I think Lars Rasmusson wrote:
>
> From the README file:
>
> Note that a not pattern cannot contain any variables that are not bound
> before that pattern...
>
> Wouldn't it be useful to allow named variables in the not pattern, as
> it is in clips? For instance
>
> (a ?x)
> (not (b ?y ?y&~?x))
This would actually work in the engine, but the parser is
overzealously vetoing it. I believe I could change it to allow this
without too much trouble. Look for this in a future release.
>
> Could the fact that 'not' is not completely implemented be the reason
> that jess could outperform CLIPS when it was jit-compiled?
No, not really. The syntax accepted by the front end bears little
relationship to what the engine itself executes. Basically if there's
a test written in a rule (either in Jess or in CLIPS) it is executed;
if there isn't it's not; so the complexity of expressions accepted by
the front end doesn't affect the speed of the engine. This kind of
thing is really a deficiency in my parser and network builder, all of
which run before I start my timer, not in the engine.
---------------------------------------------------------
Ernest Friedman-Hill
Distributed Systems Research Phone: (510) 294-2154
Sandia National Labs FAX: (510) 294-2234
Org. 8920, MS 9214 [EMAIL PROTECTED]
PO Box 969 http://herzberg.ca.sandia.gov
Livermore, CA 94550
---------------------------------------------------------------------
To unsubscribe, send the words 'unsubscribe jess-users [EMAIL PROTECTED]'
in the BODY of a message to [EMAIL PROTECTED], NOT to the
list. List problems? Notify [EMAIL PROTECTED]
---------------------------------------------------------------------
