[jBPM Users] - Re: [jBPM4] final state when async node is used
by kukeltje
Your original question was:
anonymous wrote : My question is: how do we know that a workfow is ended ?
My answer was:
anonymous wrote :
| if you retrieve the processinstance by id and it is null it was either not there to begin with or it is ended. If you are not sure, you can always check in the history. It should be there then.
This is valid AND the truth. (Ab) using the getState is not the correct way, since if a process is really ended, the assertTrue(processInstance.isEnded());
would produce a(n?) NPE since it is directly moved to the history table.
So your reply seemed not related to me and I did not respond.
Btw, If you post in the forum and make a Jira issue which is related, you can point that out in the jira issue. There is a specific field for that.
Regarding tests in general, yes, for tests you have to know the workflow. That is what tests are for. So that is not an issue either (imo)
If you could refine the test some more so the last statement does throw an NPE and the getState() still is not equal to 'completed', please get back to us.
Hope this helps to clarify some things.
View the original post : http://www.jboss.org/index.html?module=bb&op=viewtopic&p=4255145#4255145
Reply to the post : http://www.jboss.org/index.html?module=bb&op=posting&mode=reply&p=4255145
16 years, 7 months
[Persistence] - ClassNotFoundException Moving Persistence.xml From EJB/META-
by wluque
Hi, i have a application with the next structure (work fine un jboss 4.2.2.ga) :
..FMK-EAR
......fmk-war
......fmk-seam ( Seam Actions Ejb )
......fmk-sessionEjb ( Ejb beans )
......fmk-zentities ( Ejb with the entities definition )
......META-INF
............application.xml
............jboss-app.xml
............persistence.xml ( move recently )
In fmksessionEjb i have @PersistenceContext(unitName = "ejb3FMK") , so I move the persistence.xml from fmk-zentities/META-INF to FMK-EAR/META-INF.
I read in some post, that this is necessary for persistence unit is visible to fmk-sessionEjb.
Now i have the next error :
javax.persistence.PersistenceException: [PersistenceUnit: ejb3FMK] class or package not found
....
Caused by: java.lang.ClassNotFoundException: fmk-zentities.jar.com.multiSistemas.fmk.entity.security.UserSecuritySpaceAccess from ...
The class com.multiSistemas.fmk.entity.security.UserSecuritySpace exists in fmk-zentities.jar .
What is wrong? any idea?
Thank in advance
View the original post : http://www.jboss.org/index.html?module=bb&op=viewtopic&p=4255137#4255137
Reply to the post : http://www.jboss.org/index.html?module=bb&op=posting&mode=reply&p=4255137
16 years, 7 months
[Javassist Users] - Field access like in GluonJ
by Corifeo
I try to implement a field access as the @Get and @Set Listener-Annotations in GluonJ, directly with Javassist.
I looked into the source code how GluonJ performs this, but I can't get it running - and don't get the exact point.
e.g. Class A
{
String myString;
}
Does anybody know:
- is it possible to only modify Class A, e.g. by injecting a bytecode method invocation, to 'redirect' every possible access (set/get) on myString? Does GluonJ so? I found snippets like 'constPool.addMethodrefInfo(classIndex, nt); or 'codeConverter2.redirectFieldAccess(superOf, m.getDeclaringClass(), fname);
- Or is it only possible to substitute the field accesses in specified classes/methods, and GluonJ substitute these brute-force in every class that is processed by the class loader...which is not so elegant, but perhaps the only possibility
- What are the necessary Javassist invocations to achieve such Listeners? I can't find an according snippet in the tutorial. Is it possible to paste one here?
Thanks in advance!
Christian
View the original post : http://www.jboss.org/index.html?module=bb&op=viewtopic&p=4255136#4255136
Reply to the post : http://www.jboss.org/index.html?module=bb&op=posting&mode=reply&p=4255136
16 years, 7 months