Rakefile in buftok-0.1 vs Rakefile in buftok-0.2.0

@@ -1,31 +1,66 @@
-require 'rake'
-require 'rake/rdoctask'
-require 'rake/gempackagetask'
-require 'spec/rake/spectask'
+require 'bundler'
+require 'rdoc/task'
+require 'rake/testtask'
 
-Spec::Rake::SpecTask.new(:spec) do |task|
-    task.spec_files = FileList['**/*_spec.rb']
-end
+task :default => :test
 
-Rake::RDocTask.new(:rdoc) do |task|
-    task.rdoc_dir = 'doc'
-    task.title    = 'BufferedTokenizer'
-    task.rdoc_files.include('lib/**/*.rb')
+Bundler::GemHelper.install_tasks
+
+RDoc::Task.new do |task|
+  task.rdoc_dir = 'doc'
+  task.title    = 'BufferedTokenizer'
+  task.rdoc_files.include('lib/**/*.rb')
 end
 
-spec = Gem::Specification.new do |s|
-  s.name = %q{buftok}
-  s.version = "0.1"
-  s.date = %q{2006-12-18}
-  s.summary = %q{BufferedTokenizer extracts token delimited entities from a sequence of arbitrary inputs}
-  s.email = %q{[e-mail]}
-  s.homepage = %q{http://buftok.rubyforge.org}
-  s.rubyforge_project = %q{buftok}
-  s.has_rdoc = true
-  s.authors = ["Tony Arcieri","Martin Emde"]
-  s.files = ["Rakefile", "lib", "lib/buftok.rb"]
+Rake::TestTask.new :test do |t|
+  t.libs << 'lib'
+  t.test_files = FileList['test/**/*.rb']
 end
 
-Rake::GemPackageTask.new(spec) do |pkg|
-  pkg.need_tar = true
+desc "Benchmark the current implementation"
+task :bench do
+  require 'benchmark'
+  require File.expand_path('lib/buftok', File.dirname(__FILE__))
+
+  n = 50000
+  delimiter = "\n\n"
+
+  frequency1 = 1000
+  puts "generating #{n} strings, with #{delimiter.inspect} every #{frequency1} strings..."
+  data1 = (0...n).map do |i|
+    (((i % frequency1 == 1) ? "\n" : "") +
+      ("s" * i) +
+      ((i % frequency1 == 0) ? "\n" : "")).freeze
+  end
+
+  frequency2 = 10
+  puts "generating #{n} strings, with #{delimiter.inspect} every #{frequency2} strings..."
+  data2 = (0...n).map do |i|
+    (((i % frequency2 == 1) ? "\n" : "") +
+      ("s" * i) +
+      ((i % frequency2 == 0) ? "\n" : "")).freeze
+  end
+
+  Benchmark.bmbm do |x|
+    x.report("1 char, freq: #{frequency1}") do
+      bt1 = BufferedTokenizer.new
+      n.times { |i| bt1.extract(data1[i]) }
+    end
+
+    x.report("2 char, freq: #{frequency1}") do
+      bt2 = BufferedTokenizer.new(delimiter)
+      n.times { |i| bt2.extract(data1[i]) }
+    end
+
+    x.report("1 char, freq: #{frequency2}") do
+      bt3 = BufferedTokenizer.new
+      n.times { |i| bt3.extract(data2[i]) }
+    end
+
+    x.report("2 char, freq: #{frequency2}") do
+      bt4 = BufferedTokenizer.new(delimiter)
+      n.times { |i| bt4.extract(data2[i]) }
+    end
+
+  end
 end