From: Intransition <transfire@...>
Date: 2010-05-09T20:51:37+09:00
Subject: Re: how to remove duplicates from hash



On May 9, 7:30 am, Lucky Nl <lakshmi2...@gmail.com> wrote:
> Hi ,
>   i have the values in hash is like
>  list = {"1" => "ab" ,"2" => "ab","3" => "hjk"}
>  now i want to remove duplicate value "ab"
> i want the result like
>
> list = {"1" => "ab" ,"3" => "hjk"}

Tricky, b/c order becomes important here, which traditionally is not
"Hash". But if you are using Ruby 1.9+ insertion order is preserved so
you can do it. The simplest way is probably:

  list.invert.invert

But in that case you get:

  {"2" => "ab" ,"3" => "hjk"}

If you can't adjust for that, then you can do something like

  list.to_a.reverse.to_h.invert.invert

Where #to_h comes from Ruby Facets. Nice but not very efficient. We
can do it ourselves a bit faster:

  class Hash
    def reverse_invert
      h = {}
      reverse_each{ |k,v| h[v] = k }
      h.invert
    end
  end

~Trans