find duplicate files by md5 with ruby

duplicate_finder.rb

#!/usr/bin/env ruby
require 'pp'
files = {}
ctr = 0
print "Searching "
Dir.glob(ARGV[0]).each do |f|
  s = File.stat(f)
  if s.file?
    md5 = (`md5 -q #{f}`).chomp

    files[md5] ||= []
    files[md5] << {:file => f, :stat => s}
    ctr +=1
  end
  if (ctr % 100) == 0
    print '.'
  end
end
print "Done\n"

duplicates = files.select{|k,v| v.length > 1}

puts "Searched #{files.length} files"
puts "Found #{duplicates.length} dups"

duplicates.each do |md5, finfo|
  puts "[#{md5}]"
  finfo.sort_by{|f|f[:file].length}.each do |f|
    puts "  #{f[:file]}"
  end
end