Re: [HACKERS] postgres_fdw bug in 9.6

(2018/01/16 12:00), Etsuro Fujita wrote:
> (2018/01/16 11:17), Tom Lane wrote:
>> Etsuro Fujita<fujita(dot)etsuro(at)lab(dot)ntt(dot)co(dot)jp> writes:
>>> (2018/01/16 1:47), Robert Haas wrote:
>>>> Hmm, I was thinking that bar and baz wouldn't be constrained to return
>>>> just one tuple in that case, but I'm wrong: there would just be one
>>>> tuple per relation in that case. However, that would also be true for
>>>> a full join, wouldn't it?
>>
>>> Consider:
>>
>>> postgres=# create table bar (a int, b text);
>>> postgres=# create table baz (a int, b text);
>>> postgres=# insert into bar values (1, 'bar');
>>> postgres=# insert into baz values (2, 'baz');
>>> postgres=# select * from bar full join baz on bar.a = baz.a;
>>> a | b | a | b
>>> ---+-----+---+-----
>>> 1 | bar | |
>>> | | 2 | baz
>>> (2 rows)
>>
>>> Both relations have one tuple, but the full join produces two join
>>> tuples. I think it would be possible that something like this happens
>>> when executing a local join plan for a foreign join that performs a full
>>> join remotely.
>>
>> Doesn't really matter though, does it? Each of those join rows will
>> be processed as a separate EPQ event.
>
> I assume that such a local join plan is executed as part of a FOR UPDATE
> query like the one shown by Robert (the bar/baz foreign join part in
> that query), so I am thinking that those join rows will be processed as
> a single event.

I realized I am wrong; the local join execution plan would never produce
multiple tuples in a single event.

Best regards,
Etsuro Fujita