> On Fri, May 09, 2025 at 02:35:33PM GMT, Michael Paquier wrote:
> On Fri, May 09, 2025 at 11:05:43AM +0800, Junwang Zhao wrote:
> > Why not a location and a length, it should be more natural, it
> > seems we use this convention in some existing nodes, like
> > RawStmt, InsertStmt etc.
>
> These are new concepts as of Postgres 18 (aka only on HEAD), chosen
> mainly to match with the internals of pg_stat_statements as far as I
> recall. Doing the same here would not hurt, but it may be better
> depending on the cases to rely on a start/end. I suspect that
> switching from one to the other should not change much the internal
> squashing logic.
Right, switching from start/length to start/end wouldn't change much for
squashing. I didn't have any strong reason to go with start/end from my
side, so if start/length is more aligned with other nodes, let's change
that.
