xml.rb

#!/usr/bin/env ruby

require 'nokogiri'

def process_example(message, xml_text, use_noblanks_option)
  puts message
  puts "XML text: #{xml_text.inspect}"
  doc = Nokogiri::XML(xml_text) { |config| use_noblanks_option ? config.noblanks : config }
  puts 'Resulting XML document:'
  puts doc.inspect; puts; puts
end

process_example(
    "Without the noblanks option, the separating whitespace is parsed into Text elements.",
    "<xml>\n  <something/>\n</xml>",
    false)

process_example(
    "With the noblanks option, the separating whitespace is *not* parsed into Text elements.",
    "<xml>\n  <something/>\n</xml>",
    true)                

process_example(
    "But when the whitespace is an attribute it will be preserved.",
    %Q{<xml><something an_attribute=" \n " /></xml>},
    true)
                                           
process_example(
    "And when the whitespace is the value of an element it will also be preserved.",
    %Q{<xml><something> \n </something></xml>},
    true)

=begin

Output is:
    
XML text: "<xml>\n  <something/>\n</xml>"
Resulting XML document:
#<Nokogiri::XML::Document:0x8c name="document" children=[#<Nokogiri::XML::Element:0x78 name="xml" children=[#<Nokogiri::XML::Text:0x3c "\n  ">, #<Nokogiri::XML::Element:0x50 name="something">, #<Nokogiri::XML::Text:0x64 "\n">]>]>


With the noblanks option, the separating whitespace is *not* parsed into Text elements.
XML text: "<xml>\n  <something/>\n</xml>"
Resulting XML document:
#<Nokogiri::XML::Document:0xc8 name="document" children=[#<Nokogiri::XML::Element:0xb4 name="xml" children=[#<Nokogiri::XML::Element:0xa0 name="something">]>]>


But when the whitespace is an attribute it will be preserved.
XML text: "<xml><something an_attribute=\" \n \" /></xml>"
Resulting XML document:
#<Nokogiri::XML::Document:0x118 name="document" children=[#<Nokogiri::XML::Element:0x104 name="xml" children=[#<Nokogiri::XML::Element:0xf0 name="something" attributes=[#<Nokogiri::XML::Attr:0xdc name="an_attribute" value="   ">]>]>]>


And when the whitespace is the value of an element it will also be preserved.
XML text: "<xml><something> \n </something></xml>"
Resulting XML document:
#<Nokogiri::XML::Document:0x168 name="document" children=[#<Nokogiri::XML::Element:0x154 name="xml" children=[#<Nokogiri::XML::Element:0x140 name="something" children=[#<Nokogiri::XML::Text:0x12c " \n ">]>]>]>

=end