Ed L. wrote:
> I ran a simple select-for-update test on 8.1.2 and was curious as
> to why the semantics are what they are. Specifically, when you
> have multiple select-for-update-limit queries on the same rows,
> why are rows selected by the blocked query before knowing if
> some of those rows will be removed/eliminated by a preceding
> select-for-update-limit?
This is how it is just because of "historical reasons", i.e. it's a
known misfeature that no one has bothered to fix.
--
Alvaro Herrera http://www.CommandPrompt.com/
PostgreSQL Replication, Consulting, Custom Development, 24x7 support