news.eclipse.rt.eclipselink

[news.eclipse.rt.eclipselink] Re: MOXy - mapping enums

g8a@xxxxxxx (Matti Hansson) — Fri, 18 Sep 2009 09:47:48 GMT

Sorry. I have a much trickier transformation mapping which uses a similar structure that I'm getting the same error with, and I was hoping that getting help with this simpler mapping would lead me in the right direction on the other issue. However, it turns out that the error came from the tricky mapping all along and had nothing to do with the enum mapping.

I'll probably be back with questions about the other mapping soon. :P
/Matti

[news.eclipse.rt.eclipselink] not support executeThreadRuntime

mauro-flores@xxxxxxx (Mauro ) — Thu, 17 Sep 2009 22:14:41 GMT

I'm getting the following message using eclipslink version 1.1.2.v20090612-r4475 with weblogic 10.3.

[EL Warning]: 2009-09-17 19:03:30.215--Version of WebLogic does not support executeThreadRuntime - using ClassLoader: weblogic.management.NoAccessRuntimeException: Access not allowed for subject: principals=[], on ResourceType: ThreadPoolRuntime Action: execute, Target: getExecuteThread

Does any body know what does it mean? Did I forget to configure something? May I not worry about it?

Thanks. Mauro.

[news.eclipse.rt.eclipselink] MOXy - mapping enums

g8a@xxxxxxx (Matti Hansson) — Thu, 17 Sep 2009 13:54:45 GMT

Hi!
I'm trying to map an enumeration with MOXy using a transformation mapping but I can't get it to work. Here's what I got:

Java:
public enum Detail { GENERAL, DETAILED }

public class MyClass {
   Detail detail;
   /* getters & setters */
}
XML:
http://example.com"/>
Configuration:
private XMLDescriptor createDescriptor() {
   XMLDescriptor descriptor = super.buildDescriptor(MyClass.class);
   descriptor.setDefaultRootElement("app:MyClass");
   descriptor.add(createDetailMapping());
   return descriptor;
}

private XMLTransformationMapping createDetailMapping() {
   XMLTransformationMapping mapping = new XMLTransformationMapping();
   mapping.setAttributeName("detail");
   EnumTransformer transformer = new EnumTransformer();
   mapping.setAttributeTransformer(transformer);
   mapping.addFieldTransformer("@detail", transformer);
   return mapping;
}
Transformer:
public class EnumTransformer implements AttributeTransformer, FieldTransformer {
 private AbstractTransformationMapping mapping;
 private XMLField detailField;

 public void initialize(AbstractTransformationMapping mapping) {
   this.mapping = mapping;
   NamespaceResolver namespaceResolver = ((XMLDescriptor)mapping.getDescriptor()).getNamespaceResolver();
   detailField = new XMLField("@detail");
   detailField.setNamespaceResolver(namespaceResolver);
 }

public Object buildAttributeValue(Record record, Object object, Session session) { String detail = (String) record.get(detailField); return Detail.valueOf(detail); } public Object buildFieldValue(Object instance, String fieldName, Session session) { return mapping.getAttributeValueFromObject(instance); } }

When I try to run it I get this stack trace:
java.lang.NullPointerException
	at org.eclipse.persistence.oxm.record.UnmarshalRecord.startElement(UnmarshalRecord.java:551)
	at org.eclipse.persistence.internal.oxm.record.SAXUnmarshallerHandler.startElement(SAXUnmarshallerHandler.java:273)
	at com.sun.org.apache.xerces.internal.parsers.AbstractSAXParser.startElement(AbstractSAXParser.java:501)
	at com.sun.org.apache.xerces.internal.impl.XMLNSDocumentScannerImpl.scanStartElement(XMLNSDocumentScannerImpl.java:400)
	at com.sun.org.apache.xerces.internal.impl.XMLNSDocumentScannerImpl$NSContentDriver.scanRootElementHook(XMLNSDocumentScannerImpl.java:626)
	at com.sun.org.apache.xerces.internal.impl.XMLDocumentFragmentScannerImpl$FragmentContentDriver.next(XMLDocumentFragmentScannerImpl.java:3095)
	at com.sun.org.apache.xerces.internal.impl.XMLDocumentScannerImpl$PrologDriver.next(XMLDocumentScannerImpl.java:921)
	at com.sun.org.apache.xerces.internal.impl.XMLDocumentScannerImpl.next(XMLDocumentScannerImpl.java:648)
	at com.sun.org.apache.xerces.internal.impl.XMLNSDocumentScannerImpl.next(XMLNSDocumentScannerImpl.java:140)
	at com.sun.org.apache.xerces.internal.impl.XMLDocumentFragmentScannerImpl.scanDocument(XMLDocumentFragmentScannerImpl.java:510)
	at com.sun.org.apache.xerces.internal.parsers.XML11Configuration.parse(XML11Configuration.java:807)
	at com.sun.org.apache.xerces.internal.parsers.XML11Configuration.parse(XML11Configuration.java:737)
	at com.sun.org.apache.xerces.internal.parsers.XMLParser.parse(XMLParser.java:107)
	at com.sun.org.apache.xerces.internal.parsers.AbstractSAXParser.parse(AbstractSAXParser.java:1205)
	at com.sun.org.apache.xerces.internal.jaxp.SAXParserImpl$JAXPSAXParser.parse(SAXParserImpl.java:522)
	at org.eclipse.persistence.internal.oxm.record.XMLReader.parse(XMLReader.java:99)
	at org.eclipse.persistence.internal.oxm.record.SAXUnmarshaller.unmarshal(SAXUnmarshaller.java:215)
	at org.eclipse.persistence.internal.oxm.record.SAXUnmarshaller.unmarshal(SAXUnmarshaller.java:192)
	at org.eclipse.persistence.oxm.XMLUnmarshaller.unmarshal(XMLUnmarshaller.java:274)
	...


Anyone knows what's going on?
Thanks!

[news.eclipse.rt.eclipselink] Re: EclipseLink-7242

mauro-flores@xxxxxxx (Mauro ) — Wed, 16 Sep 2009 22:35:51 GMT

If a call a business method, I must know previously what it returns, I must know if it was implemented to bring a relationship or not. That would be enough.

But for openjpa null collection means that the It was not initialized and empty collection means that there is no elements indeed.

Mauro.

[news.eclipse.rt.eclipselink] Re: EclipseLink-7242

mauro-flores@xxxxxxx (Mauro ) — Wed, 16 Sep 2009 22:14:32 GMT

Doug,

I really appreciate your reply.

"I am unsure how the application would differentiate this state"
That's the backstage of openjpa. My application don't worry about it.

I made I simple example that can show the difference of behavior. It shows the change of values in 3 scenarios traced before and after the serialization (web context) 1)after the find and before the execution of method get of relationship 2)after the get and before the method size 3)after the size

In openjpa, if a relationship was no initialized (scenario 1) , its serialized with null. As soon as a method getRelationship is executed its filled with a subclass of ArrayList with 0 or more elements(scenario 2 e 3). So a null collection means that relationship was not initialized and an empty collection means that the relationship doesn't exist.

In eclipselink, in all scenarios the relationship has an InternalList. In scenario 1 and 2 the internal list has the property isInstantiated set to false and throws an exception when serialized In scenario 3, the property isInstantiated is set to true and relationship is serialized.

Eclipselink wouldn't have an configured option sinalizing to serialize relationship as null when the property isInstatiated is set to false???? Wouldn't it be a good approach?

I'm working on a proxy (Ejb interceptor) that makes exactly this. Each collection of a returned entity that has an InternalList not instantiated is set to null before the serialization. I'm not sure if that's the best approach.

Example
=======

--------------------------------------------------- EJB Class: ---------------------------------------------------

@Stateless(mappedName = "ARN")
@TransactionAttribute(TransactionAttributeType.REQUIRED)
public class ARNImpl implements ARN {

	protected final Log logger = LogFactory.getLog(getClass());

@Resource SessionContext context; @PersistenceContext(unitName="TESTE") EntityManager em;

@TransactionAttribute(TransactionAttributeType.REQUIRED) public A consultWithoutGet(int code) { System.out.println("Code" + code); A a = em.find(A.class, code); return a; } @TransactionAttribute(TransactionAttributeType.REQUIRED) public A consultWithoutSize(int code) { System.out.println("Code" + code); A a = em.find(A.class, code); List listaB = a.getListB(); if (listaB == null) { System.out.println("SessionBean Before size() - null relationship"); return a; } System.out.println("SessionBean Before size() - Class relationship: " + listaB.getClass().getName()); System.out.println("SessionBean Before size() - Super Class relationship: " + listaB.getClass().getSuperclass().getName()); if (listaB instanceof IndirectList) { System.out.println("SessionBean Before size() - Eclipselink Instantiated: " + ((IndirectList)listaB).isInstantiated()); } return a; } @TransactionAttribute(TransactionAttributeType.REQUIRED) public A consultWithSize(int code) { System.out.println("Code" + code); A a = em.find(A.class, code); int size = a.getListB().size(); List listaB = a.getListB(); if (listaB == null) { System.out.println("SessionBean After size() - null relationship"); return null; } listaB = a.getListB(); System.out.println("SessionBean size - : " + size); System.out.println("SessionBean After size() - Class relationship: " + listaB.getClass().getName()); System.out.println("SessionBean After size() - Super Class relationship: " + listaB.getClass().getSuperclass().getName()); if (listaB instanceof IndirectList) { System.out.println("After size() - Eclipselink Instantiated: " + ((IndirectList)listaB).isInstantiated()); } return a; }

---------------------------------------------- Pojo Method called from JSP that calls the EJB ----------------------------------------------

public void consult() { ARN arn = (ARN) ServiceLocator.getEJB3(ARN.class); A a = arn.consultWithoutGet(code); try { if (a.getListB() != null) { System.out.println("Web context - before get() - class " + a.getListB().getClass().getName()); System.out.println("Web context - before get() - size " + a.getListB().size()); } else { System.out.println("Web context - before get() - relationship null"); } } catch (Exception e) { System.out.println("Web context - before size() - exeption: " + e.getClass()); } try { a = arn.consultWithoutSize(code); if (a.getListB() != null) { System.out.println("Web context - before size() - class " + a.getListB().getClass().getName()); System.out.println("Web context - before size() - size " + a.getListB().size()); } else { System.out.println("Web context - before size() - relationship null"); } } catch (Exception e) { System.out.println("Web context - before size() - exeption: " + e.getClass()); } try { a = arn.consultWithSize(code); System.out.println("Web context - after size() - class " + a.getListB().getClass().getName()); System.out.println("Web context - after size() - size " + a.getListB().size()); } catch (Exception e) { System.out.println("Web context - after size() - exeption: " + e.getClass()); } }

------------------ Entity A ------------------ @Entity @Table(name = "A") public class A implements Serializable{

@OneToMany(fetch = FetchType.LAZY, mappedBy = "a") private List listB = new ArrayList();

}

======================== Running with eclipselink ========================

Web context - before get() - class org.eclipse.persistence.indirection.IndirectList Web context - before get() - exeption: class org.eclipse.persistence.exceptions.ValidationException

SessionBean Before size() - Class relationship: org.eclipse.persistence.indirection.IndirectList SessionBean Before size() - Super Class relationship: java.util.Vector SessionBean Before size() - Eclipselink Instantiated: false Web context - before size() - class org.eclipse.persistence.indirection.IndirectList Web context - before size() - exeption: class org.eclipse.persistence.exceptions.ValidationException

SessionBean size - : 2 SessionBean After size() - Class relationship: org.eclipse.persistence.indirection.IndirectList SessionBean After size() - Super Class relationship: java.util.Vector After size() - Eclipselink Instantiated: true Web context - after size() - class org.eclipse.persistence.indirection.IndirectList Web context - after size() - size 2

==================== Running with openjpa ====================

Web context - before get() - relationship null

SessionBean Before size() - Class relationship: org.apache.openjpa.util.java$util$ArrayList$proxy SessionBean Before size() - Super Class relationship: java.util.ArrayList Web context - before size() - class org.apache.openjpa.util.java$util$ArrayList$proxy Web context - before size() - size 2

SessionBean size - : 2 SessionBean After size() - Class relationship: org.apache.openjpa.util.java$util$ArrayList$proxy SessionBean After size() - Super Class relationship: java.util.ArrayList Web context - after size() - class org.apache.openjpa.util.java$util$ArrayList$proxy Web context - after size() - size 2

[news.eclipse.rt.eclipselink] Re: EclipseLink-7242

douglas.clarke@xxxxxxx (Doug Clarke) — Tue, 15 Sep 2009 14:37:12 GMT

Mauro,

I have been assisting several groups in doing similar migrations and would be interested in any all feedback or notes on your efforts so we can publish them on the wiki to assist the entire community.

On this specific issue I am not sure I understand the functionality you are seeing in OpenJPA. During detachment (through serialization) if a null reference or empty collection is returned I am unsure how the application would differentiate this state from the scenario where the entity actually had a null or empty collection. We have always sided with throwing an exception versus returning an entity in a state which may be incorrect.

Can you provide more insight as to how your application differentiates a null value or empty collection caused by serialization versus one that is actually null or empty?

Doug

[news.eclipse.rt.eclipselink] Re: release plans

douglas.clarke@xxxxxxx (Doug Clarke) — Tue, 15 Sep 2009 14:32:00 GMT
We are currently reviewing our release plans. The published plans for 1.2.0, 2.0.0, and 1.1.3 will most likely all be adjusted shortly. The schedule change is due to changes in the overall Java EE 6 schedule of which EclipseLink is delivering the JPA 2.0 reference implementation as well as some delays getting IP reviewed for the originally scheduled 1.2.0 release.

I am hoping to propose an updated schedule for final review at this week's committer meeting (Thursday) and will update the published release dates following this meeting.

Once we have the proposed schedule published for the committers to review I will post a link here as well so we can get additional community feedback.

Doug

[news.eclipse.rt.eclipselink] Re: release plans

my.mailing.lists@xxxxxxx (Marc Schlegel) — Mon, 14 Sep 2009 16:45:09 GMT

Hi Thanks for asking this, I am also not sure about the releases right now. And I've got one more question: Version 2.0 is scheduled for this fall and I am very keen on that since I want to dig into JPA 2 and not bothering with the old spec to long. Is it possible to use EclipseLink 2.0 in an RCP app right away, or do we have to wait for the next eclipse simultaneous release next summer? > Hi eclipselink team, > > could you please specify the release plans for the next weeks. > In the master timeline, there is a 1.2.0 release scheduled for the > 9/30/2009, but up to now there is no milestone or RC available. > > In my company we have a release scheduled for 9/30/2009, and the 1.1.2 > code that we use currently contains at least on blocker: > https://bugs.eclipse.org/bugs/show_bug.cgi?id=288620 > and with a potential workaround: > https://bugs.eclipse.org/bugs/show_bug.cgi?id=221389 > > How stable are the milestones? For example the 1.1.3 M2 passes all the > unit tests in the JPA section. > > Could you please also explain what are the differences between the 1.1.3 > and the 1.2.0 release. > And will be the JPA 1.0 handling in the 2.0 release backwards compatible > to the 1.1.2/1.2.0 versions? > > Will the 1.1.3 release contain the bugfixes for the earlier 1.2.0 release? > > thanks in advance for your answers. > > best regards, > Felix >

[news.eclipse.rt.eclipselink] Re: ddl-generation

james.sutherland@xxxxxxx (James) — Mon, 14 Sep 2009 15:04:13 GMT

Are you just seeing errors logged, or is an error occurring?

With drop-and-create EclipseLink will log errors if the drop fails, but will ignore the errors and try to create the tables, errors will only be thrown if the create fails.

With create EclipseLink will log errors if the create fails, but ignore the errors and assume the database is already created.

You can turn logging off, if you wish to not see the errors logged. Sometimes they are useful, as a drop may have failed because of a constraint error.

[news.eclipse.rt.eclipselink] release plans

hansharz_bugzilla@xxxxxxx (Hans Harz) — Mon, 14 Sep 2009 12:42:52 GMT

Hi eclipselink team,

could you please specify the release plans for the next weeks. In the master timeline, there is a 1.2.0 release scheduled for the 9/30/2009, but up to now there is no milestone or RC available.

In my company we have a release scheduled for 9/30/2009, and the 1.1.2 code that we use currently contains at least on blocker: https://bugs.eclipse.org/bugs/show_bug.cgi?id=288620 and with a potential workaround: https://bugs.eclipse.org/bugs/show_bug.cgi?id=221389

How stable are the milestones? For example the 1.1.3 M2 passes all the unit tests in the JPA section.

Could you please also explain what are the differences between the 1.1.3 and the 1.2.0 release. And will be the JPA 1.0 handling in the 2.0 release backwards compatible to the 1.1.2/1.2.0 versions?

Will the 1.1.3 release contain the bugfixes for the earlier 1.2.0 release?

thanks in advance for your answers.

best regards, Felix