[jboss-jira] [JBoss JIRA] Commented: (JBRULES-1804) Logical insertion lingers after the rule that inserted it no longer supports it (with testcase patch)

[Geoffrey De Smet (JIRA)]

    [ https://jira.jboss.org/jira/browse/JBRULES-1804?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12434297#action_12434297 ] 

Geoffrey De Smet commented on JBRULES-1804:
-------------------------------------------

Mark, it's indeed what we discussed month ago, but I didn't think I explained it well enough, so I hoped a testcase would clarify the issue.
It's been bugging me in drools-solver ever since, consistently giving me hard to track down corrupted scores, and apparently other drools-solver users, such as Tim, too:
  "exactly! that is the behavior i was expecting (=> the new behaviour). it took me quite some time to find out why my program did not find a solution."

After a good night sleep, I still don't get it.
1) Could you give me one example where the original behavior would actually be a good thing?

"Jess approach is to not preserve a logical assertion on a modify, so when you do a modify on the retract part of the modify all the logical assertions are removed too, and won't be re-added until the consequence fires. We could possible allow this, would that suffice?"
2) Would the testcase in subversion end up with temperatures of 150 and 200 and not 100? If yes, that would suffice (and would make my day :)).

> Logical insertion lingers after the rule that inserted it no longer supports it (with testcase patch)
> -----------------------------------------------------------------------------------------------------
>
>                 Key: JBRULES-1804
>                 URL: https://jira.jboss.org/jira/browse/JBRULES-1804
>             Project: JBoss Drools
>          Issue Type: Bug
>      Security Level: Public(Everyone can see) 
>          Components: drools-core  (expert)
>    Affects Versions:  5.0.0.M2
>            Reporter: Geoffrey De Smet
>            Assignee: Edson Tirelli
>            Priority: Critical
>             Fix For: 5.0.0.M3
>
>
> This one got me headscratching for a while, untill I realized what the problem is.
> Tim reported on the mailing list he ran into it to.
> I've commmited a testcase in subversion, in TruthMaintenanceTest, called
>    disabled_testLogicalInsertionsModifySameRuleGivesDifferentLogicalInsertion
> Remove the "disabled_" to enable it and try it out.
> Here's what happens in the testcase:
> 3 sensors with tempature 100, 200 and 200 are inserted.
> There's a rule which logically inserts all encountered tempatures, so 100 and 200.
> Then the first sensor modifies to a tempature of 150.
> As it turns out, the encoutered tempatures should be 150 and 200 now, but instead it's 100, 150 and 200.
> Even though there is no sensor that has a tempature of 100.

-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira