Thank you Raul, that model is intuitive

 ([ [ _1 Z: 5 < -) F:. ] p: i.20

3 5 7 11 13 17 19 23

its not obvious why this terminates.  _1&Z: description makes reference to x 
even if it doesn't exist ("next new value of x" makes no sense if y is the one 
with items) .  I thought it would do something similar to the following.

([ [ 0 Z: 5 < -) F:. ] p: i.20

3 5 7 11 13 17 19 23 31 41 43 47 61 71

([ [ 0 Z: 5 >: -) F:. ] p: i.20

29 37 53 59 67






On Wednesday, August 28, 2019, 06:07:25 p.m. EDT, Henry Rich 
<[email protected]> wrote: 





Your imagination about how to use this matches mine.

But please suggest improvements as they occur to you!  This is being put 
out as a J addon right now, so it will be easy to change as we gain 
experience.  In a few months, when we deem the design solid, the 
implementation will be put into C and frozen there.

Henry Rich

On 8/28/2019 5:59 PM, Raul Miller wrote:
> On Tue, Aug 27, 2019 at 10:47 PM 'Pascal Jasmin' via Beta
> <[email protected]> wrote:
>> I think all references for Z: should refer to u verb?  Would Z: ever be 
>> called in v verb?  _1&Z: makes no sense.
>> As a simple example, find first prime whose next prime is greater than 5 
>> higher?
>>
>>    [`(_2 Z: 1:)@.(5 < -) F.. ] p: i.20
>> 23
> Is this equivalence worth noting?
>
>    ([ [ _2 Z: 5 < -) F.. ] p: i.20
> 23
>
> The right argument to Z: is a on/off control value -- 0 means Z: does
> nothing that time.
>
> With this fold implementation, I anticipate seeing constructs like:
>    f [ opcode Z: test
>
>> How might you get the first prime whose previous neighbour is less than 5 
>> higher? (29)
> Presumably, you were hinting at a variation on this:
>
>    ([ [ 1 Z: 5 < -) F.. ] p: i.20
> 29
>
> Thanks,
>


---
This email has been checked for viruses by AVG.
https://www.avg.com


----------------------------------------------------------------------
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