Roger posted a link to this question of Dijkstra's:
> How would you represent a more complex operation,
> for example, the sum of all elements of a matrix
> which are equal to the sum of the corresponding row
> and column indices?
Oleg pondered:
> What is the standard J answer?
Whenever I want a matrix of indicies I think first of the idiom (#:i.)@:$ .
So, in this case, I would write something along the lines of:
(+/@:#&,~ ( = (+/@#: i.)@:$)) i. 3 4
6
(+/@:#&,~ ] = (+/@#: i.)@:$) i. 3 4
6
(+/@ #&,~ ] = (+/@#: i.)@ $) i. 3 4
6
with the particular style chosen to suit the objective. In Dyalog APL (at
least), monadic iota (the analog to i.y ) already produces a matrix of
(enclosed) indicies, so the analogous verb might be briefer. I should really
buy one of those cheap "personal use" licenses, so I could play with ideas like
this.
-Dan
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
