find_duplicates.rb

# Encoding: BINARY
require "digest/md5"

abort "usage #$0 path1 [path2] [path3] ..." if ARGV.empty?

hashes = Hash.new{|h,k| h[k] = [] }

roots = ARGV.map{|arg| File.expand_path(arg) }.uniq
roots.each do |root|
  glob = File.join(root, '/**/*')

  Dir.glob glob do |path|
    next unless File.file?(path)
    puts "Analyzing #{path}"
    hash = Digest::MD5.hexdigest(File.read(path).encode!)
    hashes[hash] << path
  end
end

hashes.each do |hash, paths|
  next if paths.size < 2
  puts "\nDuplicate files with checksum: #{hash}", *paths
end