Marko Tiikkaja <marko.tiikkaja@cs.helsinki.fi> writes:
> On 2010-11-08 7:26 PM +0200, Tom Lane wrote:
>> The alternative is to artificially serialize the DML CTEs, which
>> while it does have some advantages doesn't seem like a win overall.
> We've discussed this before and the consensus was that as long as we
> don't change the results, we can optimize the materialization away.
No, because the problem is mainly about what might happen if
user-defined functions choose to look at the target tables. We can't
really tell what triggers are going to do, to take one item that the
planner has no access to.
I think we have to decide up front which implementation behavior
it's going to be, and if we go with serialized queries, we'll be
stuck with that.
regards, tom lane