I would use a fork, and 'bind' the left argument to the result of f. There is
a difficulty, because you can't compose adverbs with verbs. Perhaps something
like this:
c=. (*@] {{x}y}} ,:) f
c i.10
0 1 2 3 1 2 6 1 2 9
You can get rid of the explicit definition using boxes, if you really care to,
but I would not.
-E
On Sun, 29 May 2022, Brian Schott wrote:
> I have a problem similar to the one shown below.
> Both the verbs g and h depend on calculating f first so this seems like an
> inefficient way of doing it where f produces only nonnegative integers from
> its argument.
> Is there a better approach?
>
> g=:,:f
> h=:*@f
> f=: 3&| NB. not the real f, but f is more compute heavy
> 3 : '(h y)}g y' i.10 NB. an example calculation
> 0 1 2 3 1 2 6 1 2 9
> ]`h`g } i. 10 NB. another "tacit" calculation
> 0 1 2 3 1 2 6 1 2 9
>
>
>
> --
> (B=) <-----my sig
> Brian Schott
> ----------------------------------------------------------------------
> For information about J forums see http://www.jsoftware.com/forums.htm
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
