Re: [HACKERS] advanced partition matching algorithm for partition-wise join

On Wed, 8 Apr 2020 at 15:42, Etsuro Fujita <etsuro(dot)fujita(at)gmail(dot)com> wrote:

> Hi Kuntal,
>
> On Wed, Apr 8, 2020 at 4:30 PM Kuntal Ghosh <kuntalghosh(dot)2007(at)gmail(dot)com>
> wrote:
> > I'm getting the following warning during compilation.
> >
> > partbounds.c: In function ‘partition_bounds_merge’:
> > partbounds.c:1024:21: warning: unused variable ‘inner_binfo’
> [-Wunused-variable]
> > PartitionBoundInfo inner_binfo = inner_rel->boundinfo;
> > ^
> > For fixing the same, we can declare inner_binfo as
> > PG_USED_FOR_ASSERTS_ONLY as it is not used for any other purpose.
>
> I'd propose to remove an assertion causing this (and the
> outer_binfo/inner_binfo variables) from partition_bounds_merge(),
> rather than doing so, because the assertion is redundant, as we have
> the same assertion in merge_list_bounds() and merge_range_bounds().
> Please find attached a patch.
>

Oh, I didn't see this mail before sending my other mail.

I think it's better to have the assertion in all the three places and also
in merge_hash_bounds() whenever that comes along. The assertion in
merge_*_bounds() will be good to in case those functions are called from
places other than partition_bounds_merge(). The assertion in
partition_bounds_merge() will make sure that when the individual
merge_*_bounds() functions are called based on one of the bounds both of
the bounds have same strategy.
--
Best Wishes,
Ashutosh