Re: [Bug-apl] Strange behaviour of ,/

[Elias Mårtenson]

Jay,

Thank you. This made me aware of the fact that I still don't understand this. You are right that the rank is zero, shown by the fact that ⍴,/1 2 3 returns an empty result:

      8⎕CR ⍴,/1 2 3

┌⊖┐

│0│

└─┘

Can you explain to me why that is the case? When I look at the result from ,/1 2 3 it looks like an array that contains a single element: another array with the values 1 2 3 in it. But it isn't. The result is rank zero, and I can't take the first element from it:

      (,/1 2 3)[1]

RANK ERROR

      (,/1 2 3)[1]

But wait, I can take the first element from it:

      ↑,/1 2 3

1 2 3

What's going on here?

Regards,

Elias

On 3 February 2014 18:07, Jay Foad <address@hidden> wrote:

On 31 January 2014 17:37, Elias Mårtenson <address@hidden> wrote:
>       8⎕CR ,/1 2 3 4
> ┌─────────┐
> │┌→──────┐│
> ││1 2 3 4││
> │└───────┘│
> └∊────────┘

I just have one comment to add to Kacper's excellent explanation: a
useful thing to remember about this implementation of *reduction* is
that it always *reduces* the rank of the argument by one (except
there's a special case for scalar arguments).

The vector 1 2 3 4 has rank 1, so the result must have rank 0, so you
get an enclosed vector.

Jay.