Comments on: Memo #3: Advanced usage of Ruby hashes and arrays

By: Martin

Martin — Sun, 25 Jan 2009 17:48:29 +0000

I blogged about it too, a few months ago. And Ilya Grigorik had an interesting post with some other nice Ruby fu's.

By: Peter Szinek

Peter Szinek — Sun, 25 Jan 2009 10:20:21 +0000

Ah sure… I didn’t notice that the elements of the original array are not unique (should have read the example better, counting the duplicates was the idea… doh)

By: Dmytro Shteflyuk

Dmytro Shteflyuk — Sun, 25 Jan 2009 10:12:46 +0000

Peter, you code would not work. You will got the same array with numbers 1, 2, 3 for all fruits. Hash will be instantiated once and all keys will refer the same instance.

By: Peter Szinek

Peter Szinek — Sun, 25 Jan 2009 09:58:55 +0000

Or in this case (i.e. single default value):

1	h = Hash.new([])

By: Dmytro Shteflyuk

Dmytro Shteflyuk — Sun, 25 Jan 2009 07:46:30 +0000

In reply to john. Wow, it looks much better. Thank you!

By: john

john — Sun, 25 Jan 2009 01:59:31 +0000

What about . . . (for your example with h = Hash.new { |h, key| h[key] = [] }):

1
2
3
4
5
6

a = %w(apple banana apple)
h = Hash.new { [] }
a.each_with_index do |fruit, i|
h[fruit] <<= i
end
puts h.inspect