Nice. Much more maintainable IMO and quite close to what I was looking for. Thanks a lot for the suggestion, I will definitely try it/implement it right away. Still has some redundancy compared to an hypothetical nowait modifier but I think it's the very best alternative so far.
Eduardo On Thu, Sep 8, 2011 at 1:22 PM, Merlin Moncure <mmonc...@gmail.com> wrote: > On Thu, Sep 8, 2011 at 4:01 AM, Eduardo Piombino <drak...@gmail.com> > wrote: > > Hi, would it be possible to implement a nowait modifier to the update > > statement in order to tell it not to wait and raise an error -just like a > > select for update nowait would-, instead of defaulting to waiting forever > > until the lock becomes available? > > > > The lack of such a modifier nowadays forces me to do a select for update > > before every update on which I need the fastest response possible, and it > > would be great if it could be integrated into the command itself. > > > > Just an idea. > > +1 > > note you may be able to emulate this by sneaking a nolock into the > update statement in a highly circuitous fashion with something like: > update foo set v = 2 from (select 1 from foo where id = 1 for update > nowait) q where id = 1; > > merlin >
