Message-ID: <b1e217e4-0527-48ba-9d8a-74106d47f216@oss.nttdata.com>
Date: Fri, 11 Jul 2025 09:12:02 +0900
MIME-Version: 1.0
User-Agent: Mozilla Thunderbird
Subject: Re: pg_dump --with-* options
Content-Language: en-US
To: Jeff Davis <pgsql@j-davis.com>, Robert Haas <robertmhaas@gmail.com>,
 Nathan Bossart <nathandbossart@gmail.com>
Cc: Peter Eisentraut <peter@eisentraut.org>,
 pgsql-hackers <pgsql-hackers@postgresql.org>
References: <7cc52488-f876-4ad3-affd-6e4b0ef0cb09@eisentraut.org>
 <aEMLzPQtjRbj2R7X@nathan>
 <CA+TgmoactJEUHeXuxHLKNXHzm6-XrQFigPur-4nHTMFSiSRkiQ@mail.gmail.com>
 <aErpuM6RXSlK1FBW@nathan>
 <7bf0246542713d80f12c738c1d365d1365750c18.camel@j-davis.com>
 <5e398522803da9bf3f9fc33ca43084ec21ab0fcd.camel@j-davis.com>
 <aFLenfMRS9jFINxq@nathan>
 <031558c60e84362898922caa6a90587e7fdf2a57.camel@j-davis.com>
 <CA+TgmoYW4ycbzZehdS9nOw60ZdtwzWZ-qoVpBe3+QCpeBxvKug@mail.gmail.com>
 <aFrWvlsWwE92Mcpr@nathan>
 <CA+TgmoYuP7TvwThJERO2X7CHJCrTMpiv=Fd=EPOcyWwL=8J4Hg@mail.gmail.com>
 <3cc4f166-0b58-47e2-a03c-7c2873d62b14@oss.nttdata.com>
 <d64d902780ab6f95ea30093b4985d097d7b29605.camel@j-davis.com>
From: Fujii Masao <masao.fujii@oss.nttdata.com>
In-Reply-To: <d64d902780ab6f95ea30093b4985d097d7b29605.camel@j-davis.com>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: quoted-printable
Archived-At: 
 <https://www.postgresql.org/message-id/b1e217e4-0527-48ba-9d8a-74106d47f216%40oss.nttdata.com>
Precedence: bulk


On 2025/07/11 2:57, Jeff Davis wrote:
> On Wed, 2025-06-25 at 08:18 +0900, Fujii Masao wrote:
>> For pg_dump and pg_dumpall, I agree with Jeff's idea in [1],
>> but if the statistics is skipped by default, I don't think
>> we need a --no-statistics option. So, here's how I think
>> the options should work:
>>
>>  =C2=A0=C2=A0=C2=A0=C2=A0 * Keep: --schema-only, --data-only, --statis=
tics-only, --no-
>> schema, --no-data, -and -statistics
>>  =C2=A0=C2=A0=C2=A0=C2=A0 * Remove: --no-statistics, --with-schema, an=
d --with-data
>=20
> ...
>=20
>> For pg_restore, I believe there's agreement to restore statistics
>> by default if they exist in the archive. So:
>>
>>  =C2=A0=C2=A0=C2=A0=C2=A0 * Keep: --schema-only, --data-only, --statis=
tics-only, --no-
>> schema, --no-data, and --no-statistics
>>  =C2=A0=C2=A0=C2=A0=C2=A0 * Remove: --with-schema, --with-data, and --=
statistics
>=20
> That means pg_dump will accept --statistics and reject --no-statistics;
> and pg_restore will accept --no-statistics and reject --statistics.
> Other options are mostly the same between them, so I'm not sure it's a
> good idea for them to diverge.

I agree it would be better to have the same options in both pg_dump
and pg_restore, if possible.

But to do that, we'd either need to make pg_dump dump statistics
by default, or allow redundant options like --statistics in pg_restore,
even though it already restores statistics by default.

As I understand it, the rough consensus so far is that we'd prefer to
avoid both of these approaches. I know some want to change the default
behavior about statistics in pg_dump, though.

But, are you suggesting we go with one of them?

Regards,

--=20
Fujii Masao
NTT DATA Japan Corporation