From: Tom Lane <tgl@sss.pgh.pa.us>
To: Tomas Vondra <tomas@vondra.me>
cc: Jim Jones <jim.jones@uni-muenster.de>,
        Alex Liapychev <coder.sam@gmail.com>,
        matheusssilv97 <matheusssilv97@gmail.com>,
        Chao Li <li.evan.chao@gmail.com>,
        "masao.fujii" <masao.fujii@gmail.com>,
        "david.g.johnston" <david.g.johnston@gmail.com>,
        "huseyin.d3r" <huseyin.d3r@gmail.com>,
        pgsql-hackers <pgsql-hackers@lists.postgresql.org>,
        chochoforwork@gmail.com, 44973863@qq.com
Subject: Re: COMMENTS are not being copied in CREATE TABLE LIKE
In-reply-to: <6cfde153-bb97-4fa5-a7e2-6b7c2056ae5d@vondra.me>
References: <e08cb97f-0364-4002-9cda-3c16b42e4136@uni-muenster.de>
 <tencent_D0A4135EDF6E96A0D4478DABF0656D6DB408@qq.com>
 <2d41978b-3bca-4756-b0a0-b6e707ad7793@uni-muenster.de>
 <tencent_FF05EDF384FEB7852516F57BE32C7C61DB06@qq.com>
 <88b16128-d6f4-4a7b-a29a-fa0cd1c2084f@uni-muenster.de>
 <tencent_28945AA8110B3EA5AD8A366542E61BCA5205@qq.com>
 <65963A8C-C1EE-431E-9513-C4AD16AEE789@gmail.com>
 <223a6be3-3917-47a4-9b5d-51720c60ad58@vondra.me>
 <c8525e6e-5789-464e-a70d-b3916e4168eb@uni-muenster.de>
 <CAD4C109-0CAB-4D25-9C3B-FDC8CE375E4D@gmail.com>
 <36F56BBE-1499-4F82-AC03-05AAAE65D657@gmail.com>
 <6cc4ef4b-f997-4453-ab40-bc012d342fa0@uni-muenster.de>
 <6cfde153-bb97-4fa5-a7e2-6b7c2056ae5d@vondra.me>
Comments: In-reply-to Tomas Vondra <tomas@vondra.me>
	message dated "Thu, 01 Oct 2026 19:25:49 +0200"
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-ID: <1113327.1790876738.1@sss.pgh.pa.us>
Date: Thu, 01 Oct 2026 13:45:38 -0400
Message-ID: <1113328.1790876738@sss.pgh.pa.us>
Archived-At: 
 <https://www.postgresql.org/message-id/1113328.1790876738%40sss.pgh.pa.us>
Precedence: bulk

Tomas Vondra <tomas@vondra.me> writes:
> Yes, I can construct a made-up example, but I struggle to create an
> example where copying table comments would be meaningful.

> I think this lack of use case is partially due to the origin of the
> patch. If it was written because of genuine need for the feature, we'd
> have the use case by definition. But it was written "because it's
> missing" and so no use case.

Yeah, this argument is what's bugging me.  We are struggling to figure
out what the patch should do with multiple comments because we don't
have a clear use-case in mind.  Any behavior we choose might be the
wrong one for whatever usage is out there.  So I'd just as soon pass
on implementing this functionality until we have some field requests
and we can quiz those people about their needs.

			regards, tom lane