current R CMD check failures: argument "what" is missing, with no default about spades.core HOT 7 CLOSED

This was an old issue. inherits doesn't work on simList. It was LONG ago changed to class( ). I am undoing the portion of the commit that made this 0129288 . It must stay class or we need to diagnose why inherits fails and fix that. For now, class is fine.

EDIT: I am incorrect on this diagnosis. There was something that happened a while ago with this, but I am mis-remembering the details. This error may simply be a case of correcting the use of inherits, which takes 2 arguments. Working on it...

from spades.core.

It may have been a speed issue, but I think both are "fast enough"

microbenchmark::microbenchmark(inherits(sim, "simList"), class(sim)== "simList", is(sim, "simList"), times = 1e4)
Unit: nanoseconds
                     expr  min   lq      mean median   uq   max neval cld
 inherits(sim, "simList")  880 1174 1778.6009   1467 1760 23462 10000  b 
  class(sim) == "simList"  293  587  967.7314    880  880 43698 10000 a  
       is(sim, "simList") 4986 6159 7612.0034   6452 6746 50443 10000   c

from spades.core.

I've fixed it in my branch.

Note we use inherits(sim, "simList") elsewhere with no issue.

from spades.core.

if it's a speed issue, should we change all instances of inherits(sim, "simList") to class(sim) == "simList"?

from spades.core.

I think the difference is only relevant if the function is being run EVERY event. Even then, probably fine. I believe inherits(...) is "more correct", but I don't really know the strong arguments.

from spades.core.

See discussion here: https://stackoverflow.com/questions/27923345/whats-the-difference-between-is-and-inherits
(is vs. inherits). They note the integer difference ... I will look for discussions about class vs inherits

from spades.core.

Yes, I believe inherits is more correct.

But since it's in the doEvent, I'll make the change based on speed.

from spades.core.