[Jprogramming] ^:n always returning atoms for n<_1

[Henry Rich]

Fixed for the next beta.

When n has negative values, u^:n computes all the powers in an array and 
then performs

($u) $ array

to get the result.  The code was actually computing (($u) ($,) array).  
So if u was not a list, and the items of the array were not atoms, the 
result would have the wrong shape.

Henry Rich

On 5/21/2021 8:15 AM, ethiejiesa via Programming wrote:
>         9!:14''
>      j902-b/j64avx2/linux/release/GPL3/guix.gnu.org/1970-01-01T00:00:00/clang-9-0-1/SLEEF=1
>
>         +&1^:_1 i.3
>      _1 0 1
>         +&1^:_2 i.3
>      _2
>         $$ +&1^:_2 i.3
>      0
>
> Similar results for other negative exponents smaller than _1.
>
> My beta is a bit behind, but I see the same there, while my j901 is producing
> the expected results. Specific version info below:
>
>    j903-h/j64avx2/linux/beta/GPL3/guix.gnu.org/1970-01-01T00:00:00/clang-9-0-1/SLEEF=1
>    j901-f/j64avx2/linux/release/GPL3/guix.gnu.org/1970-01-01T00:00:00
>
> Can anyone reproduce?
> ----------------------------------------------------------------------
> For information about J forums see http://www.jsoftware.com/forums.htm


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