The left argument defines the structure of the result. The count of the nub of the left argument defines the number of items of the result. The order of the nub defines the order of the items of the result.
The values of the left argument specify which item of the right argument go into which item of that result. And, the order of the items from the right argument that match a specific value of the left argument which define the order of the right argument items used to build that specific result item (for example, using X </. Y). So ... it's the "only the nub" which is relevant here, if you do not care about the calculation which produces the result. Thanks, -- Raul On Mon, Oct 14, 2019 at 12:42 AM 'Jim Russell' via Programming <[email protected]> wrote: > > One more comment re group by: > Don't know k or the connection machine (and struggle with take always, and > goezinta's) but "group by" maybe helps me understand something. Implicit in a > "group by" is that the grouping isn't by the referenced columns, it is by the > unique, or distinct values in those columns. So it's neat that you can use ~ > to throw the complete "column" at #/. , but it is only the unique values, the > nub, that is defining the categories. > > Or an I misunderstanding that too? > ---------------------------------------------------------------------- > For information about J forums see http://www.jsoftware.com/forums.htm ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
