[JBoss JIRA] Created: (JBPM-2478) jBPM4/Spring: unintended wiring PropertiesFactoryBean to RepositoryCacheImpl.deployments field
by Michael Feichtegger (JIRA)
jBPM4/Spring: unintended wiring PropertiesFactoryBean to RepositoryCacheImpl.deployments field
----------------------------------------------------------------------------------------------
Key: JBPM-2478
URL: https://jira.jboss.org/jira/browse/JBPM-2478
Project: jBPM
Issue Type: Bug
Security Level: Public (Everyone can see)
Components: Runtime Engine
Affects Versions: jBPM 4.0
Environment: Windows XP Service …
[View More]Pack 3, JDK 1.6.0_13, Eclipse Galileo JavaEE, jBPM4
Reporter: Michael Feichtegger
If you define a spring bean (<bean name="configProperties" class="org.springframework.beans.factory.config.PropertiesFactoryBean" ) this bean will be an instance of java.util.Hashtable (java.util.Properties) at runtime.
During jBPM initialization various ObjectDescriptiors create instances they describe. So do the descriptor of org.jbpm.pvm.internal.repository.RepositoryCacheImpl.
After creation the auto-wire mechanism tries to set-up the fields. In case of RepositoryCacheImpl there is just one field: Map<String, Object> deployments.
The "configProperties" bean will be found in the SpringContext since java.util.Hashtable implements java.util.Map and the field "RepositoryCacheImpl.deployments" will be set because there is just one object of the requested type (java.util.Map). The isAutoWireEnabled attribute is set to "true" (hardcoded) at RepositoryCacheBinding.
This behavior causes a NullPointerException when deleting a deployment
RepositoryCacheImpl.java line 45
deployments.put(deploymentId, null);
That statement works on java.util.HashMap (created when instantiate RepositoryCacheImpl) but not on java.util.Hashtable.
That behaviour might influence several other components.
--
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
[View Less]
14 years, 10 months
[JBoss JIRA] Created: (JBPM-2537) A triggered timer does not end the task which is left
by Alexis Seigneurin (JIRA)
A triggered timer does not end the task which is left
-----------------------------------------------------
Key: JBPM-2537
URL: https://jira.jboss.org/jira/browse/JBPM-2537
Project: jBPM
Issue Type: Bug
Security Level: Public (Everyone can see)
Components: Runtime Engine
Affects Versions: jBPM 4.0, jBPM 4.1
Reporter: Alexis Seigneurin
In my process definition, "task1" has a transition with a due date. …
[View More]When this timer is triggered by the job executor, the transition takes the process to "task2" but "task1" still looks active. For instance, when I run a "task query", instead of only finding "task2", both "task1" and "task2" are found.
Here is my process definition:
<?xml version="1.0" encoding="UTF-8"?>
<process name="test1" xmlns="http://jbpm.org/4.0/jpdl">
<start g="-9,192,48,48" name="start1">
<transition g="-44,-18" name="to task3" to="task3" />
</start>
<end g="490,192,48,48" name="end1" />
<task g="272,189,92,52" name="task1">
<transition g="-42,-18" name="to end1" to="end1" />
<transition g="-45,-18" name="to java1" to="reminder">
<timer duedate="10 seconds" />
</transition>
</task>
<java class="fr.idm.SimpleReminder" g="272,316,92,52" method="remind" name="reminder">
<transition to="task2" />
</java>
<end g="626,319,48,48" name="end2" />
<task g="454,316,92,52" name="task2">
<transition g="-42,-18" name="to end2" to="end2" />
</task>
<task g="113,189,92,52" name="task3">
<transition g="-44,-18" to="task1" name="to end1" />
</task>
</process>
For testing purpose, I introduced "task3" before "task1". It is not listed by the task query, which makes me think that, once "task1" is left, it should no longer be active.
This looks like https://jira.jboss.org/jira/browse/JBPM-967 but the suggested solution does not seem applicable to JBPM 4.x.
I will attached the whole source code.
--
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
[View Less]
14 years, 10 months
[JBoss JIRA] Created: (JBPM-1162) Problem with exception handling within the Decision nodes
by Jon Folland (JIRA)
Problem with exception handling within the Decision nodes
---------------------------------------------------------
Key: JBPM-1162
URL: http://jira.jboss.com/jira/browse/JBPM-1162
Project: JBoss jBPM
Issue Type: Bug
Components: Core Engine
Affects Versions: jBPM jPDL 3.2.2
Environment: Vaio laptop running Windows Vista, Java 1.5, JBPM 3.2.2
Reporter: Jon Folland
Assigned To: Tom Baeyens
I …
[View More]believe there may be a bug in the Decision class where exception handling is concerned which causes this failure:
If the line: String transitionName = decisionHandler.decide(executionContext); ...
...in the method public void execute(ExecutionContext executionContext)
throws an exception, which is caught here in the same method:
raiseException(exception, executionContext);
We miss the block:
if (transition==null) {
transition = getDefaultLeavingTransition();
log.debug("decision didn't select transition, taking default "+transition);
}
So when executionContext.leaveNode(transition); is called at the bottom, there wil be no transition set.
I believe we need to check for a tranistion==null condition regardless of whether an excpetion has been handled. Otherwise you will never be able to handle an exception in a Decision without another unhandled exception being thrown (which kind of defeits the point :-)
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators: http://jira.jboss.com/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
[View Less]
14 years, 11 months