Re: Two noncritical bugs of pg_waldump

On Thu, Jan 27, 2022 at 01:23:06PM +0900, Kyotaro Horiguchi wrote:
> So the issue there is neither EndRecPtr and ReadRecPtr always points
> to the current read LSN. The first proposal from Nathen was to use
> currRecPtr but it was a private member.  But after discussion, it
> seems to me it is (at least now) exactly what we need here so if we
> ofccially exposed the member the problem would be blown away.  Do you
> want to conitnue that?

Presumably there is a good reason for keeping it private, but if not, maybe
that is a reasonable approach to consider.  I don't think we wanted to
proceed with the approaches discussed on the old thread, anyway.

-- 
Nathan Bossart
Amazon Web Services: https://aws.amazon.com/