[rules-users] Inconsistent behavior between LHS and RHS ...

[Wolfgang Laun]

The basic behaviour of the engine is to re-evaluate all rules where a
new, updated or deleted fact is referenced via its (sub)class or some
interface name. If the LHS is true, the rule is fired.

To insert a complex fact requiring setter calls, make sure to do the
insert() after all the settings.

There are some variations due to rule attributes, see the docs.

-W


On Tue, Nov 10, 2009 at 1:37 PM, Javier Vera Gomez
<Javier.Vera at andorratelecom.ad> wrote:
> Man... you made my day!
>
> Such a Dumb! Thats true, first is inserted the Task and after that the Param is added. Im wondering if I update the fact, all the rules fired before updating it would fire again? Or just the rules which conditions are changed?
>
> I guess:
>
> R1
> When
>   T
>   T.X
> Then
>   ...
>
> R2
> When
>   T
>   T.Y
> Then
>   ...
>
> R3
> When
>   T
> Then
>   ...
>
> If T is asserted but not T.X neither T.Y, R3 is fired. If after this, T.X is asserted (I mean updating T), I guess R3 will not be fired again... and if T is updated again but retracting T.X, then R1 should perform rollback of the RHS clausure... is this the actual behavior?
>
> -----Mensaje original-----
> De: rules-users-bounces at lists.jboss.org [mailto:rules-users-bounces at lists.jboss.org] En nombre de Wolfgang Laun
> Enviado el: dimarts, 10 / novembre / 2009 12:41
> Para: Rules Users List
> Asunto: Re: [rules-users] Inconsistent behavior between LHS and RHS ...
>
> There are a couple of scenarios that might produce this effect:
>
> One is:
>  // session is running !
>  Task t = new Task();
>  session.insert ( t );
>  t.getTechnicalParams().add(...)
>
> Can you exclude this or a similar scenario, where Parameters are added
> to an existing Task() fact, without update() of the fact?
>
> -W
>
>
> On Tue, Nov 10, 2009 at 9:22 AM, Javier Vera Gomez
> <Javier.Vera at andorratelecom.ad> wrote:
>> Thanks for answering,
>>
>> Further detail:
>> The actual output is the rule is fired (ie, the param is not found) and then the param is printed out. So, in fact, the param does exists.
>>
>> The Task is created as usual, nothing special here (insert (new Task...)
>>
>> Code fragment from Task.java:
>>
>>    public String getParameter(String name){
>>        for (Parameter param : getTechnicalParams()) {
>>                        if( name.equals(param.getName()) ) {
>>                                return param.getValue();
>>                        }
>>                }
>>        return null;
>>    }
>>
>> @XmlElement(name="parameter",nillable = true)
>>    @XmlElementWrapper(name="technicalParams")
>> protected List<Parameter> technicalParams;
>>
>>   public List<Parameter> getTechnicalParams() {
>>        if (technicalParams == null) {
>>            technicalParams = new ArrayList<Parameter>();
>>        }
>>        return this.technicalParams;
>>    }
>>
>>
>> It is important to note that in other flows of similar rules this is working perfect... I know I have not giving so much input but I do not know what else could provide.
>>
>> Any clues ?
>>
>> -----Mensaje original-----
>> De: rules-users-bounces at lists.jboss.org [mailto:rules-users-bounces at lists.jboss.org] En nombre de Wolfgang Laun
>> Enviado el: dilluns, 9 / novembre / 2009 13:56
>> Para: Rules Users List
>> Asunto: Re: [rules-users] Inconsistent behavior between LHS and RHS ...
>>
>> It's difficult to assess this case without the code from Task.java,
>> especially mehtods getParameter(), getTechnicalParams(); the code used
>> to create the Task fact, and the actual output.
>> -W
>>
>> 2009/11/9 Javier Vera Gomez <Javier.Vera at andorratelecom.ad>:
>>> Hi,
>>>
>>> I've been using drools 4.07 for a while. I have found a weird case. It seems
>>> that when evaluating the LHS some fact is true but the RHS probes it wasn't
>>> actually true.
>>>
>>> The example code  is at the end. It is very weird to me I have no clue about
>>> how to solve this or what could be happening!
>>>
>>> rule "A-Rule"
>>>         no-loop true
>>>         ruleflow-group "validate-task"
>>>         salience -9999999
>>>         when
>>>                 >$task : Task( type == "Creacio_Profile_Radius" )
>>>                 >eval($task.getParameter("PRIMARYDNS")==null)
>>>         then
>>>                 >System.out.println(">>>DBGPar0:
>>> "+$task.getTechnicalParams().get(0).getName());
>>>                 >System.out.println(">>>DBGPar1:
>>> "+$task.getTechnicalParams().get(1).getName());
>>>                 >System.out.println("Getparam PRIMARYDNS:
>>> "+$task.getParameter("PRIMARYDNS"));
>>> end
>>>
>>>
>>> further explanation:
>>>
>>> The Task class has a List<Parameter> technicalParams, originally the eval
>>> part was:
>>> not(Parameter (name=="PRIMARYDNS") from $task.technicalParams())
>>>
>>> it didn't work and I coded this "getParameter" method just to make some
>>> tests. So, it seems the collection appears "empty" to the LHS but in fact it
>>> isnt!
>>>
>>> any clue?
>>>
>>> thanks a lot,
>>>
>>> Javier
>>>
>>> _______________________________________________
>>> rules-users mailing list
>>> rules-users at lists.jboss.org
>>> https://lists.jboss.org/mailman/listinfo/rules-users
>>>
>>>
>>
>> _______________________________________________
>> rules-users mailing list
>> rules-users at lists.jboss.org
>> https://lists.jboss.org/mailman/listinfo/rules-users
>>
>>
>> _______________________________________________
>> rules-users mailing list
>> rules-users at lists.jboss.org
>> https://lists.jboss.org/mailman/listinfo/rules-users
>>
>
> _______________________________________________
> rules-users mailing list
> rules-users at lists.jboss.org
> https://lists.jboss.org/mailman/listinfo/rules-users
>
>
> _______________________________________________
> rules-users mailing list
> rules-users at lists.jboss.org
> https://lists.jboss.org/mailman/listinfo/rules-users
>