RE: Taxonomic reasoning

Ok, here's what happens when I run this example, using clisp:

CL-USER> (in-package :lisa-user)
#<PACKAGE LISA-USER>
LISA-USER> (consider-taxonomy)
T
LISA-USER> (setf (consider-taxonomy) nil)
NIL
LISA-USER> (consider-taxonomy)
NIL
LISA-USER> (defclass mysuper () ())
#<STANDARD-CLASS MYSUPER>
LISA-USER> (defclass mysub (mysuper) ())
#<STANDARD-CLASS MYSUB>
LISA-USER> (defrule match-mysuper () (mysuper) => (format t
"match-mysuper~%"))
#<RULE MATCH-MYSUPER>
LISA-USER> (defrule match-mysub () (mysub) => (format t "match-mysub~%"))
#<RULE MATCH-MYSUB>
LISA-USER> (reset)
T
LISA-USER> (assert ((make-instance 'my-super)))
; Evaluation aborted
LISA-USER> (assert ((make-instance 'mysuper)))
#<FACT F-1, MYSUPER, CF is 0.000>
LISA-USER> (assert ((make-instance 'mysub)))
#<FACT F-2, MYSUB, CF is 0.000>
LISA-USER> (run)
match-mysuper
match-mysub
2
LISA-USER> 

-----Original Message-----
From: Young, David [mailto:[email protected]]
Sent: Friday, September 09, 2005 12:40 PM
To: 'Paolo Amoroso'; LISA
Subject: RE: [Lisa-users] Taxonomic reasoning


Just glancing at this code, it should do as you (and I) expect. There
shouldn't be any surprises here. I'll take a look when I have a moment...

dey

-----Original Message-----
From: Paolo Amoroso [mailto:[email protected]]
Sent: Friday, September 09, 2005 12:25 PM
To: LISA
Subject: [Lisa-users] Taxonomic reasoning


I am trying to understand how taxonomic reasoning works.  Consider the
following class definitions:

(in-package :lisa-user)

(defclass mysuper ()
  ())

(defclass mysub (mysuper)
  ())

I would like to write a rule that matches only mysuper facts, and a
different rule only for mysub facts, i.e. I don't want the rule for
mysuper to also match mysub:

(defrule match-mysuper ()
  (mysuper)
  =>
  (format t "~&Matched MYSUPER~%"))

(defrule match-mysub ()
  (mysub)
  =>
  (format t "~&Matched MYSUB~%"))

If I understand correctly, this should do what I want if I set
consider-taxonomy to nil.  But LISA works in a different way.  Here is
what happens with the default of t for consider-taxonomy (see my other
message):

LISA-USER> (reset)
T
LISA-USER> (assert (mysuper))
#<FACT F-1, MYSUPER, CF is 0.000>
LISA-USER> (assert (mysub))
#<FACT F-2, MYSUB, CF is 0.000>
LISA-USER> (run)
Matched MYSUPER
Matched MYSUPER
Matched MYSUB
3

And here is what happens when I toggle consider-taxonomy:

LISA-USER> (reset)
T
LISA-USER> (setf (consider-taxonomy) nil)
NIL
LISA-USER> (assert (mysuper))
#<FACT F-1, MYSUPER, CF is 0.000>
LISA-USER> (assert (mysub))
#<FACT F-2, MYSUB, CF is 0.000>
LISA-USER> (run)
Matched MYSUPER
Matched MYSUPER
Matched MYSUB
3
LISA-USER> 

Setting consider-taxonomy has apparently non effect.  Is this the
expected behavior?  Are my rules correct?  If not, do I have to
rewrite match-mysuper like this?  It feels quite kludgey.

(defrule match-mysuper ()
  (mysuper)
  (not (mysub))
  =>
  (format t "~&Matched MYSUPER~%"))

I am using CMUCL Snapshot 2005-09 (19B) under Linux with the latest
LISA CVS sources.


Paolo
-- 
Lisp Propulsion Laboratory log - http://www.paoloamoroso.it/log


-------------------------------------------------------
SF.Net email is Sponsored by the Better Software Conference & EXPO
September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices
Agile & Plan-Driven Development * Managing Projects & Teams * Testing & QA
Security * Process Improvement & Measurement * http://www.sqe.com/bsce5sf
_______________________________________________
Lisa-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/lisa-users

This email message is for the sole use of the intended recipients(s) and may
contain confidential and privileged information of Bloodhound Software,
Inc.. Any unauthorized review, use, disclosure is prohibited. If you are not
the intended recipient, please contact the sender by reply email and destroy
all copies of the original message.


-------------------------------------------------------
SF.Net email is Sponsored by the Better Software Conference & EXPO
September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices
Agile & Plan-Driven Development * Managing Projects & Teams * Testing & QA
Security * Process Improvement & Measurement * http://www.sqe.com/bsce5sf
_______________________________________________
Lisa-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/lisa-users

This email message is for the sole use of the intended recipients(s) and may contain confidential and privileged information of Bloodhound Software, Inc.. Any unauthorized review, use, disclosure is prohibited. If you are not the intended recipient, please contact the sender by reply email and destroy all copies of the original message.


-------------------------------------------------------
SF.Net email is Sponsored by the Better Software Conference & EXPO
September 19-22, 2005 * San Francisco, CA * Development Lifecycle Practices
Agile & Plan-Driven Development * Managing Projects & Teams * Testing & QA
Security * Process Improvement & Measurement * http://www.sqe.com/bsce5sf