Yep got it. I was overlooking the fact that J matches the arguments and you
need to use / (Table) to apply atoms of the left argument to all of the atoms
of the right argument
1 2 x:"0 [ 3 4 NB. applies 1 x: 3 and 2 x: 4
3 0
4 1
1 2 x:"0 [ 3 NB. applies 1 x: 3 and 2 x: 3
3 0
3 1
I haven't made that mistake for a while, but it does remind me that there are
very simple rules which, when forgotten, really leave you in the dust. It is
refreshing to visit beginners mind!
Thanks for the prodding.
Cheers, bob
> On Nov 1, 2018, at 12:02 PM, Brian Schott <[email protected]> wrote:
>
> Bob,
>
> "0 behaves like "0 0 on dyadic verbs.
>
> On Thu, Nov 1, 2018 at 2:48 PM 'robert therriault' via Programming <
> [email protected]> wrote:
>
>> Any ideas on why I am getting this result? I am running the current
>> version of j807 but this seems to be the same all the way back to J701,
>> which makes me think that there is something that I am missing in the way
>> that rank is applied.
>>
>> a
>> 0 1 2
>> 3 4 5
>> 1 2 x:"0 _ a NB. this is expected
>> 0 1 2
>> 3 4 5
>> 0 0 0
>>
>> 0 0 0
>> 0 0 0
>> 0 0 0
>>
>>
>> 0 1 0
>> 1 1 0
>> 2 1 0
>>
>> 3 1 0
>> 4 1 0
>> 5 1 0
>> 1 2 x:"0 a NB. this is was unexpected. Shouldn't it be the same as
>> the result of 1 2 x:"0 _ a
>> 0 0
>> 1 0
>> 2 0
>>
>> 3 1
>> 4 1
>> 5 1
>> 2 x:"0 a NB. the constituent parts behave as expected with rank 0
>> 0 1
>> 1 1
>> 2 1
>>
>> 3 1
>> 4 1
>> 5 1
>> 1 x:"0 a
>> 0 1 2
>> 3 4 5
>>
>> Cheers, bob
>> ----------------------------------------------------------------------
>> For information about J forums see http://www.jsoftware.com/forums.htm
>
>
>
> --
> (B=) <-----my sig
> Brian Schott
> ----------------------------------------------------------------------
> For information about J forums see http://www.jsoftware.com/forums.htm
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
