"Jeroen T. Vermeulen" <jtv@xs4all.nl> writes:
> On Mon, Oct 20, 2003 at 03:39:31PM -0400, Tom Lane wrote:
>> Whether the other approach would be any more future-proof is not clear
>> either.
> How standardized is row locking anyway?
Not at all, which is one reason I think that you might be fooling
yourself to imagine that a solution dependent on row locking is more
future-proof than one that depends explicitly on some Postgres-specific
behavior. The pg_locks approach is at least likely to be more obvious
about it if it breaks ...
regards, tom lane