Re: regclass without error?

Tatsuo Ishii <ishii@postgresql.org> writes:
>> It's not generally safe to suppress errors like that. You could leak
>> locks or tuple descriptors etc. And if the error is not "no scuh
>> table", but e.g. out of memory, you don't want to suppress it anyway.

> Thanks. I will create more "invasive" patch.

Why is any of this necessary?  It sure looks like you are solving a
problem at the wrong level.
        regards, tom lane