eval() changes %# to caller

MUX does the same thing. It's either deliberate or a bug that's been around for over a decade.

Quoting Penndev (pennmush-bugs@pennmush.org):
> #7401: eval() changes %# to caller
> ---------------------+------------------------------------------------------
>   Reporter:  Talvo   |       Owner:     
>       Type:  bug     |      Status:  new
>   Priority:  minor   |   Milestone:     
> Resolution:          |    Keywords:     
> Visibility:  Public  |  
> ---------------------+------------------------------------------------------
> Comment (by raevnos):
> 
>  MUX does the same thing. It's either deliberate or a bug that's been
>  around for over a decade.

Or a little of neither. It's a historical artifact that's
been lost from Penn's help file. The TM 2.2.4 help for
GET_EVAL() notes this venerable behavior:

  The get_eval function returns the specified attribute from the named object
  (just like the get function), except that function references and
  %-substitutions have already been performed.  In function references,
  'me' refers to the object being looked at, and %-substitutions that refer
  to the enactor (such as %n, %#, etc) refer to the object making the get_eval
  call.  If the attribute is not present on <object>, its parent is searched
  for the attribute.

People using get_eval/eval relied on this behavior in some cases. But
people soon wanted the other behavior in many cases. As a result, U()
was born -- in part, to have an eval-like function that maintained %#
as the original enactor.

I recommend updating Penn's help file to provide better documentation.

-- 
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Javelin@M*U*S*H, once Paul@DuneMUSH | Alan Schwartz <dunemush@pennmush.org>
   (mush.pennmush.org 4201)         |     
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-

I'm not sure if I should be scared or impressed that you have Tiny 2.2 sitting around.

Help files mention %# change now.