greg heil wrote: > John > > Implementing an in place optimization that covered all of C. would be > very worthwhile. However as the main goal is time efficiency, there > will be a cross over point, as the number of items to be shuffled > approaches n, where it may just be simpler to punt and copy the whole > structure.
I agree. That's why I suggested (-p) C. name for short positive p. For example, Forth has rot as a primitive correpsonding to p=:2 3 1 (or possibly its inverse). > i think postscript retained all the forth stack operators? PostScript has considerably fewer (e.g. it does not have nip, tuck, or rot). Best wishes, John ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
