I played with x u}y some more. It seems that y is treated as a vector (rank 1) no matter what rank it actually is. x and u must be either rank 0 or 1 and both the same. Also then number of items in x must match the number of items produced by u. Not sure what is going on. but it does not seem to work as described in nuvoc. The vocabulary does not address this form at all, just gerunds.
On Wed, Dec 28, 2016 at 6:33 AM, Cliff Reiter <[email protected]> wrote: > A long, long, time ago in a J far away, indexing was done in i. order. So > indices to a 3 by 2 array would be given by > i.3 2 > > 0 1 > > 2 3 > > 4 5 > > Is it possible that x u}y is an artifact from that era? > > > > > On 12/27/2016 3:06 PM, Don Guinn wrote: > >> Just played with the original expression a little and got: >> >> 2 4 (0 2"_)}3 2$0 >> 2 0 >> 4 0 >> 0 0 >> 2 4 (0 3"_)}3 2$0 >> 2 0 >> 0 4 >> 0 0 >> 2 4 (0 4"_)}3 2$0 >> 2 0 >> 0 0 >> 4 0 >> 2 4 (0 5"_)}3 2$0 >> 2 0 >> 0 0 >> 0 4 >> 2 4 (0 6"_)}3 2$0 >> |index error >> | 2 4 (0 6"_)}3 2$0 >> 2 4 (0 _1"_)}3 2$0 >> 2 0 >> 0 0 >> 0 4 >> >> On Tue, Dec 27, 2016 at 12:31 PM, Raul Miller <[email protected]> >> wrote: >> >> The verb form is different from the gerund form, for whatever reasons. >>> But this is documented in the dictionary (and in NuVoc, like Henry >>> mentioned earlier). >>> >>> I'm not sure what other answer can be given, other than going into >>> detailed explanations of how they are different, and/or why both forms >>> are useful. Or, perhaps, if we need "moral justification", we could go >>> into examples of quirks in other programming languages (but all >>> programming languages have quirks - they seem to be unavoidable - so >>> that might be boring). >>> >>> -- >>> Raul >>> >>> >>> On Tue, Dec 27, 2016 at 2:20 PM, Brian Schott <[email protected]> >>> wrote: >>> >>>> This is what the OP needed, I think. >>>> Except that instead of the u} form this is the gerund form. >>>> So I don't think it is the answer, either. >>>> >>>> 2 4 [`(0 1"_)`]} 3 2$0 >>>> 2 4 >>>> 2 4 >>>> 0 0 >>>> >>>> >>>> On Tue, Dec 27, 2016 at 2:09 PM, Henry Rich <[email protected]> >>>> >>> wrote: >>> >>>> I have removed the deprecation - it was my misunderstanding of something >>>>> Roger wrote. >>>>> >>>>> The gerund form would be, in simplest form, >>>>> >>>>> 2 4 [`0:`]} 3 2$0 >>>>> >>>>> -- >>>>> >>>> (B=) >>>> ---------------------------------------------------------------------- >>>> For information about J forums see http://www.jsoftware.com/forums.htm >>>> >>> ---------------------------------------------------------------------- >>> For information about J forums see http://www.jsoftware.com/forums.htm >>> >>> ---------------------------------------------------------------------- >> For information about J forums see http://www.jsoftware.com/forums.htm >> > > ---------------------------------------------------------------------- > For information about J forums see http://www.jsoftware.com/forums.htm > ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
