On 7/13/06, R&S HUI <[EMAIL PROTECTED]> wrote:
For swapping items, x A. y has a high "gee whiz" factor, but (<i,j)C. y is more practical.  For 
example, to swap last two items, one could say 1&A.  or  (<_1 _2)&C.  .  To swap the first two 
items,  (<0 1)&C.  does it.  To do it using A. is less straightforward.

Yes. But assuming in place operations are easier at the distal end...?
The A. is quite perspicuous - more terse, much like o. For stack
operations, one end is good enough.

If you could do in place swap operations more generically say those
which preserved element sizes and array shape ... well the whole world
of group theory is your oyster. i could use a lot of efficient in
place operations, eg moving sprites around in a bitmap...

~greg
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Reply via email to