Ruby script to do tests of parsing xml with Nokogiri, Hpricot parsers.

test_xml.rb

# **************************************
#  XML パーステスト
# **************************************
#
require 'hpricot'
require 'nokogiri'

class TestXml
  FILE_XML = "sitemap.xml"
  N = 100

  def proc_main
    begin
      # Nokogiri
      t1 = Time.now
      N.times { |i| parse_nokogiri }
      t2 = Time.now
      printf("[Nokogiri] Time: %6.2f sec.\n", t2 - t1)

      # Hpricot
      t1 = Time.now
      N.times { |i| parse_hpricot }
      t2 = Time.now
      printf("[Hpricot ] Time: %6.2f sec.\n", t2 - t1)
    rescue => e
      puts "[#{e.class}] #{e.message}"
      e.backtrace.each{|trace| puts "\t#{trace}"}
      exit
    end
  end

  private

  def parse_nokogiri
    begin
      doc = Nokogiri::XML(File.read(FILE_XML))
      ns = {"ns" => "http://www.sitemaps.org/schemas/sitemap/0.9"}
      #locs = doc.css('urlset url loc')  # <= 遅い
      locs = doc.xpath('/ns:urlset/ns:url/ns:loc', ns)
    rescue => e
      raise
    end
  end

  def parse_hpricot
    begin
      doc = Hpricot::XML(File.read(FILE_XML))
      #locs = doc/:urlset/:url/:loc   # <= 遅い
      #locs = doc/('urlset url loc')  # <= 遅い
      locs = doc/('/urlset/url/loc')
    rescue => e
      raise
    end
  end
end

TestXml.new.proc_main