Davide is correct, for our analysis to work, the constraint must not be inside
an eval, and any constraint that does not start with a field name is internally
rewritten as an eval.
Mark
On 11 Jun 2014, at 18:48, Davide Sottara <dso...@gmail.com> wrote:
> I think I know what is happening here.
> I assume your supportFT class is @propertyReactive.
> (btw, you should follow bean conventions and capitalize class names)
> Looking at Rule 1, you don't set the value directly, but you do it through
> a modify. Your rule 2, which checks for the value, uses the function
> "containsAny"
> rather than the field direclty as in "value IN (..)".
> At the moment, the engine has no way to realize that the function
> involves the field "value" - I'm not even sure it is possible in general -
> Property reactivity will ignore the update since, from its perspective, none
> of the fields relevant to the rule has been affected, hence rule 2 will
> not hit.
> If you use eval in the constraint, property reactivity is disabled, so
> the rule
> WILL fire the first time, but now you are vulnerable to infinite loops,
> as if
> you did not have propertyreactive.
> You may have to use the @watch() annotation explicitly to control which
> modifies will cause reevaluations and which ones won't (see the manual
> for this)
> Adding "eval" everywhere is not a good idea.
> Best,
> Davide
>
>
>
>
>
> On 06/11/2014 01:53 PM, brachi wrote:
>> example of rule that doesn't work without eval:
>>
>> /* rule "1"
>> salience -1
>> agenda-group "agenda1"
>> when
>> $conclusion: supportFT()
>> then
>> if($conclusion.getValue()==null){
>> modify($conclusion) { setValue(new ArrayList())};
>> }
>> $conclusion.getValue().add("supportedValue");
>> modify($conclusion) { setValue($conclusion.getValue()) };
>> end
>>
>>
>> rule "2"
>> salience -2
>> agenda-group "agenda2"
>> when
>> supportFT(Operators.containsAny(value,new
>> String[]{"supportedValue","otherValue"}))
>> $conclusion: ConclusionFt()
>> then
>> modify($conclusion){setValue("success")};
>> end*/
>>
>> rule "2" doesn't hit, works only with eval in rule "2".
>> in this example I can add eval, because the Fact Types model is different,
>> so the rule doesn't reevaluated.
>> because I had this case, I decided to add eval for all constraints, but
>> unfortunately I had an infinite loop.
>>
>>
>>
>>
>> --
>> View this message in context:
>> http://drools.46999.n3.nabble.com/Adding-eval-on-rule-cause-it-to-run-in-infinite-loop-tp4029966p4029984.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
_______________________________________________
rules-users mailing list
rules-users@lists.jboss.org
https://lists.jboss.org/mailman/listinfo/rules-users
