Hi,
On Thu, Nov 06, 2025 at 10:06:13AM +0100, Álvaro Herrera wrote:
> On 2025-Nov-06, Bertrand Drouvot wrote:
>
> > Subject: [PATCH v5 1/4] Introduce XLogRecPtrIsValid() and replace
> > XLogRecPtrIsInvalid() calls
>
> > XLogRecPtrIsInvalid() is inconsistent with the affirmative form of other
> > *IsValid() macros and leads to awkward double negative.
> >
> > This commit introduces XLogRecPtrIsValid() and replace all the
> > XLogRecPtrIsInvalid() calls.
> >
> > It also adds a comment mentioning that new code should use XLogRecPtrIsValid()
> > instead of XLogRecPtrIsInvalid() and that XLogRecPtrIsInvalid() could be
> > deprecated in the future.
>
> I think we should do this in two steps. First, introduce
> XLogRecPtrIsValid(), don't use it anywhere, backpatch this one. This
> would alleviate potential backpatching pains when using the new macro in
> future bugfixes.
I see, I would have introduced XLogRecPtrIsInvalid() on the back branches only
if there is a need to (a bugfix that would make use of it). But yeah, I agree
that would add extra "unnecessary" work, so done as you suggested in the
attached. I checked that 0001 apply on the [14-18]_STABLE branches successfully.
> The uppercase name looks a bit ugly. We use lowercase for other uses of
> __attribute__, e.g. pg_attribute_aligned(). Also, probably add
> "attribute" to the name, for consistency with those.
Right, replaced by pg_attribute_deprecated() in the attached.
Regards,
--
Bertrand Drouvot
PostgreSQL Contributors Team
RDS Open Source Databases
Amazon Web Services: https://aws.amazon.com
