From: Tom Lane <tgl@sss.pgh.pa.us>
To: Andrew Dunstan <andrew@dunslane.net>
cc: "Dean Gibson (DB Administrator)" <postgresql@mailpen.com>,
        pgsql-performance@postgresql.org
Subject: Re: pg_dumpall --exclude-database case folding,
 was Re: AWS forcing PG upgrade from v9.6 a disaster
In-reply-to: <b468416e-645c-d9fd-c415-05884d6535bd@dunslane.net>
References: <88dec70b-eb63-0b72-0b5d-253851ef127b@mailpen.com>
 <92cc8563-195f-8727-d887-5f2bb63f231f@mailpen.com>
 <20210601041635.GQ2082@telsasoft.com>
 <a38c45cf-3bbd-b43a-0f8a-11539c1c6dc1@mailpen.com>
 <a7e9da9d-7433-ee02-9e35-db6b19acdcd4@mailpen.com>
 <18f77a0c-dade-b86a-cbeb-6f85d3d137e1@dunslane.net>
 <ec18a226-6bef-7ba6-e176-c751e189270d@mailpen.com>
 <CAEudQAodj-ywxc-3fgh8wK7oP8WXCoNSs0j8EF4HK+uUixTqMA@mail.gmail.com>
 <4abbbebd-1496-7376-136f-38eb91773cd2@mailpen.com>
 <1682562.1623348017@sss.pgh.pa.us>
 <a1916786-21e4-e336-47d6-71057a45ded0@dunslane.net>
 <b468416e-645c-d9fd-c415-05884d6535bd@dunslane.net>
Comments: In-reply-to Andrew Dunstan <andrew@dunslane.net>
	message dated "Mon, 14 Jun 2021 09:21:30 -0400"
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-ID: <197325.1623677541.1@sss.pgh.pa.us>
Date: Mon, 14 Jun 2021 09:32:21 -0400
Message-ID: <197326.1623677541@sss.pgh.pa.us>
Archived-At: 
 <https://www.postgresql.org/message-id/197326.1623677541%40sss.pgh.pa.us>
Precedence: bulk

Andrew Dunstan <andrew@dunslane.net> writes:
> On 6/10/21 2:23 PM, Andrew Dunstan wrote:
>> Ouch. That looks like a plain old bug. Let's fix it. IIRC I just used
>> the same logic that we use for pg_dump's --exclude-* options, so we need
>> to check if they have similar issues.

> Peter Eisentraut has pointed out to me that this is documented, albeit a
> bit obscurely for pg_dumpall. But it is visible on the pg_dump page.

Hmm.

> Nevertheless, it's a bit of a POLA violation as we've seen above, and
> I'd like to get it fixed, if there's agreement, both for this pg_dumpall
> option and for pg_dump's pattern matching options.

+1, but the -performance list isn't really where to hold that discussion.
Please start a thread on -hackers.

			regards, tom lane