[ https://issues.jboss.org/browse/JBRULES-3144?page=com.atlassian.jira.plug... ] Geoffrey De Smet commented on JBRULES-3144: ------------------------------------------- Here's what I'd like to be able to do: {code} rule "serviceConflict" when $leftProcessAssignment : MrProcessAssignment($service : service, $machine : machine, $leftId : id) $rightProcessAssignment : MrProcessAssignment(service == $service, machine == $machine, id > $leftId) then scoreCalculator.addToHardScore(1); // No insertLogical or accumulate of those ConstraintOccurence's needed undo-then scoreCalculator.substractFromHardScore(1); end {code} And here's a more complex example where the weight isn't a static "1": {code} rule "serviceLocationSpread" when $service : MrService($locationSpread : locationSpread) $spreadCount : Number(intValue < $locationSpread) from accumulate( $location : MrLocation() and exists MrProcessAssignment(service == $service, location == $location), count($location) ) then scoreCalculator.addToHardScore($locationSpread - $spreadCount.intValue()); undo-then scoreCalculator.substractFromHardScore($locationSpread - $spreadCount.intValue()); // $spreadCount should still be the same as it used to be end {code} -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.jboss.org/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira

[ https://issues.jboss.org/browse/JBRULES-3144?page=com.atlassian.jira.plug... ] Geoffrey De Smet updated JBRULES-3144: -------------------------------------- Summary: undo-then: when rule becomes no longer true execute a callback" (was: Experiment with "when rule becomes no longer true execute a callback" feature in Drools Expert) -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.jboss.org/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira

[ https://issues.jboss.org/browse/JBRULES-3144?page=com.atlassian.jira.plug... ] Geoffrey De Smet updated JBRULES-3144: -------------------------------------- Description: Here's what I'd like to be able to do: {code} rule "serviceConflict" when $leftProcessAssignment : MrProcessAssignment($service : service, $machine : machine, $leftId : id) $rightProcessAssignment : MrProcessAssignment(service == $service, machine == $machine, id > $leftId) then scoreCalculator.addToHardScore(1); // No insertLogical or accumulate of those ConstraintOccurence's needed undo-then scoreCalculator.substractFromHardScore(1); end {code} And here's a more complex example where the weight isn't a static "1": {code} rule "serviceLocationSpread" when $service : MrService($locationSpread : locationSpread) $spreadCount : Number(intValue < $locationSpread) from accumulate( $location : MrLocation() and exists MrProcessAssignment(service == $service, location == $location), count($location) ) then scoreCalculator.addToHardScore($locationSpread - $spreadCount.intValue()); undo-then scoreCalculator.substractFromHardScore($locationSpread - $spreadCount.intValue()); // $spreadCount should still be the same as it used to be end {code} was: <_conan_> ge0ffrey: can do unfalse situation <ge0ffrey> conan: is there a way to get a "when the rule no longer applies, execture this code"? <_conan_> ge0ffrey: i.e. when a ruel becomes untrue execute a callback <ge0ffrey> conan: yes! <_conan_> yes <_conan_> it's trivial to do actually <ge0ffrey> conan: what's the code name for that feature? <_conan_> not related to TMS <ge0ffrey> yep <_conan_> you can potentially hack one for yourself <_conan_> in about 10 minutes <ge0ffrey> I 'd also need to have a value which was caclulated in hte the orginal RHS <_conan_> but the feature over all will take a lot of thought, due to language design <_conan_> but for your use case <_conan_> it's trivial to do <_conan_> in the rule you have the Activation <_conan_> the impl is the AgendaItem <ge0ffrey> conan: would love to do this (espeically if I get some appointed time for it :) <_conan_> you can do kcontext.getActivation <_conan_> it'll take you 15 minutes to do it <_conan_> change AgendaItem to have new field reference Callable or what ever you want to use. <ge0ffrey> k <_conan_> in the consequnce just do agendaItem.addCallable( new Callable() {............ } ); <_conan_> i.e. use an anonymous class <_conan_> the only thing you need to do now <_conan_> is in RuleTerminalNode in retractLeftTuple is agendaItem.executeCallables() <_conan_> that's it <_conan_> it's that trivial <ge0ffrey> (copy pasting htis in an issue to look at later, as my workspace is open on my fork today) <_conan_> I just don't want to expose that as a generic feature <_conan_> because the issue needs more thought <ge0ffrey> of course <_conan_> with regards to language design <_conan_> but for your use case <_conan_> and as you are an expert <_conan_> it's fine <_conan_> I'd suggest you just maintain a "patched" version of drools for that feature <ge0ffrey> but if the exeperiment works well, we can look at language design <_conan_> yes <_conan_> I need this anyway <_conan_> becausae rule engines need to tell peopl when things stop being true <ge0ffrey> conan: not going to start using for released planner until it's in drools expert for real -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.jboss.org/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira

[ https://issues.jboss.org/browse/JBRULES-3144?page=com.atlassian.jira.plug... ] Geoffrey De Smet edited comment on JBRULES-3144 at 12/16/11 10:55 AM: ---------------------------------------------------------------------- I wrote a buggy undo-then implementation and ran some benchmarks to see how much faster it is: for 256 queens for 100 steps (with undoMove assertion) average calculate count per second (higher is better): - classic planner: 111537 - undo-then planner: 205672 (84% faster) for 512 queens for 100 steps (with undoMove assertion) average calculate count per second (higher is better): - classic planner: 111472 (-0.06%) - undo-then planner: 187462 (68% faster, -8.85%) It's faster, but scaling doesn't seem to be affected. Note that it doesn't lower scaling as those number might make you think, because it just means that the insertLogical's part is not the deciding factor as it scales out. was (Author: ge0ffrey): I wrote a buggy undo-then implementation and ran some benchmarks to see how much faster it is: for 256 queens for 100 steps (with undoMove assertion) average calculate count per second: - classic planner: 111537 - undo-then planner: 205672 for 512 queens for 100 steps (with undoMove assertion) average calculate count per second: - classic planner: 111472 (-0.06%) - undo-then planner: 187462 (-8.85%) It's faster, but scaling doesn't seem to be affected. Note that it doesn't lower scaling as those number might make you think, because it just means that the insertLogical's part is not the deciding factor as it scales out. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.jboss.org/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira

[ https://issues.jboss.org/browse/JBRULES-3144?page=com.atlassian.jira.plug... ] Geoffrey De Smet reassigned JBRULES-3144: ----------------------------------------- Assignee: Mark Proctor (was: Geoffrey De Smet) My experiments proved that planner can be a lot faster if undo-then is implemented in drools expert, so I hope it's considered in the DRL language design to include undo-then. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.jboss.org/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira

[ https://issues.jboss.org/browse/JBRULES-3144?page=com.atlassian.jira.plug... ] Mark Proctor updated JBRULES-3144: ---------------------------------- Summary: UnMatch Listener for Compensation Actions (was: undo-then: when rule becomes no longer true execute a callback) -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.jboss.org/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira

[ https://issues.jboss.org/browse/JBRULES-3144?page=com.atlassian.jira.plug... ] Mark Proctor commented on JBRULES-3144: --------------------------------------- see AgendaTest: {code} AgendaItem item = ( AgendaItem ) knowledgeHelper.getActivation(); final Cheese cheese = ( Cheese ) item.getTuple().getHandle().getObject(); final int oldPrice = cheese.getPrice(); cheese.setPrice( 100 ); item.setActivationUnMatchListener( new ActivationUnMatchListener() { public void unMatch(org.drools.runtime.rule.WorkingMemory wm, org.drools.runtime.rule.Activation activation) { cheese.setPrice( oldPrice ); } } ); {code} -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.jboss.org/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira