Re: [rules-users] Adding @PropertyReactive causes the other rules removed from the stack
Davide Sottara wrote
The goal of @propertyReactive is exactly to prevent rules from refiring
on a modify, based on what properties are constrained or @watched.
This is irrelevant with respect to the order of the rules.
This may or may not be a bug, we'd need to see the rules.
Davide
Example attached
ReactiveTest.java
package com.sample;
import org.drools.KnowledgeBase;
import org.drools.KnowledgeBaseFactory;
import org.drools.builder.KnowledgeBuilder;
import org.drools.builder.KnowledgeBuilderFactory;
import org.drools.builder.ResourceType;
import org.drools.io.ResourceFactory;
import org.drools.logger.KnowledgeRuntimeLogger;
import org.drools.logger.KnowledgeRuntimeLoggerFactory;
import org.drools.runtime.StatefulKnowledgeSession;
public class ReactiveTest {
public static final void main(String[] args) {
KnowledgeBuilder kbuilder =
KnowledgeBuilderFactory.newKnowledgeBuilder();
kbuilder.add(ResourceFactory.newClassPathResource(Sample.drl),
ResourceType.DRL);
KnowledgeBase kbase = KnowledgeBaseFactory.newKnowledgeBase();
kbase.addKnowledgePackages(kbuilder.getKnowledgePackages());
StatefulKnowledgeSession ksession =
kbase.newStatefulKnowledgeSession();
ksession.insert(new MyClass());
ksession.fireAllRules();
}
}
MyClass.java
package com.sample;
import org.drools.definition.type.PropertyReactive;
@PropertyReactive
public class MyClass {
private String value;
public String getValue() { return value; }
public void setValue(String value) { this.value = value; }
}
Sample.drl
package com.sample;
rule 1
when MyClass(value == null)
then System.out.println(Rule 1 fired);
end
rule 2
when m : MyClass(value == null)
then modify(m) { setData(test) }
end
rule 3
when MyClass(value == null)
then System.out.println(Rule 3 fired);
end
So you can see from the example, rule 1, rule 2 and rule 3 are all
added to the stack ready to be fired.
Rules are then fired starting from the bottom, where rule 3 fired and then
rule 2 fired. But rule 1 is being removed from the stack after rule 2
has fired, which is a mystery to me.
Try to remove modify(m) to just m.setData(test) in rule 2, all three
rules fired.
So I agree to your prevent rules from refiring, but this case is not even
fired yet.
--
View this message in context:
http://drools.46999.n3.nabble.com/Adding-PropertyReactive-causes-the-other-rules-removed-from-the-stack-tp4027374p4027382.html
Sent from the Drools: User forum mailing list archive at Nabble.com.
___
rules-users mailing list
rules-users@lists.jboss.org
https://lists.jboss.org/mailman/listinfo/rules-users
Re: [rules-users] Adding @PropertyReactive causes the other rules removed from the stack
Of course, a workaround is to add @watch(value) to rule 1, but thats really weird. -- View this message in context: http://drools.46999.n3.nabble.com/Adding-PropertyReactive-causes-the-other-rules-removed-from-the-stack-tp4027374p4027383.html Sent from the Drools: User forum mailing list archive at Nabble.com. ___ rules-users mailing list rules-users@lists.jboss.org https://lists.jboss.org/mailman/listinfo/rules-users
Re: [rules-users] Adding @PropertyReactive causes the other rules removed from the stack
I turned your code into a unit test and added it to the 6.0 codebase. see the
method “testModifyAfterInsertWithPropertyReactive” in the commit:
https://github.com/droolsjbpm/drools/commit/53ca46d3b
It works for PHREAK and RETE mode. Everything works for 6.0, we’ll try it
against 5.6.CR1 soon
Mark
On 20 Dec 2013, at 09:14, Sonata plz.write...@gmail.com wrote:
Davide Sottara wrote
The goal of @propertyReactive is exactly to prevent rules from refiring
on a modify, based on what properties are constrained or @watched.
This is irrelevant with respect to the order of the rules.
This may or may not be a bug, we'd need to see the rules.
Davide
Example attached
ReactiveTest.java
package com.sample;
import org.drools.KnowledgeBase;
import org.drools.KnowledgeBaseFactory;
import org.drools.builder.KnowledgeBuilder;
import org.drools.builder.KnowledgeBuilderFactory;
import org.drools.builder.ResourceType;
import org.drools.io.ResourceFactory;
import org.drools.logger.KnowledgeRuntimeLogger;
import org.drools.logger.KnowledgeRuntimeLoggerFactory;
import org.drools.runtime.StatefulKnowledgeSession;
public class ReactiveTest {
public static final void main(String[] args) {
KnowledgeBuilder kbuilder =
KnowledgeBuilderFactory.newKnowledgeBuilder();
kbuilder.add(ResourceFactory.newClassPathResource(Sample.drl),
ResourceType.DRL);
KnowledgeBase kbase = KnowledgeBaseFactory.newKnowledgeBase();
kbase.addKnowledgePackages(kbuilder.getKnowledgePackages());
StatefulKnowledgeSession ksession =
kbase.newStatefulKnowledgeSession();
ksession.insert(new MyClass());
ksession.fireAllRules();
}
}
MyClass.java
package com.sample;
import org.drools.definition.type.PropertyReactive;
@PropertyReactive
public class MyClass {
private String value;
public String getValue() { return value; }
public void setValue(String value) { this.value = value; }
}
Sample.drl
package com.sample;
rule 1
when MyClass(value == null)
then System.out.println(Rule 1 fired);
end
rule 2
when m : MyClass(value == null)
then modify(m) { setData(test) }
end
rule 3
when MyClass(value == null)
then System.out.println(Rule 3 fired);
end
So you can see from the example, rule 1, rule 2 and rule 3 are all
added to the stack ready to be fired.
Rules are then fired starting from the bottom, where rule 3 fired and then
rule 2 fired. But rule 1 is being removed from the stack after rule 2
has fired, which is a mystery to me.
Try to remove modify(m) to just m.setData(test) in rule 2, all three
rules fired.
So I agree to your prevent rules from refiring, but this case is not even
fired yet.
--
View this message in context:
http://drools.46999.n3.nabble.com/Adding-PropertyReactive-causes-the-other-rules-removed-from-the-stack-tp4027374p4027382.html
Sent from the Drools: User forum mailing list archive at Nabble.com.
___
rules-users mailing list
rules-users@lists.jboss.org
https://lists.jboss.org/mailman/listinfo/rules-users
___
rules-users mailing list
rules-users@lists.jboss.org
https://lists.jboss.org/mailman/listinfo/rules-users
Re: [rules-users] Adding @PropertyReactive causes the other rules removed from the stack
small correction to the unit test, which I paste in full below:
@Test
public void testModifyAfterInsertWithPropertyReactive() {
String rule1 =
\n +
package com.sample;\n +
import + MyClass.class.getCanonicalName() + ;\n +
global java.util.List list;\n +
rule r0\n +
then insert( new MyClass() );\n +
end\n +
rule r1 salience 1\n +
when +
MyClass(value == null)\n +
then +
list.add( 1 );\n +
end\n +
\n +
rule r2 salience 2\n +
when +
m : MyClass(value == null)\n +
then +
modify(m) { setData(\test\) }\n +
list.add( 2 );\n +
end\n +
\n +
rule r3 salience 3\n +
when +
MyClass(value == null)\n +
then +
list.add( 3 );\n +
end;
KnowledgeBuilder kbuilder = KnowledgeBuilderFactory.newKnowledgeBuilder();
kbuilder.add( ResourceFactory.newByteArrayResource(rule1.getBytes()),
ResourceType.DRL );
if ( kbuilder.hasErrors() ) {
fail( kbuilder.getErrors().toString() );
}
KnowledgeBase kbase = KnowledgeBaseFactory.newKnowledgeBase();
kbase.addKnowledgePackages( kbuilder.getKnowledgePackages() );
StatefulKnowledgeSession ksession = kbase.newStatefulKnowledgeSession();
List list = new ArrayList();
ksession.setGlobal(list, list);
assertEquals(4, ksession.fireAllRules());
assertEquals(3, list.size());
assertEquals(3, list.get(0));
assertEquals(2, list.get(1));
assertEquals(1, list.get(2));
}
On 20 Dec 2013, at 15:14, Mark Proctor mproc...@codehaus.org wrote:
I turned your code into a unit test and added it to the 6.0 codebase. see the
method “testModifyAfterInsertWithPropertyReactive” in the commit:
https://github.com/droolsjbpm/drools/commit/53ca46d3b
It works for PHREAK and RETE mode. Everything works for 6.0, we’ll try it
against 5.6.CR1 soon
Mark
On 20 Dec 2013, at 09:14, Sonata plz.write...@gmail.com wrote:
Davide Sottara wrote
The goal of @propertyReactive is exactly to prevent rules from refiring
on a modify, based on what properties are constrained or @watched.
This is irrelevant with respect to the order of the rules.
This may or may not be a bug, we'd need to see the rules.
Davide
Example attached
ReactiveTest.java
package com.sample;
import org.drools.KnowledgeBase;
import org.drools.KnowledgeBaseFactory;
import org.drools.builder.KnowledgeBuilder;
import org.drools.builder.KnowledgeBuilderFactory;
import org.drools.builder.ResourceType;
import org.drools.io.ResourceFactory;
import org.drools.logger.KnowledgeRuntimeLogger;
import org.drools.logger.KnowledgeRuntimeLoggerFactory;
import org.drools.runtime.StatefulKnowledgeSession;
public class ReactiveTest {
public static final void main(String[] args) {
KnowledgeBuilder kbuilder =
KnowledgeBuilderFactory.newKnowledgeBuilder();
kbuilder.add(ResourceFactory.newClassPathResource(Sample.drl),
ResourceType.DRL);
KnowledgeBase kbase = KnowledgeBaseFactory.newKnowledgeBase();
kbase.addKnowledgePackages(kbuilder.getKnowledgePackages());
StatefulKnowledgeSession ksession =
kbase.newStatefulKnowledgeSession();
ksession.insert(new MyClass());
ksession.fireAllRules();
}
}
MyClass.java
package com.sample;
import org.drools.definition.type.PropertyReactive;
@PropertyReactive
public class MyClass {
private String value;
public String getValue() { return value; }
public void setValue(String value) { this.value = value; }
}
Sample.drl
package com.sample;
rule 1
when MyClass(value == null)
then System.out.println(Rule 1 fired);
end
rule 2
when m : MyClass(value == null)
then modify(m) { setData(test) }
end
rule 3
when MyClass(value == null)
then System.out.println(Rule 3 fired);
end
So you can see from the example, rule 1, rule 2 and rule 3 are all
added to the stack ready to be fired.
Rules are then fired starting from the bottom, where rule 3 fired and then
rule 2 fired. But rule 1 is being removed from the stack after rule 2
has fired, which is a mystery to me.
Try to remove modify(m) to just m.setData(test) in rule 2, all three
rules fired.
So I agree to your prevent rules from refiring, but this case is not even
fired yet.
--
View this message in context:
http://drools.46999.n3.nabble.com/Adding-PropertyReactive-causes-the-other-rules-removed-from-the-stack-tp4027374p4027382.html
Sent from the Drools: User forum mailing list archive at Nabble.com.
___
rules-users mailing list
rules-users@lists.jboss.org
https://lists.jboss.org/mailman/listinfo/rules-users
[rules-users] how to restore the knowledgeBase when newly created kb has compilationFailed error
Hi there, I am using KnowledgeAgent to get the knowledgeBase and have notifierService started to monitor the rules resource. New knowledgeBase is being created when kAgent finds changes on the rules. However, I don't want to replace the old knowledgebase if newly created knowledgebase has some rules with compilation error. Is this something possible? I am using KnowledgeAgentEventListener to detect the compilation error and using drools 5.3 version. thanks in advance. -- View this message in context: http://drools.46999.n3.nabble.com/how-to-restore-the-knowledgeBase-when-newly-created-kb-has-compilationFailed-error-tp4027387.html Sent from the Drools: User forum mailing list archive at Nabble.com. ___ rules-users mailing list rules-users@lists.jboss.org https://lists.jboss.org/mailman/listinfo/rules-users
Re: [rules-users] how to restore the knowledgeBase when newly created kb has compilationFailed error
There is a newInstance option in the KA configuration that determines whether the changes are applied to the existing KB, or a new one is to be created. Please see the documentation for more details. As a side note, the KA was refactored and improved in 5.5 and 5.6 to fix some bugs and missing features. Let me know if you have more questions Best Davide On 12/20/2013 11:27 AM, bhochhi wrote: Hi there, I am using KnowledgeAgent to get the knowledgeBase and have notifierService started to monitor the rules resource. New knowledgeBase is being created when kAgent finds changes on the rules. However, I don't want to replace the old knowledgebase if newly created knowledgebase has some rules with compilation error. Is this something possible? I am using KnowledgeAgentEventListener to detect the compilation error and using drools 5.3 version. thanks in advance. -- View this message in context: http://drools.46999.n3.nabble.com/how-to-restore-the-knowledgeBase-when-newly-created-kb-has-compilationFailed-error-tp4027387.html Sent from the Drools: User forum mailing list archive at Nabble.com. ___ rules-users mailing list rules-users@lists.jboss.org https://lists.jboss.org/mailman/listinfo/rules-users ___ rules-users mailing list rules-users@lists.jboss.org https://lists.jboss.org/mailman/listinfo/rules-users
Re: [rules-users] how to restore the knowledgeBase when newly created kb has compilationFailed error
Thanks for the reply, so far from documentation, I believe newInstance=true(default) replaces the old kb as a whole and =false will update the existing kb. But I didn't see anywhere explaining that if there is a compilation error on rules, it will not update the kb. And I verified. It updates the kb. I don't know if that is expected or bugs in this version. Anyway, I found the workthrough by assigning the kb to the variable and update it only if there is not resourceCompilationFailed using knowledgeBaseUpdated event on KnowledgeAgentEventListener. Anyway, thanks for your help. -- View this message in context: http://drools.46999.n3.nabble.com/how-to-restore-the-knowledgeBase-when-newly-created-kb-has-compilationFailed-error-tp4027387p4027389.html Sent from the Drools: User forum mailing list archive at Nabble.com. ___ rules-users mailing list rules-users@lists.jboss.org https://lists.jboss.org/mailman/listinfo/rules-users
Re: [rules-users] how to restore the knowledgeBase when newly created kb has compilationFailed error
That feature was definitely not available in 5.3... I would have to check in 5.6, and I agree that it would be a nice feature to have. I'll see if we are still in time to add it. Davide On 12/20/2013 12:35 PM, bhochhi wrote: Thanks for the reply, so far from documentation, I believe newInstance=true(default) replaces the old kb as a whole and =false will update the existing kb. But I didn't see anywhere explaining that if there is a compilation error on rules, it will not update the kb. And I verified. It updates the kb. I don't know if that is expected or bugs in this version. Anyway, I found the workthrough by assigning the kb to the variable and update it only if there is not resourceCompilationFailed using knowledgeBaseUpdated event on KnowledgeAgentEventListener. Anyway, thanks for your help. -- View this message in context: http://drools.46999.n3.nabble.com/how-to-restore-the-knowledgeBase-when-newly-created-kb-has-compilationFailed-error-tp4027387p4027389.html Sent from the Drools: User forum mailing list archive at Nabble.com. ___ rules-users mailing list rules-users@lists.jboss.org https://lists.jboss.org/mailman/listinfo/rules-users ___ rules-users mailing list rules-users@lists.jboss.org https://lists.jboss.org/mailman/listinfo/rules-users
