Interesting. It would probably help if the docs made that clear. :)
I’m off to get myself some sleep just now, but I’ll do a push of the repo to GitHub
tomorrow and point you at the tests.
A summary of the behaviour is that I was seeing multiple activations of a rule, which went
away when I added “lock-on-active” to that rule. However, that didn’t prevent other rules
from activating afterwards. Unfortunately in a separate test, which added some more rules
which would activate first, that rule ceased to activate at all. Despite the fact that it
was the only rule with “lock-on-active” and no rules had an "agenda-group"
attribute.
This is using 5.5. I’ll probably be upgrading to the 5.6 CR tomorrow or at the weekend, so
I should be able to confirm what happens there.
I came across it because I was experimenting with mechanisms to ensure that a rule only
activates once. It’s something that I find quite useful in a stateless session.
Steve
On 7 Nov 2013, at 22:29, Davide Sottara <dsotty(a)gmail.com> wrote:
Lock-on-active was very recently the subject of a bad bug,
DROOLS-281,
which has been fixed a few days ago.
This said, all rules that do not have an explicit group set end up in
the "MAIN" (or "DEFAULT", I don't remember)
agenda group and then behave accordingly.
Could you post the Drools version number and some more details on the
example and the "unexpected" behavior?
Thanks
Davide
On 11/07/2013 03:01 PM, Stephen Masters wrote:
> Hi folks,
>
> According to the user guide, lock-on-active “inhibits additional activations of all
rules with this flag set within the same rule flow or agenda group”.
>
> I was doing a little testing of some rules earlier today, and noticed that
lock-on-active seems to change behaviour when applied to rules which don’t have an
agenda-group or rules flow-group defined. It also seemed to have a slightly inconsistent
effect, although that may just be me not realising what it’s supposed to do.
>
> There doesn’t appear to be any documentation of what the attribute means when a rule
is not part of a rule flow or agenda group. So I was wondering whether perhaps there is an
expected/official behaviour, which is just not documented. Or is lock-on-active without a
rule flow or agenda group an error? In which case is there a reason why it doesn’t cause a
compilation error when the knowledge base is built?
>
> Yours curiously...
>
> Steve
> _______________________________________________
> rules-users mailing list
> rules-users(a)lists.jboss.org
>
https://lists.jboss.org/mailman/listinfo/rules-users
>
_______________________________________________
rules-users mailing list
rules-users(a)lists.jboss.org
https://lists.jboss.org/mailman/listinfo/rules-users