Re: [rules-dev] Why are fundamental API functions not part of "stable"?
On 21/08/2010 07:54, Wolfgang Laun wrote:
On 20 August 2010 21:56, Mark Proctor<mproctor(a)codehaus.org>
wrote:
> try this:
> for ( KnowledgePackage pkg : kbase.getKnowledgePackages() ) {
> for ( org.drools.definition.rule.Rule rule : pkg.getRules() ) {
> String name = rule.getName();
> org.drools.rule.Rule realRule = ( org.drools.rule.Rule
> ) ((KnowledgePackageImp)pkg).getRule( name );
> System.out.println( realRule.getAgendaGroup() );
> }
> }
Very well hidden - almost an 'Easter egg ;-)
Thank you!
Sorry I probably didn't make myself clear, all of the old api is
available via one way or another. In the above we cast to the
implementation to return a org.drools.definition.rule.Rule that is
castable. In some other cases the old session interface of rulebase
interface is availlable as a field on the ksession impl or kbase impl.
So it's all there, it's just "internal". But yes in general you'll
need
to look at the impl to figure out what can be done where.
Mark
> At the drl it'll stay, for backwards comptability for a
while, we'll map
> it onto the new planned constructs. I have plans over all for something
> more generic, orthogonal and thus more flexible and powerful.
At one time, I thought that the metadata map would be the place where
things are kept. - I canimagine that a selection mechanism based on
s.th. like general property predicates would give you all you've got now
(possibly even with long-term backward compatibility) and plenty of new
possibilities. "Sharing" of rule sets among focusable grouds is one thing
I'd very much like to have...
-W
_______________________________________________
rules-dev mailing list
rules-dev(a)lists.jboss.org
https://lists.jboss.org/mailman/listinfo/rules-dev