I think you're right, Elijah.

(0 p: 0) or (0 p: 1) breaks a lot of things.
But AFAICT not (0 p: n) for n>1

For instance:

list i.3

0 1 2

wcsize''

60 36

0 p: 0

1

list i.3

wcsize''

0 36

JVERSION

Engine: j903/j64avx2/darwin

Release-b: commercial/2022-01-28T04:20:11

Library: 9.03.08

Qt IDE: 1.9.5/5.15.2(5.15.2)

Platform: Darwin 64

Installer: J903 install

InstallPath: /applications/j903

Contact: www.jsoftware.com

This behavior is different in j902.

But (0 p: 0) does alter (wcsize'')


Haven't time right now for further testing.

On Sat, 26 Feb 2022 at 05:44, Elijah Stone <[email protected]> wrote:

> I can consistently reproduce this by running 0 p: 0, cf:
>
>     0 p: 0
> 1
>     (0&p: -: 1&p:) i.1e6
> 1
>
> (new session)
>
>     (0&p: -: 1&p:) i.1e6
> 0
>
> I would guess maybe there is some cache somewhere, and 0 is messing it up?
>
>   -E
>
> On Fri, 25 Feb 2022, Raul Miller wrote:
>
> > https://www.jsoftware.com/help/dictionary/dpco.htm
> > or https://code.jsoftware.com/wiki/Vocabulary/pco
> >
> > 0 p: y is supposed to be -. 1 p: y
> >
> > But, currently (j903), 0 p: y is the same as 1 p: y
> >
> > That said, I should perhaps note that I was looking for a "composite"
> > test -- a concise way of expressing {{ (1 < y) * 0 = 1 p: y }}
> >
> > FYI,
> >
> > --
> > Raul
> > ----------------------------------------------------------------------
> > 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

Reply via email to