Hibernate version: 3.2.2

I have an existing schema with items split into two identical tables - 'active_tasks' and 'completed_tasks'. Upon completion of a task it is moved from the active_tasks table to the completed_tasks table.

What is the best way to map a Task object (or hierarchy) onto these two tables?

I finally got InheritanceType.TABLE_PER_CLASS to work with an abstract Task class and two subclasses, but the SQL generated is very poorly performing with a moderately large completed_tasks table.

The SQL is basically:

select * from (select * from active_tasks union select * from completed_tasks) where task_id=?

Is there anything I can do to improve this? This has to be a fairly common problem of having identical items stored in multiple tables (say tables for each month, etc.).

Thanks,

--Kevin

Well, I may have figured out a way to get around this, but it is not pretty and is fragile should hibernate internals change (change clazz_ to something else), etc.



But, this performs much better for my scenario where the number of tables is small (2), but the quantity of data in them is large.

I would still like to know the "right" way to do this if there is one.

--Kevin