Map reduce unit tests for hadoop

example.rb

#!/usr/bin/env ruby

require './rubydoop'

HADOOP_HOME = '/usr/local/Cellar/hadoop/0.21.0/libexec/'

map do |location, line|
  line.split(/\s+/).each do |word|
    next unless word.strip.length > 0
    emit word.strip.downcase.gsub(/^\(|[^a-zA-Z]$/, ''), location
  end
end

reduce do |key, values|
  emit key, values.join(",")
end

example_test.rb

require 'minitest/unit'
require 'open3'

MiniTest::Unit.autorun

def map_reduce_results(script, input)
  Open3.pipeline_rw("./#{script} map", "sort", "./#{script} reduce") do |in_io, out_io, wt|
    in_io.print input
    in_io.close
    out_io.readlines
  end
end

class ExampleTest < MiniTest::Unit::TestCase
  def test_it
    input = ["file@1\t t this is a sentence", "file@2\t t this is another sentence\n"].join("\n")
    result_list = map_reduce_results('./example.rb', input)
    puts result_list.inspect
    assert(result_list.include?("sentence\tfile@1,file@2\n"))
    assert(result_list.include?("this\tfile@1,file@2\n"))
  end
end