** warning: This is a rant, but it just might save you some head scratching if you run into these issues ** I really hate writing rules. Not, well not rules, but rules in the DRL language. There are so many edge conditions that fail silently. Here just two from today: ... then retract(f1) // some comment retract(f2) end In the above retract(f2) will not be invoke. No message about any problems -- it just doesn't happen. Remove the comment and f2 is retracted. And, where 'ksession' is a global ... then update(ksession.getFactHandle(f1), new MyF1(10, 10)) end In this one the constructor used for MyF1 does not exists. No message, just silent failure. Change the rule to: ... then MyF1 fi_new = MyF1(10, 10) update(ksession.getFactHandle(f1), f1_new) end And now the error message about the missing constructor is produced. -- View this message in context: http://n3.nabble.com/drl-s-and-silent-failures-tp704973p704973.html Sent from the Drools - User mailing list archive at Nabble.com. _______________________________________________ rules-users mailing list rules-users(a)lists.jboss.org https://lists.jboss.org/mailman/listinfo/rules-users

No jira yet, my brain was too frazzled late last night. Dialect was mvel against 5.0.1. Also ran into a few internal exceptions wrt to window:length(1) and doing an explicit retract of an event from that window. In 5.0.1 it yielded an NPE, in 5.1.M1 it yielded an internal error (can't build trunk). Should be able to reproduce easily, but I want to try to figured these out myself as part of learning the internals a bit more. -- View this message in context: http://n3.nabble.com/drl-s-and-silent-failures-tp704973p706069.html Sent from the Drools - User mailing list archive at Nabble.com. _______________________________________________ rules-users mailing list rules-users(a)lists.jboss.org https://lists.jboss.org/mailman/listinfo/rules-users