Question

I will just paste down a simple example i tried so that it would be clear to those who read this.

irb(main):001:0> h =  { }
=> {}
irb(main):002:0> a=[1,2,3]
=> [1, 2, 3]
irb(main):003:0> a.object_id
=> 69922343540500


irb(main):004:0> h[a] = 12          #Hash with the array as a key
=> 12
irb(main):005:0> a << 4             #Modified the array
=> [1, 2, 3, 4]
irb(main):006:0> a.object_id        #Object id obviously remains the same.
=> 69922343540500
irb(main):007:0> h[a]               #Hash with the same object_id now returns nil.
=> nil
irb(main):008:0> h                  #Modified hash
=> {[1, 2, 3, 4]=>12}
irb(main):009:0> h[[1,2,3,4]]       #Tried to access the value with the modified key -
=> nil


irb(main):011:0> h.each { |key,value| puts "#{key.inspect} maps #{value}" }
[1, 2, 3, 4] maps 12
=> {[1, 2, 3, 4]=>12}

Now when i iterate over the hash, its possible to identify the map between the key and the value.

Could some one please explain me this behaviour of ruby hash and what are the properties of hash keys.

1) As i mentioned above, the object_id hasn’t changed – then why is the value set to nil.

2) Is there any possible way so that i can get back the value ’12’ from the hash ‘h’ because h[[1,2,3,4]] as mentioned above returns nil.

Answer 1

Hash keys are checked using the #eql? method, and since [1, 2, 3] isn’t .eql? to [1, 2, 3,4] your hash lookup has a different result.

Maybe you want to be using something other than an Array as your Hash key if the semantics aren’t working for you?