Community
Participate
Working Groups
Build Identifier: 2.1.3 and newer The patch of bug 243993 causes the regression that newly persisted entities (and maybe other changes as well) which are made in the event listener during a preCalculateUnitOfWorkChangeSet event are lost or at least not written properly to the database. The reason is that EL is now cloning the clone mapping which causes that persists made in the event listener will update the original map only - but not the created clone which gets used in calculateChanges() and in subsequent database actions. As a consequence we cannot use EclipseLink releases > 2.1.2. For us, it's a blocker. Is there a workaround available? BTW: bug 332723 would also help me in this case - so I don't have to misuse preCalculateUnitOfWorkChangeSet for reacting on a flush(). Reproducible: Always Steps to Reproduce: 1. create and register a event listener which persists an entity during the preCalculateUnitOfWorkChangeSet event 2. call flush() or commit() on the EntityManager 3. new entity can not be found in the database
Setting target and priority. See the following page for the meanings of these fields: http://wiki.eclipse.org/EclipseLink/Development/Bugs/Guidelines Community: Please vote for this bug if it is important to you. Votes are one of the main criteria we use to determine which bugs to fix next.
Created attachment 199753 [details] Proposed changes
Created attachment 199769 [details] 2.3.1 patch
Created attachment 199770 [details] trunk patch
Changes have been submitted to 2.2.1, 2.3.1 and trunk. Verified by: Chris Delahunt Tests: New test (testSessionEventListenerOnBus) added to LifecycleCallbackJunitTest
The Eclipselink project has moved to Github: https://github.com/eclipse-ee4j/eclipselink
