[rules-users] Question about the Nature of Ruleflows

[Wolfgang Laun]

On 21/08/2012, BenjaminWolfe <benjamin.e.wolfe at gmail.com> wrote:
> I'm trying to wrap my mind around the nature of ruleflows -- and I think
> this
> question sums up a bit of my confusion.
>
> I have a stateful knowledge session.  The stateful knowledge session has a
> ruleflow associated with it.  That ruleflow diverges into two branches --
> branch A and branch B -- based on a rule constraint.  It's an XOR, and the
> mvel rule constraints are something like this:
>
> to branch A
> $u : User( $i : userId )
> exists ( Dog( owner == $i, hasDogTag == false) )
>
> to branch B
> $u : User( $i : userId )
> not ( Dog( owner == $i, hasDogTag == false) )
>
> Then the two branches converge again.  I insert two facts: user 1, whose
> dogs all have dog tags, and user 2 whose third dog is missing a tag.  Then,
> after I insert both facts, I fire all rules.

If you don't insert any Dogs, only the condition for branch B is true,
since there no Dogs at all, tag or no tag.

If you insert Dogs according to your statement, both conditions are
true, A for user 2 and B for user 1 (which is not what you describe
below, but I think this is an oversight).

Anyway, a "flow" will not cycle automatically if there are several
fact groups, with some of them requiring this or that processing. To
notify Users with untagged dogs, a simple rule with the conditions
you've posted and with a consequence notifying the user would be
sufficient.

For an overall decision between "there are tagless dogs" and "all dogs
are tagged" you'd need conditions like these:

exists( Dog( hasDogTag == false) )

not( Dog( hasDogTag == false) )

-W


>
> Should the flow be like this?
>
> user 1: start --> diverge --> branch A (tell him to buy dog tags) -->
> converge --> end
> user 2: start --> diverge --> branch B (leave him alone) --> converge -->
> end
>
> Or does a ruleflow look at the whole session rather than each fact -- in
> which case either both users would pass through branch A, or both would
> pass
> through branch B?
>
> If I want the first case to happen, is there something I'm missing in the
> implementation?  And are there any more details I can add to make my post
> more specific?  Thank you for your help.
>
>
>
> --
> View this message in context:
> http://drools.46999.n3.nabble.com/Question-about-the-Nature-of-Ruleflows-tp4019292.html
> Sent from the Drools: User forum mailing list archive at Nabble.com.
> _______________________________________________
> rules-users mailing list
> rules-users at lists.jboss.org
> https://lists.jboss.org/mailman/listinfo/rules-users
>