='#{to_s().gsub(/'/, '&apos;')}'"
  end
end
value() click to toggle source

Returns the UNNORMALIZED value of this attribute. That is, entities have been expanded to their values

# File lib/rexml/attribute.rb, line 136
def value
  return @unnormalized if @unnormalized
  @unnormalized = Text::unnormalize( @normalized, doctype )
  @normalized = nil
  @unnormalized
end
write( output, indent=-1 ) click to toggle source

Writes this attribute (EG, puts 'key=“value”' to the output)

# File lib/rexml/attribute.rb, line 170
def write( output, indent=-1 )
  output << to_string
end
xpath() click to toggle source
# File lib/rexml/attribute.rb, line 184
def xpath
  path = @element.xpath
  path += "/@#{self.expanded_name}"
  return path
end