From: | Alena Rybakina <lena(dot)ribackina(at)yandex(dot)ru> |
---|---|
To: | Richard Guo <guofenglinux(at)gmail(dot)com> |
Cc: | PostgreSQL-development <pgsql-hackers(at)postgresql(dot)org> |
Subject: | Re: Simplify create_merge_append_path a bit for clarity |
Date: | 2023-10-24 10:00:13 |
Message-ID: | 0cf04512-016f-4fcc-b3f9-cefda0390a19@yandex.ru |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
Hi!
On 11.08.2023 05:31, Richard Guo wrote:
> As explained in the comments for generate_orderedappend_paths, we don't
> currently support parameterized MergeAppend paths, and it doesn't seem
> like going to change anytime soon. Based on that, we could simplify
> create_merge_append_path a bit, such as set param_info to NULL directly
> rather than call get_appendrel_parampathinfo() for it. We already have
> an Assert on that in create_merge_append_plan.
>
> I understand that the change would not make any difference for
> performance, it's just for clarity's sake.
I agree with you, and we can indeed directly set the param_info value to
NULL, and there are enough comments here to explain.
I didn't find anything else to add in your patch.
--
Regards,
Alena Rybakina
From | Date | Subject | |
---|---|---|---|
Next Message | Drouvot, Bertrand | 2023-10-24 10:05:09 | Re: Synchronizing slots from primary to standby |
Previous Message | PG Bug reporting form | 2023-10-24 09:42:28 | BUG #18167: cannot create partitioned tables when default_tablespace is set |