Alvaro Herrera <alvherre@alvh.no-ip.org> writes:
> On 2022-Sep-16, David Rowley wrote:
>> I kinda disagree with Alvaro's fix in 05fb5d661. I think indexlist is
>> the place to store these details. That commit added the following
>> comment:
>>
>> /*
>> * Ignore partitioned indexes, since they are not usable for
>> * queries.
>> */
>>
>> But neither are hypothetical indexes either, yet they're added to
>> RelOptInfo.indexlist.
>>
>> I think the patch should be changed so that the existing list is used
>> and we find another fix for the problems Alvaro fixed in 05fb5d661.
>> Unfortunately, there was no discussion marked on that commit message,
>> so it's not quite clear what the problem was. I'm unsure if there was
>> anything other than CLUSTER that was broken.
> After a bit of trawling through the archives, I found it here:
> https://www.postgresql.org/message-id/20180124162006.pmapfiznhgngwtjf%40alvherre.pgsql
> I think there was insufficient discussion and you're probably right that
> it wasn't the best fix. I don't object to finding another fix.
FWIW, I don't see any big problem with what you did. We'd need to
do something more like what David suggests if the planner ever has
a reason to consider partitioned indexes. But as long as it does
not, why expend the time to build data structures representing them?
And we'd have to add code in quite a few places to ignore them,
once they're in indexlist.
regards, tom lane
