On 2010-11-12 8:25 PM +0200, I wrote:
> I'm going to take some time off this weekend to get a patch with this
> behaviour to the next commitfest.
.. and a wild patch appears.
This is almost exactly the patch from 2010-02 without
CommandCounterIncrement()s. It's still a bit rough around the edges and
needs some more comments, but I'm posting it here anyway.
This patch passes all regression tests, but feel free to try to break
it, there are probably ways to do that. This one also has the "always
run DMLs to completion, and exactly once" behaviour.
Regards,
Marko Tiikkaja