One of the benefits of the Rete algorithm is its capability (given that
the network is built properly) of not having to evaluate identical
conditions occuring in several rules individually. This would be
thwarted by blocking the evaluation of all rules except the ones in
the current group.
Also, consider what should happen after one group becomes active,
some rules fire and another group is activated. Should all updates
be ignored for the rules of the first group? So, when this group becomes
active again, all rules have to be evaluated from scratch?
There might be scenarios where the approach for evaluating rules
in batches would be more efficient, but these might be solved using
a completely different approach.
-W
On 09/05/2013, Sonata <plz.write.to(a)gmail.com> wrote:
Thats ok, because if you dont call your function, you dont know if
that
rule
should be fired or not.
But I am concerning about agenda-group, where I know exactly one or more
groups or rules will not be fired but they are still being evaluated.
I ended up moving all the matching/checking to the "then" part for all my
rules to speed things up.
Now my "when" part is super clear
--
View this message in context:
http://drools.46999.n3.nabble.com/Agenda-group-in-fact-insert-time-tp4023...
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