Re: InvokeObjectPostAlterHook() vs. CommandCounterIncrement()

On Sun, Jul 21, 2013 at 4:44 AM, Ants Aasma <ants(dot)aasma(at)eesti(dot)ee> wrote:
> On Jul 21, 2013 4:06 AM, "Noah Misch" <noah(at)leadboat(dot)com> wrote:
>> If these hooks will need to apply to a larger operation, I
>> think that mandates a different means to reliably expose the before/after
>> object states.
>
> I haven't checked the code to see how it would fit the API, but what about
> taking a snapshot before altering and passing this to the hook. Would there
> be other issues besides performance? If the snapshot is taken only when
> there is a hook present then the performance can be fixed later.

I had the idea of finding a way to pass either the old tuple, or
perhaps just the TID of the old tuple. Not sure if passing a snapshot
is better.

--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company