Hi Michael,
Its a long story, but you can read about it here:
http://n2.nabble.com/%40OneToMany-%40ManyToOne%2C-Bidirectional%2C-Composite-Key-BUG-tc210672.html#a221083
http://n2.nabble.com/%40OneToMany-%40ManyToOne%2C-Bidirectional%2C-Composite-Key-BUG-tc210672.html#a221083
In the mean time, I've simplified my test case like so:
public void updateTblFndmst(TblFndmst tblFndmst){
try {
oem.getTransaction().begin();
em.merge(tblFndmst);
oem.getTransaction().commit();
} catch (Exception e) {
e.printStackTrace();
}
}
...and the version number is still incrementing.
The relavant OpenJPA code appears to be:
StateManagerImpl: line 1572
if (active) {
if (_broker.getOptimistic()) //<-- here
setPCState(_state.beforeOptimisticWrite(this, field,
mutate.booleanValue()));
else
setPCState(_state.beforeWrite(this, field,
mutate.booleanValue()));
}
PNonTransState: line 80
PCState beforeOptimisticWrite(StateManagerImpl context, int field,
boolean mutate) {
if (context.getBroker().getAutoClear() == AutoClear.CLEAR_ALL)
return beforeWrite(context, field, mutate, true);
return PDIRTY; //returns PDIRTY
}
Will debug some more tomorrow.
--
View this message in context:
http://n2.nabble.com/Non-dirty-entity-version-field-update-SUPER-URGENT%21%21%21-tp1120307p1129303.html
Sent from the OpenJPA Users mailing list archive at Nabble.com.
