On Sat, 2009-01-24 at 19:45 +0000, Greg Stark wrote:
> There already is quite an extensive discussion of how FOR UPDATE
> behaves including these kinds of violations.
Not in the documentation, that I can see. And I think it's important
that it be there for the reasons I mentioned.
Can you refer me to the dicussion that you're talking about? I don't
remember any discussion that points out that FOR UPDATE/FOR SHARE is
broken in the simple case of a simple WHERE clause.
> What you propose is interesting though. It would have been impossible
> before subtransactions but it's doable now. Still the performance
> might be unusable for complex queries. It's basically generalizing the
> logic a serializable transaction would take to a read committed command.
It might be effective for queries that are highly selective on large
tables. Still has strange deadlock possibilities, but I think that's the
case already.
Regards,Jeff Davis