On Tue, Mar 19, 2024, at 8:57 AM, Peter Eisentraut wrote:
On 19.03.24 12:26, Shlok Kyal wrote:
>
> I have added a top-up patch v30-0003. The issue in [1] still exists in
> the v30 patch. I was not able to come up with an approach to handle it
> in the code, so I have added it to the documentation in the warning
> section. Thoughts?
Seems acceptable to me. pg_createsubscriber will probably always have
some restrictions and unsupported edge cases like that. We can't
support everything, so documenting is ok.
Shlok, I'm not sure we should add a sentence about a pilot error. I added a
comment in check_subscriber that describes this situation. I think the comment
is sufficient to understand the limitation and, if it is possible in the future,
a check might be added for it. I didn't include v31-0004.