Hi, imagine having a tree of entities and some of them having changed data. In order to save the entities to the database they have to be validated first. During validation it is necessary to fetch some additional validation data from the database. Now if OpenJPA comes to fire the first select it tries to persist the changed data per default (FlushBeforeQueries=true) and an exception is raised because in our case some important entity fields are set shortly before commit using a callback listener (@PrePersist and @PreUpdate). If these fields aren't set the database throws an error.
Why is FlushBeforeQueries=true the default and what is best practice to cope with that situation? The flush should execute if the business method which started the transaction ends. In my opinion our process (changing data, fetching data to perform validation, persisting) is not really an exceptional case. Which combination of FlushBeforeQueries and IgnoreChanges can we use to execute queries during transaction? Thanks for your help! -- View this message in context: http://openjpa.208410.n2.nabble.com/Executing-queries-during-transaction-tp7580793.html Sent from the OpenJPA Users mailing list archive at Nabble.com.
