[jBPM] New message: "Re: BPMN 2.0 Service Task"
by Ronald van Kuijk
User development,
A new message was posted in the thread "BPMN 2.0 Service Task":
http://community.jboss.org/message/528617#528617
Author : Ronald van Kuijk
Profile : http://community.jboss.org/people/kukeltje
Message:
--------------------------------------------------------------
> From what I can see in the example unit test this creates a new instance of the specified java class (JavaService) when the process executes, in fact if you call the method twice it creates the object twice.
Correct, that is the behaviour. The default service type in BPMN2.0 (http://www.omg.org/spec/BPMN/2.0/) is 'webservice' (you can see the high BPEL type overhead here). What the behaviour of services should be is not explicitly mentioned in the spec afaik (did not look at it lately) but webservices are mostly stateless, as are java services (tasks) in jPDL, so it is not strange that they are to be seen as stateless.
> Surely a service is more likely to have a lifecycle external to the process?
No, not in this scope. That is more going towards orchestrations or conversations in BPMN
> How can I 'inject' a prexisting service instance into the process?
Injecting them is not possible (yet?) but resolving classes, variables etc like in jPDL will probably be implemented one way or another in the future. More flexibility will be achieved then
> Specifically is it possible to access externally defined spring beans?
No, not afaik, but see the answer to the previous question.
Cheers,
Ronald
--------------------------------------------------------------
To reply to this message visit the message page: http://community.jboss.org/message/528617#528617
15 years, 10 months
[JBoss Microcontainer Development] New message: "Re: VFS3 and symlinks"
by John Bailey
User development,
A new message was posted in the thread "VFS3 and symlinks":
http://community.jboss.org/message/528610#528610
Author : John Bailey
Profile : http://community.jboss.org/people/johnbailey
Message:
--------------------------------------------------------------
Another possible solution (that I have not proven or even finished thinking about) is to detect a symlinked path at mount time and mount in both the symlinked path and the canonical path. You wouldn't really need to mount multiple filesystems, as it would really be the same filesystem mounted in multiple locations in the VFS. This would allow either path to be used from that point forward and both paths would resolve to the same filesysem entries. There would be no need to perform excessive getCanonicalFile calls once the mount has occurred.
Some things to think about:
1. This only works if the mount occurs from the symlinked path however
2. It would need to make sure the canonical path is not currently mounted with another filesystem
3. Unmounting would need to remove both
Just a thought.
--------------------------------------------------------------
To reply to this message visit the message page: http://community.jboss.org/message/528610#528610
15 years, 10 months
[JBoss Microcontainer Development] New message: "Re: CL requirements and domains not in-synch"
by Adrian Brock
User development,
A new message was posted in the thread "CL requirements and domains not in-synch":
http://community.jboss.org/message/528588#528588
Author : Adrian Brock
Profile : http://community.jboss.org/people/adrian@jboss.org
Message:
--------------------------------------------------------------
I assume you mean the commented out test? It looks like it is working as it is supposed to out-of-the-box.
The purpose of Hierarchical domains is to reproduce what was done in JBoss4.x with the Hierarchical loader repositories.
So by default they work the same way.
The jboss-classloading.xml only lets you specify parent-first=true/false,
i.e. ParentPolicy.BEFORE or ParentPolicy.AFTER_BUT_JAVA_BEFORE
If you want more control on what is visible in the parent,
then you need to create the domain manually to specify your filter.
e.g. Your test works if I change it as follows:
[ejort@warjort jboss-cl]$ svn diff classloading/src/test/java/org/jboss/test/classloading/dependency/test/HierarchicalDomainUnitTestCase.java
Index: classloading/src/test/java/org/jboss/test/classloading/dependency/test/HierarchicalDomainUnitTestCase.java
===================================================================
--- classloading/src/test/java/org/jboss/test/classloading/dependency/test/HierarchicalDomainUnitTestCase.java (revision 101491)
+++ classloading/src/test/java/org/jboss/test/classloading/dependency/test/HierarchicalDomainUnitTestCase.java (working copy)
@@ -23,6 +23,7 @@
import junit.framework.Test;
import org.jboss.classloader.spi.ClassLoaderSystem;
+import org.jboss.classloader.spi.ParentPolicy;
import org.jboss.classloading.spi.dependency.policy.mock.MockClassLoadingMetaData;
import org.jboss.classloading.spi.metadata.ClassLoadingMetaDataFactory;
import org.jboss.classloading.spi.metadata.ExportAll;
@@ -478,7 +479,8 @@
public void testExplicitRequirementsInNewDomain() throws Exception
{
- // FIXME - testExplicitRequirementsInDomain("SomeNewDomain");
+ system.createAndRegisterDomain("SomeNewDomain", ParentPolicy.BEFORE_BUT_JAVA_ONLY, system.getDefaultDomain());
+ testExplicitRequirementsInDomain("SomeNewDomain");
}
protected void testExplicitRequirementsInDomain(String domain) throws Exception
But in that case, there is little point in the hierarchical domain. ;-)
system.createAndRegisterDomain("SomeNewDomain", ParentPolicy.BEFORE_BUT_JAVA_ONLY);
would give you a new top-level domain and also make your test work.
NOTE: In real life, you would use ParentPolicy.BEFORE in the last example,
but if you used that in this test, it would load the support classes from the classpath
rather than the Mock classloader. :-)
--------------------------------------------------------------
To reply to this message visit the message page: http://community.jboss.org/message/528588#528588
15 years, 10 months
[JBoss Portal] New message: "Re: Resources for using portal api"
by Monkey Den
User development,
A new message was posted in the thread "Resources for using portal api":
http://community.jboss.org/message/528587#528587
Author : Monkey Den
Profile : http://community.jboss.org/people/monkeyden
Message:
--------------------------------------------------------------
Thanks again Wesley. I've been able to get everything working correctly within my code but transaction committal fails after the request. Seam doesn't know anything about the persistence context the portal API is using so when it tries to commit it fails with:
org.jboss.util.NestedSQLException: Could not enlist in transaction on entering meta-aware object!
...
Caused by: org.hibernate.exception.GenericJDBCException: Cannot open connection
Here is the bit of code that works, until the response:
UserModule userMod = (UserModule) new InitialContext().lookup("java:portal/UserModule");
user = userMod.findUserByUserName(username);
Any idea of how to get around this?
--------------------------------------------------------------
To reply to this message visit the message page: http://community.jboss.org/message/528587#528587
15 years, 10 months