Compare nokogiri xml parsing vs. marc4j4r xml parsing with a roundtrip through marchash

marc_xml_test.rb

require 'marc'
require 'marc4j4r'
require 'benchmark'
 

iterations = 1
 
xmlsourcefile = 'topics.xml' # 18k records as a MARC-XML collection

puts RUBY_DESCRIPTION
puts "#{iterations} iteration(s)\n"
Benchmark.bm do |x|
  x.report("nokogiri            ") do
    title_length = 0
    iterations.times do
      reader =  MARC::XMLReader.new(xmlsourcefile, :parser=>'nokogiri')
      reader.each do |r|
        title_length += r['245'].value.size
      end
    end
  end
  
  x.report("marc4j4r with marchash") do
    title_length = 0
    iterations.times do
      reader =  MARC4J4R::Reader.new(xmlsourcefile, :marcxml)
      reader.each do |m4jr|
        r = MARC::Record.new_from_marchash(m4jr.to_marchash)
        title_length += r['245'].value.size
      end
    end
  end
  
  x.report("marc4j4r raw") do
    title_length = 0
    iterations.times do
      reader =  MARC4J4R::Reader.new(xmlsourcefile, :marcxml)
      reader.each do |r|
        title_length += r['245'].value.size
      end
    end
  end  
end

results.rxt

/tmp> jruby --server -J-Djruby.compile.invokedynamic=false  t.rb
jruby 1.7.4 (1.9.3p392) 2013-05-16 2390d3b on Java HotSpot(TM) 64-Bit Server VM 1.7.0_40-ea-b27 [darwin-x86_64]
1 iteration(s)
       user     system      total        real
nokogiri                56.680000   0.420000  57.100000 ( 54.955000)
marc4j4r with marchash  44.800000   0.590000  45.390000 ( 35.089000)
marc4j4r raw            19.170000   0.370000  19.540000 ( 14.428000)