On Sun, Jun 9, 2013 at 10:51 AM, Simon Riggs <simon@2ndquadrant.com> wrote:
> AFAICS there are weird cases where changing the way FKs execute will
> change the way complex trigger applications will execute. I don't see
> a way to avoid that other than "do nothing". Currently, we execute the
> checks following the normal order of execution rules for triggers.
> Every idea we've had so far changes that in some way; variously in
> major or minor ways, but changed nonetheless.
The obvious case to handle would be if someone has a trigger to
automatically create any missing references.
--
greg