Horton

Re: Dropped and generated columns might cause wrong data on subs when REPLICA IDENTITY FULL

Amit Kapila <amit.kapila16@gmail.com>

View thread View on postgr.es
From: Amit Kapila <amit.kapila16@gmail.com>
To: Önder Kalacı <onderkalaci@gmail.com>
Cc: "shiy.fnst@fujitsu.com" <shiy.fnst@fujitsu.com>, PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
Date: 2023-03-17T00:08:50Z
Lists: pgsql-hackers 
On Thu, Mar 16, 2023 at 9:33 PM Önder Kalacı <onderkalaci@gmail.com> wrote:
>
> Hi Amit, Shi Yu,
>
> > Generated column is introduced in PG12, and I reproduced generated column problem
> in PG12~PG15.
> > For dropped column problem, I reproduced it in PG10~PG15. (Logical replication
> was introduced in PG10)
>
> So, I'm planning to split the changes into two commits. The first one fixes
> for dropped columns, and the second one adds generated columns check/test.
>
> Is that the right approach for such a case?
>

Works for me.

> >  and backpatch the fix for dropped column to PG10.
>
> Still, even the first commit fails to apply cleanly to PG12 (and below).
>
> What is the procedure here? Should I be creating multiple patches per version?
>

You can first submit the fix for dropped columns with patches till
v10. Once that is committed, then you can send the patches for
generated columns.

> Or does the committer prefer to handle the conflicts? Depending on your reply,
> I can work on the followup.
>

Thanks for working on it.

-- 
With Regards,
Amit Kapila.

Commits

  1. Ignore generated columns during apply of update/delete.

  2. Ignore dropped columns during apply of update/delete.