On Tue, Aug 4, 2015 at 9:12 AM, Fabrízio de Royes Mello
<fabriziomello@gmail.com> wrote:
> Are you sure we need to do all this changes just to check the highest
> locklevel based on the reloptions?
Well, by looking at the code that's what it looks as. That's a large
change not that straight-forward.
> Or I misunderstood the whole thing?
No I think we are on the same page.
> Maybe we just check the DoLockModeConflict in the new method
> GetReloptionsLockLevel and if true then fallback to AccessExclusiveLock (as
> Michael suggested in a previous email).
Honestly that's what I would suggest for this patch, and also fix the
existing problem of tablecmds.c that does the same assumption. It
seems saner to me for now than adding a whole new level of routines
and wrappers, and your patch has IMO great value when each ALTER
COMMAND is kicked individually.
--
Michael