helpful, thanks Raul. I wasn't so much married to & as not wanting to duplicate both sides in fork. (it seemed like a job for &)
You failed on the bonus though :P ----- Original Message ----- From: Raul Miller <[email protected]> To: Programming forum <[email protected]> Cc: Sent: Thursday, September 24, 2015 10:35 PM Subject: Re: [Jprogramming] towards a reshape arguments adverb On Thu, Sep 24, 2015 at 10:05 PM, 'Pascal Jasmin' via Programming <[email protected]> wrote: > for vectors, this is ok > > 7 (([ $~ >.&# ) ; ] $~ >.&#) i.5 > ┌─────────┬─────────┐ > │7 7 7 7 7│0 1 2 3 4│ > └─────────┴─────────┘ > > I thought this should work too though. > > 7 ;&($~ >.&#) i.5 > ┌─┬─────────┐ > │7│0 1 2 3 4│ > └─┴─────────┘ > > > Is there a way to chain & to get first result? That depends, what is the purpose of this? For example, this gets the first result: 7 ((#~#) ; ] ) i.5 It doesn't use & but you could easily throw one in: 7 ((#~#) ;&] ] ) i.5 Another approach, which accomplishes something a bit different, would be: 7 ;/@,: i.5 Again, doesn't use & but "use X" specifications are broken more often than not. But, if you really want to "use &" you can always easily add it in: 7 ;&]/@,: i.5 7 ;/@,:&] i.5 Yet another approach, perhaps closer to your original result for a broader domain, would be: 7 (>.&# $&.> ;) i.5 But I don't know what you'd use that for... > can you make an adverb that reshapes the right argument such that a > rank0 verb would produce a sensible result, or more impressive, > reshape both arguments if they are incompatible to reproduce "0 1 > behaviour? ie. > > 7 6 +/ i.5 > 7 8 9 10 11 > 6 7 8 9 10 I have no idea in this example what "sensible" means. But, sure, I guess I can do this: adv=: / Thanks, -- 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
