From: | Bruce Momjian <bruce(at)momjian(dot)us> |
---|---|
To: | Jeff Davis <pgsql(at)j-davis(dot)com> |
Cc: | Michael Paquier <michael(at)paquier(dot)xyz>, Brar Piening <brar(at)gmx(dot)de>, pgsql-hackers(at)lists(dot)postgresql(dot)org, Andres Freund <andres(at)anarazel(dot)de> |
Subject: | Re: Minor documentation error regarding streaming replication protocol |
Date: | 2020-12-03 18:14:56 |
Message-ID: | 20201203181456.GK20285@momjian.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
On Thu, Dec 3, 2020 at 09:04:21AM -0800, Jeff Davis wrote:
> On Wed, 2020-12-02 at 15:16 -0500, Bruce Momjian wrote:
> > Yes, we could, but I thought the format code was not something we set
> > at
> > this level. Looking at byteasend() it is true it just sends the
> > bytes.
>
> It can be set along with the type. Attached an example.
>
> Andres objected (in a separate conversation) to forcing a binary-format
> value on a client that didn't ask for one. He suggested that we mandate
> that the data is ASCII-only (for both filename and content), closing
> the gap Michael raised[1]; and then just declare all values to be text
> format.
How do we mandate that? Just mention it in the docs and C comments?
> I am fine with either approach; but in any case, I don't see the point
> in sending an incorrect RowDescription.
Yeah, I can see that argument, particularly since you are setting binary
for the entire row, which in this case is valid, but still, kind of odd.
--
Bruce Momjian <bruce(at)momjian(dot)us> https://momjian.us
EnterpriseDB https://enterprisedb.com
The usefulness of a cup is in its emptiness, Bruce Lee
From | Date | Subject | |
---|---|---|---|
Next Message | Alexander Korotkov | 2020-12-03 19:03:03 | Re: Improving spin-lock implementation on ARM. |
Previous Message | Fujii Masao | 2020-12-03 17:53:59 | Re: Add Information during standby recovery conflicts |