This is odd. According to https://code.jsoftware.com/wiki/Vocabulary/atdot f should have rank 0 (since v is * here).
But, when I redefine f as f=: 1:`(] * [: $:&([echo) <:)@.* I see that when (f 2 3 5 7 11) is executed, f is being treated as some other rank. This does not occur for (f 2 3 5 7). Using $:@<: works around this issue by using $: with the rank of <: (the rank problem with @. would still occur, without triggering a rank error). Also, the j implementation tells me that f has infinite rank - despite what nuvoc says and despite the behavior for (f 2 3 5 7): f b.0 _ _ _ -- Raul On Thu, May 25, 2023 at 11:12 AM Martin Kreuzer <i...@airkreuzer.com> wrote: > > Dear all - > > Thought I had succeeded in deducing step by step the tacit verb for factorial, > > f=: 1:`(] * [: $: <:)@.* > > when I encountered this: > > f=. 1:`(] * [: $: <:)@.* > f 2 3 5 7 > 2 6 120 5040 > f 2 3 5 7 11 > |rank error in f, executing monad $: > | f 2 3 5 7 11 > > In contrast this seems to work alright: > > g=. 1:`(] * $: @ <:)@.* NB. dictionary entry > g 2 3 5 7 > 2 6 120 5040 > g 2 3 5 7 11 > 2 6 120 5040 39916800 > > JVERSION > Engine: j9.4.2/j64avx2/windows > Build: commercial/2023-04-10T01:19:53/clang-15-0-7/SLEEF=1 > Library: 9.4.21 > Platform: Win 64 > Installer: j9.4 install > InstallPath: d:/program files (non-sys)/j9.4 > Contact: www.jsoftware.com > > However, my verb > > f=: 1:`(] * [: $: <:)@.* > f 2 3 5 7 11 13 > 2 6 120 5040 39916800 6227020800 > > works under v8.07 as expected .... > > Could somebody shed some light on this ? > > -M > > ---------------------------------------------------------------------- > For information about J forums see http://www.jsoftware.com/forums.htm ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
