On Thu, 4 Jan 2024 at 00:01, Richard Guo <guofenglinux@gmail.com> wrote:
> As a clause's serial number is unique within the current PlannerInfo
> context, and multiple clones of the same clause have the same serial
> number, it seems to me that it's more correct to calculate the length of
> ppi_clauses by:
>
> bms_num_members(inner_path->param_info->ppi_serials)
>
> So I think maybe we can fix this issue with the attached patch.
Thanks for the patch. I looked at this again and believe the
bms_num_member() check is a good fix.
I considered if Amadeo's case is worth including as a test and I
concluded that having that as a test won't necessarily guard us
against getting this error again if there are still bugs in this area.
David