deDuper_vs3.rb

#!/usr/bin/env ruby
#from http://www.psteiner.com/2007/06/ruby-script-find-duplicate-files.html
require 'find'

library_path = ARGV[0]

dir = Dir.glob(library_path + "/**/*").select{ |x| File.file? x }


files = {}
found = {}

# read root directory from command line    
dir.each do |arg|
  Find.find(arg) do |f|
    if File.file?(f) 
      # accumulate the file names
      files[f] = File.basename(f)
    end
  end
end

# count up the number of each file name 
files.each_value do |base|
  # Ruby doesn't allow this Perl idiom: found[base]++
  found[base] = 0 if !found[base]
  found[base] += 1
end

# print the path of each file found more than once,
# prepended with rm command commented out
found.each do |name,count|
  if count > 1
    files.each do |path,filename|
      if name == filename
        puts "# rm #{path}"
      end
    end
  end
end