Barry, I did not run the code bellow, but drools uses the temporal distance algorithm to calculate how much time it has to wait on a "delaying not" before activating the rule. So, the first question I have is what are you using as timestamp for the events? the time they are inserted into the engine or a timestamp attribute?
[]s Edson 2009/8/30 Barry Kaplan <[email protected]> > > After looking at ../integrationtests/test_CEP_DelayingNot.drl I'm getting > closer, but not really understanding. I can get a greenbar with the below. > However if change the last advanceTime() value to anything less than 30 the > test fails. So it seems that the the conditional not clause does not start > its window until the first two clauses are true. > > Why would this be? Why does it not start ticking when the first StartEvent > is inserted? > > ---- > rule "[every s->f&&!a] How do I use patterns to correlate events arriving > in-order or out-of-order?" > when > $se : StartEvent($exchangeId : exchangeId) > from entry-point "stream" > > $fe : FinishedEvent(exchangeId == $exchangeId, this after[0s,30s] > $se) > from entry-point "stream" > > not (AbortedEvent(exchangeId == $exchangeId, this after[0s,30s] $se) > from entry-point "stream") > then > results.put("startEvent", $se); > results.put("finishedEvent", $fe); > end > ---- > @Test > def void > > How_do_I_use_patterns_to_correlate_events_arriving_in_order_or_out_of_order__B() > { > insert new StartEvent(id: "se1", exchangeId: "BBB") > advanceTime 10, SECONDS > fireAllRules() > assert results.isEmpty() > > advanceTime 10, SECONDS > insert new FinishedEvent(id: "fe1", exchangeId: "BBB") > fireAllRules() > assert results.isEmpty() > > advanceTime 30, SECONDS > assert results["startEvent"].id == "se1" > assert results["finishedEvent"].id == "fe1" > } > ---- > -- > View this message in context: > http://www.nabble.com/Conditional-%27not%27-invalid-for-event-streams--tp25217095p25217673.html > Sent from the drools - user mailing list archive at Nabble.com. > > _______________________________________________ > rules-users mailing list > [email protected] > https://lists.jboss.org/mailman/listinfo/rules-users > -- Edson Tirelli JBoss Drools Core Development JBoss by Red Hat @ www.jboss.com
_______________________________________________ rules-users mailing list [email protected] https://lists.jboss.org/mailman/listinfo/rules-users
