On Jul 11, 2011, at 11:55 AM, "Kevin Grittner" <Kevin.Grittner@wicourts.gov> wrote:
> Robert Haas <robertmhaas@gmail.com> wrote:
>
>> I find these responses to be a bit off point.
>
> The OP is basically looking for what Florian tried to implement.
> This is perhaps a *bit* off point, but arguably not more than
> pointing someone who is requesting planner hints in another
> direction. And someone thought the issues were related here:
>
> http://archives.postgresql.org/pgsql-hackers/2010-12/msg01792.php
>
> ;-)
Well, fair enough. I thought of the connection between this request and Florian's work, too. I would very much like to
supportwhat he proposed, but it doesn't appear viable without a heapam rewrite, or maybe a lock manager rewrite.
However,I think that's a somewhat separate question from whether we need to forbid SFU on the outer side of a join.
Tom's question seems to me to be right on target: what semantics do our competitors assign to this construct? And do
theybroadly agree with each other?
>> Evidently our competition does not agree
>
> Neither on this nor on planner hints. ;-)
Well, we are a pretty smart group of people. But I don't think we should completely ignore what other people are doing,
onany topic.
...Robert