Thread

Re: Improve docs for n_distinct_inherited

David G. Johnston <david.g.johnston@gmail.com> — 2025-10-13T11:55:36Z

On Monday, October 13, 2025, David Rowley <dgrowleyml@gmail.com> wrote:

> On Mon, 13 Oct 2025 at 17:09, David G. Johnston
> <david.g.johnston@gmail.com> wrote:
> > How about:
> >
> > n_distinct_inherited exists to allow the distinct value estimate to be
> overwritten for the children-included statistics gathered for inheritance
> parent tables, and for partitioned tables.
> >
> > I can live with “ordinarily” with that wording, it is just the negative
> of “children-included”.
>
> I've ended up reverting the text back to become more similar to how it
> was but explicitly adding ", and for the statistics gathered for
> partitioned tables.", since that's what the original bug report was
> confused about.
>
> I've kept the other document changes as they were in the previous patch.
>

WFM.  Since we allow 0 as an input I’m not convinced we should remove that
sentence, though I concur that using reset is preferred.

David J.