I upgraded from 5.5.0.Final of Drools (rule engine) to 5.6.0.Final and I am now seeing errors with globals. We have rule constraints that access globals in a DRL; like this: some.classpath.package2.MyType ( global_instance.invoke(this) == true ) I have `some.classpath.package.SomeType global_instance;` declared at the top of the DRL, but it doesn't seem to be interpeting it as a global in the rule's restriction. So, putting this into context within, I have a DRL such as: package my.package.example; global some.classpath.package.SomeType global_instance; **** rule "Example rule" when some.classpath.package2.MyType ( global_instance.invoke(this) == true ) then // do something end **** I am getting an error from the package builder originating from the "Example rule" specifying something like: "can't find method MyType.global_instance()" Once, again this did work in 5.5.0.Final, but in 5.6.0.Final I am getting a failure trying to build the knowledge package. It looks to me like the `global_instance` symbol is not being recognized as a global in the DRL and is trying to be resolved as a method of the MyType class. So I guess it would be trying to do a "MyType.global_instance().invoke(this)" or something. The intent is that `global_instance` will be set for a session with an immutable function-type of class (invoke is an instance method though; not static). The rule restriction uses this function/predicate to determine if some complex criteria is true for `this` MyType instance. I would appreciate any insight into what the issue might be. I haven't been able to find any threads out there on a similar issue. -- View this message in context: http://drools.46999.n3.nabble.com/Incorrect-resolution-of-global-symbols-... Sent from the Drools: User forum mailing list archive at Nabble.com. _______________________________________________ rules-users mailing list rules-users(a)lists.jboss.org https://lists.jboss.org/mailman/listinfo/rules-users

Thank you for your reply. That was helpful. I did notice that the workaround you proposed of only calling the `some.classpath.package2.MyType ( global_instance.invoke(this) )` was still not analyzing the constraint correctly. I think this was because #invoke() was returning an java.lang.Object type (which would be a java.lang.Boolean at runtime). This led us to the discovery that if we just flipped the order of the constraint expression to: `some.classpath.package2.MyType ( true == global_instance.invoke(this) )` it works fine. It looks like with beginning the expression with `true ==` instead of having it `== true` after the method call, was able to be analyzed correctly. I posted a Jira for this @ https://issues.jboss.org/browse/DROOLS-420 . I haven't logged a Drools Jira before, so hopefully it is filled out as expected/desired. -- View this message in context: http://drools.46999.n3.nabble.com/Incorrect-resolution-of-global-symbols-... Sent from the Drools: User forum mailing list archive at Nabble.com. _______________________________________________ rules-users mailing list rules-users(a)lists.jboss.org https://lists.jboss.org/mailman/listinfo/rules-users