remove_dups.rb

#!/usr/bin/env ruby
require 'digest'

all = Hash.new

Dir["*.jpg"].each do |file|
  md5 = Digest::MD5.hexdigest(IO.read(file))
  ary = all.fetch(md5, [])
  ary << file
  all[md5] = ary
end


all.each do |k,v|
  if v.size > 1
    puts("#{k} has #{v.size} entries: #{v}")
    while v.size > 1
      found = v.find { |f| f.include?("copy") }
      break unless found

      File.unlink(found)
      v.delete(found)
      puts "Removing #{found}"
    end
  end
end