#!/usr/bin/ruby

# this script is used like sdupes and prints found duplicates with unequal md5sums.
# depends on ruby and the md5sum command-line utility.

clusters = %x{sdupes -c}.split "\n\n"
clusters.each {|cluster|
  paths = cluster.split "\n"
  md5sums = paths.map {|path| %x{md5sum "#{path}"}.split(" ").first}
  next if 1 == md5sums.uniq.size
  paths.each_with_index {|path, index|
    puts "false positive: #{md5sums[index]} #{path}"
  }
}