Would be good to have tests showing the problem. Did you check if such method is not already covered (even by a bad test)? Stef On Sun, Dec 31, 2017 at 2:47 PM, Andy Burnett <andy.burnett@knowinnovation.com> wrote:

Ben wrote >>>

Its not so much that your first example sorted the permutations, but that the collection contained only one permutation. I've swapped the order of your examples and downsized them to the simplest case to observe. Something seems broken. It works as expected if the "copy" is uncommented.

Transcript clear. oc1 := OrderedCollection new. oc2 := OrderedCollection new. Transcript crShow:'a---'. (1 to: 3) permutationsDo: [ :each | Transcript crShow: each. oc1 add: each asOrderedCollection]. Transcript crShow:'b---'. (1 to: 3) permutationsDo: [ :each | Transcript crShow: each. oc2 add: each "copy"]. Transcript crShow:'c---'. Transcript crShow: { oc1 asSet size. oc2 asSet size}. Transcript crShow:'d---'. oc2 do: [ :x | Transcript crShow: x ].

==> a--- #(1 2 3) #(1 3 2) #(2 1 3) #(2 3 1) #(3 2 1) #(3 1 2) b--- #(1 2 3) #(1 3 2) #(2 1 3) #(2 3 1) #(3 2 1) #(3 1 2) c--- #(6 1) d--- #(1 2 3) #(1 2 3) #(1 2 3) #(1 2 3) #(1 2 3) #(1 2 3)

cheers -ben

<<<

Thanks Ben, That is really interesting. I had completely misunderstood the problem. Checking the items in oc2 shows that they are literally the same object. So, it would seem that I have to run copy, or some other method to get a unique object.

I can see that this would make sense from an efficiency perspective. Reusing the same object would presumably save memory space. However, it would probably be good to update the comment to let people know that this will happen.

What is the process for submitting suggested improvements to class comments?

Cheers Andy