you're right, an uncorrelated subquery is much better. A way of tweaking envers is to create a patch . I'll be happy to commit it, if you provide create one .
i thought it might be more
efficient to communicate the issue to the envers
team and have someone who already knows the codebase
figure out how best revise the code.
but i suppose i could take a look at the envers codebase and
try to figure it out and patch it. that would be most efficient for
the person not doing the work. :-)
thanks for the speedy reply.