I recognize that this is not a Hibernate problem per se, but no one is more informed than the people here on java-vs-database issues so I thought I'd give it a shot.

I am running into a problem where a query with an order-by including character fields is run. It turns out that the database seems to be running with a different collator assumption than java is, even when both database (Postgres in this case) and java are running on the same box under the same user.

This creates some annoying issues. Is there a way via Hibernate and/or JDBC to get the collation scheme used by Postgres, and use that to do String comparisons under java? I've never run into this problem before and a quick glance at documentation isn't helping.