Peter Kovacs wrote:> That's, then, even simpler that I originally thought. The
only thing to
> be done is to make using "real" prepared statements the default
> behaviour of the PreparedStatement instances, is not it?
>
I agree with you. I would make it the default to 7.4 backends at least...
But it is up to Barry and Dave to decide.
We will probably have to keep the client side emulation around for a while.
Even if all the supported backend versions already have server side prepared
statements that implementation is recent enogh for us to have an alternative one
around, even if for comparison purposes. Besides, someone was claiming to have,
in some situations, faster results with the client side emulation. I believe
the 7.4 backend and the V3 protocol will solve that but we must be sure that is
so before removing this code (7.6 version will only have to support 7.3 and 7.4
backends so it is a possible end of the line for it).
As we have to keep the old code around, even if not the default, we will have to
fix it so nobody gets hurt.
--
Fernando Nasser
Red Hat - Toronto E-Mail: fnasser@redhat.com
2323 Yonge Street, Suite #300
Toronto, Ontario M4P 2C9