As you might have seen from my post about "PostgreSQL JDBC vs fetchSize", there seemed to be one consistent failure with the ODMG ObjectImageTest under PostgreSQL.
However, just to calm Armin down ;-) I just found that it is another one of our favorite failures of late: SQL results ordering and non-deterministic tests.
I'm debugging the test now and it's clear that an updated object fetched back from PostgreSQL will change place in a referenced Book->Review vector.
I will check in a mini-fix for this soon.
Regards, Martin
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
