Horton

Re: Avoid incomplete copy string (src/backend/access/transam/xlog.c)

Daniel Gustafsson <daniel@yesql.se>

View thread
From: Daniel Gustafsson <daniel@yesql.se>
To: Michael Paquier <michael@paquier.xyz>
Cc: Alvaro Herrera <alvherre@alvh.no-ip.org>, Ranier Vilela <ranier.vf@gmail.com>, Yugo NAGATA <nagata@sraoss.co.jp>, Richard Guo <guofenglinux@gmail.com>, PostgreSQL Hackers <pgsql-hackers@postgresql.org>
Date: 2024-07-02T09:44:07Z
Lists: pgsql-hackers

Commits

Same data as JSON: GET /api/v1/messages/:b64id/commits the thread's linked commits as JSON, with link sources. API reference →

  1. Use safe string copy routine

> On 2 Jul 2024, at 02:33, Michael Paquier <michael@paquier.xyz> wrote:
> 
> On Mon, Jul 01, 2024 at 09:19:59PM +0200, Daniel Gustafsson wrote:
>>> The bit I don't understand about this discussion is what will happen
>>> with users that currently have exactly 1024 chars in backup names today.
>>> With this change, we'll be truncating their names to 1023 chars instead.
>>> Why would they feel that such change is welcome?
>> 
>> That's precisely what I was getting at.  Maybe it makes sense to change, maybe
>> not, but that's not for this patch to decide as that's a different discussion
>> from using safe string copying API's.
> 
> Yep.  Agreed to keep backward-compatibility here, even if I suspect
> there is close to nobody relying on backup label names of this size.

I suspect so too, and it might be a good project for someone to go over such
buffers to see if there is reason grow or contract.  Either way, pushed the
strlcpy part.

--
Daniel Gustafsson