Csaba Nagy <[EMAIL PROTECTED]> writes:
> I DO NOT CARE about which rows are deleted.
You can't possibly think that that holds true in general.
> The fact that it is
> nondeterministic can be very clearly specified in the documentation if
> you think it is such a bad thing, but nondeterministic is perfectly
> fine sometimes. There are lots of nondeterminisms in the data base
> world, starting with the ordering of selects if you don't use order
> by, then why don't we force everybody using order by ? Why don't you
> force to use order by on a select with limit ? Why there it is enough
> to say it in the docs that it WILL BE NON_DETERMINISTIC ?
I can tolerate nondeterminism in SELECT because it doesn't change the
data. If you get it wrong you can always do it over. UPDATE/DELETE
need to have higher standards though.
regards, tom lane
---------------------------(end of broadcast)---------------------------
TIP 2: Don't 'kill -9' the postmaster
