On Tue, Oct 2, 2012 at 9:18 AM, Merlin Moncure
<mmoncure@gmail.com> wrote:
> Yes but that is irrelevant to the discussion. I am comparing the speed of
> repeated table existence checks with the speed of repeated exception blocks
> that access said table.
Both approaches have to do a catalog scan (even if you've established
an exception block the server still internally has to do a catalog
scan in order to raise an appropriate error). The exception block has
the additional overhead of a subtransaction.
OK this makes sense. Thanks :-) I will go for the CREATE TABLE approach.
--
Moshe Jacobson
Nead Werx, Inc. | Senior Systems Engineer