Re: Should we add GUCs to allow partition pruning to be disabled?

On 2018/04/24 6:10, Alvaro Herrera wrote:
>> BTW, while we're at it, would it also be a good idea to consider the patch
>> you had proposed, which I then posted an updated version of, to adjust the
>> documentation in ddl.sgml (in the section 5.10. Table Partitioning)
>> regarding the relationship between constraint exclusion and declarative
>> partitioning?
>
> I looked at this one. That patch has two hunks. I applied a change
> where the first hunk is, to replace constraint_exclusion with the new
> GUC -- seemed easy enough.

Looks good.

> However, the second hunk is on "5.10.4.
> Partitioning and Constraint Exclusion" which needs major editing.

Reading 5.10.4 again, I tend to agree with this.

> Not really sure how best to handle that one. For starters, I think it need
> to stop mentioning the GUC name in the title;

Hmm, "Constraint Exclusion" that's used in the title is a concept, not a
GUC, although pretty close.

> maybe rename it to
> "Partition Pruning" or some such, and then in the text explain that
> sometimes the enable_partition_pruning param is used in one case and
> constraint_exclusion in the other, and approximately what effects they
> have. I don't think it's worth going into too much detail on exactly
> how they differ, but then I'm not 100% sure of that either.

Just a thought -- How about making 5.10.4 cover partitioning based
optimizations in general? I see that a number of partitioning-based
optimizations have been developed in this release cycle, but I only see
various enable_partition* GUCs listed in config.sgml and not much else.
Although the config.sgml coverage of the new capabilities seems pretty
good, some may find their being mentioned in 5.10 Table Partitioning
helpful. Or if we don't want to hijack 5.10.4, maybe create a 5.10.5.

Thanks,
Amit