From: Tom Lane <tgl@sss.pgh.pa.us>
To: =?UTF-8?Q?J=c3=bcrgen_Purtz?= <juergen@purtz.de>
cc: pgsql-sql@lists.postgresql.org
Subject: Re: SQL conformity regarding SQLSTATE
In-reply-to: <c5fb7c99-0f02-75cf-9b97-95693fc6adc7@purtz.de>
References: <c2c1a96d-dae9-6667-bf11-eb423ee90385@purtz.de>
 <31999.1513697414@sss.pgh.pa.us>
 <c5fb7c99-0f02-75cf-9b97-95693fc6adc7@purtz.de>
Comments: In-reply-to =?UTF-8?Q?J=c3=bcrgen_Purtz?= <juergen@purtz.de>
	message dated "Wed, 20 Dec 2017 11:08:20 +0100"
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-ID: <11557.1513819695.1@sss.pgh.pa.us>
Date: Wed, 20 Dec 2017 20:28:15 -0500
Message-ID: <11558.1513819695@sss.pgh.pa.us>
Precedence: bulk

=?UTF-8?Q?J=c3=bcrgen_Purtz?= <juergen@purtz.de> writes:
> Summary: 01008, 03000, 0B000, 39001, F0000, and F0001 do not conform to 
> the standard.

I poked around in SQL:2011 and I concur that the first four of those
no longer appear in the standard.  However, unless grep is failing me,
we aren't generating those errcodes anywhere either:

ERRCODE_WARNING_IMPLICIT_ZERO_BIT_PADDING
ERRCODE_SQL_STATEMENT_NOT_YET_COMPLETE
ERRCODE_INVALID_TRANSACTION_INITIATION
ERRCODE_E_R_I_E_INVALID_SQLSTATE_RETURNED

So we could just remove those codes and be no worse off.

As for the other two, ERRCODE_CONFIG_FILE_ERROR and
ERRCODE_LOCK_FILE_EXISTS, we certainly are using those, but
as I mentioned it seems somewhat unlikely that clients are
testing for them.  I'm tempted to propose renumbering them
as PF000 and PF001.

			regards, tom lane