String similariry in ruby. Origin http://stackoverflow.com/questions/653157/a-better-similarity-ranking-algorithm-for-variable-length-strings

string_distace.rb

def distance(str1, str2)
  str1.downcase!
  pairs1 = (0..str1.length-2).collect {|i| str1[i,2]}.reject {
    |pair| pair.include? " "}
  str2.downcase!
  pairs2 = (0..str2.length-2).collect {|i| str2[i,2]}.reject {
    |pair| pair.include? " "}
  union = pairs1.size + pairs2.size
  intersection = 0
  pairs1.each do |p1|
    0.upto(pairs2.size-1) do |i|
      if p1 == pairs2[i]
        intersection += 1
        pairs2.slice!(i)
        break
      end
    end
  end
  (2.0 * intersection) / union
end