From: Tom Lane <tgl@sss.pgh.pa.us>
To: Michael Paquier <michael@paquier.xyz>
cc: Kyotaro HORIGUCHI <horiguchi.kyotaro@lab.ntt.co.jp>,
        pavel.stehule@gmail.com, a.zakirov@postgrespro.ru,
        pgsql-hackers@postgresql.org
Subject: Re: pg_get_functiondef forgets about most GUC_LIST_INPUT GUCs
In-reply-to: <20180321050103.GD2288@paquier.xyz>
References: <2637.1521501336@sss.pgh.pa.us>
 <20180320003951.GA4963@paquier.xyz> <10037.1521515233@sss.pgh.pa.us>
 <20180320.123747.43512160.horiguchi.kyotaro@lab.ntt.co.jp>
 <32403.1521566855@sss.pgh.pa.us> <20180321050103.GD2288@paquier.xyz>
Comments: In-reply-to Michael Paquier <michael@paquier.xyz>
	message dated "Wed, 21 Mar 2018 14:01:03 +0900"
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-ID: <22578.1521610823.1@sss.pgh.pa.us>
Content-Transfer-Encoding: quoted-printable
Date: Wed, 21 Mar 2018 01:40:23 -0400
Message-ID: <22579.1521610823@sss.pgh.pa.us>
Precedence: bulk

Michael Paquier <michael@paquier.xyz> writes:
> On Tue, Mar 20, 2018 at 01:27:35PM -0400, Tom Lane wrote:
>> +       if (flags & GUC_LIST_QUOTE)
>> +               elog(FATAL, "extensions cannot define GUC_LIST_QUOTE va=
riables");

> This would be better as an ereport with ERRCODE_FEATURE_NOT_SUPPORTED I
> think.  An ERROR is better in my opinion.

I don't mind making it an ereport, but I think it needs to be FATAL
for the reason stated in the comment.

>> +   record =3D find_option(name, false, WARNING);
>> +   if (record =3D=3D NULL)

> LOG instead of WARNING?

I did it like that to match the similar code in flatten_set_variable_args.
In the end it doesn't matter, because find_option only uses its elevel
argument when create_placeholders is passed as true.

			regards, tom lane