I think Hotmail wrote:
[Charset iso-8859-1 unsupported, filtering to ASCII...]
> Hi all,
>
> I encountered the following problems when using backward chaining. Maybe I
> did something wrong?
> If the problems are real, are there better workarounds?
Thanks very much for the feedback. You're probing some of the dark
corners of backwards-chaining behaviour; see comments below.
>
> Assuming:
>
> (do-backward-chaining foo)
>
> 1) 'not' CE implies 'explicit':
> Example:
> (not (foo)) interpreted as: (not (explicit (foo)), that is no 'need-foo'
> fact is generated
>
That's right. Jess behaves this way mostly because it never occurred
to me that it could behave any other way; it could, though, I
see. Does anybody else have an opinion one way or the other?
> It might seem ok to assume 'not need-foo' means: "I don't need foo", but
> what I really need
> here is a way to check that the need-foo rule does not fire.
>
> (Workaround: another rule without the 'not' CE, that does initialization of
> need-foo)
>
I guess I don't understand what it is you want to do here -- I'd be
interested in details.
>
> 2) references passed corrupted
> Example:
>
> (defrule r1
> ?rf <-(ref-foo)
> (foo ?rf)
> ...
>
> (defrule need-foo
> (need-foo ?rf) ;rf here is corrupt
> ?rf <-(ref-foo) ;does not succeed
>
> (Workaround: match on 'id' fields instead on references - like Jess 5.x)
>
This one is a very interesting bug. Basically what's happening is that
the object ?rf that gets passed isn't the Fact reference, but a kind
of a handle to a fact. THat handle is only valid within the context of
the activation that generated it. It's erroneously being passed out of
that context into the need-foo fact. I will see about fixing this.
> 3) multi-fields passed corrupted
> (deftemplate foo (multislot mf))
>
> (defrule r1
> (ref-foo (mf $?mf))
> (foo (mf $?mf))
> ...
>
> (defrule need-foo
> (need-foo $? ?x $?) ;?x gets bound to the whole multi-field !?
>
> What happens here that the source multi-field is bound to a single
> variable within the target multi-field.
>
> (Workaround: assert helper fact with multi-field and match in another rule)
This one is quite subtle, and I'm gladd you brought it to my
attention. I will fix this for the next release.
>
>
> Thanks in advance for your help,
> Marc.
>
> ---------------------------------------------------------------------
> To unsubscribe, send the words 'unsubscribe jess-users [EMAIL PROTECTED]'
> in the BODY of a message to [EMAIL PROTECTED], NOT to the
> list (use your own address!) List problems? Notify [EMAIL PROTECTED]
> ---------------------------------------------------------------------
>
---------------------------------------------------------
Ernest Friedman-Hill
Distributed Systems Research Phone: (925) 294-2154
Sandia National Labs FAX: (925) 294-2234
Org. 8920, MS 9012 [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 (use your own address!) List problems? Notify [EMAIL PROTECTED]
---------------------------------------------------------------------
