Re: [SERVER, SECURITY] bug in set_task_perms() ?

• Date: Mon, 25 Sep 1995 14:34:04 PDT
• From:
  Rich Connamacher <phantom@baymoo.sfsu.edu>
• cc: moo-cows.PARC@xerox.com
• Content-Type: TEXT/PLAIN; charset=US-ASCII
• In-Reply-to: <199509252106.RAA10059@kanga.INS.CWRU.Edu>

> > Am I missing something?  Why would you need to build the huge stack in
> > callers() just to test top level-ness?  In all of my verbs, to test top
> > level-ness, I just make the quick and tick-friendly check 'if
> > (valid(caller))'.  A wiz can fool caller_perms(), but can't fool the
> > variable 'caller'.
> 
> Uh, how can this possibly work?  For command-line verb calls,
> caller==player, which means that caller is valid.  On the other hand, if
> it's called from another verb, it's also valid.  caller is !valid only
> when referenced within eval(), which is a very limited case indeed.

Um... all I can say is... whoops

Phantom

• Re: [SERVER, SECURITY] bug in set_task_perms() ?
• From: sir@po.cwru.edu (Seth I. Rich)

• Prev: Re: [SERVER, SECURITY] bug in set_task_perms() ?
• Next: Re: [SERVER] ALPHA release
• Index: MOO-cows
• Thread: MOO-cows