From: Tom Lane <tgl@sss.pgh.pa.us>
To: Kyotaro Horiguchi <horikyota.ntt@gmail.com>
cc: jeevan.chalke@enterprisedb.com, andrew@dunslane.net, peter@eisentraut.org,
        pgsql-hackers@lists.postgresql.org
Subject: Re: More new SQL/JSON item methods
In-reply-to: <20240129.121200.235012930453045390.horikyota.ntt@gmail.com>
References: <451195.1706216300@sss.pgh.pa.us>
 <65ac43de-8abe-f9ef-1fe8-838ea22e9c45@dunslane.net>
 <CAM2+6=V1euSMRnYPv7s7JCndFk_3rfmaSny6So2+VJs2Xxd6Hw@mail.gmail.com>
 <20240129.121200.235012930453045390.horikyota.ntt@gmail.com>
Comments: In-reply-to Kyotaro Horiguchi <horikyota.ntt@gmail.com>
	message dated "Mon, 29 Jan 2024 12:12:00 +0900"
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-ID: <1678191.1706500022.1@sss.pgh.pa.us>
Content-Transfer-Encoding: quoted-printable
Date: Sun, 28 Jan 2024 22:47:02 -0500
Message-ID: <1678192.1706500022@sss.pgh.pa.us>
Archived-At: 
 <https://www.postgresql.org/message-id/1678192.1706500022%40sss.pgh.pa.us>
Precedence: bulk

Kyotaro Horiguchi <horikyota.ntt@gmail.com> writes:
> I have two possible issues in a recent commit.
> Commit 66ea94e8e6 has introduced the following messages:

>> errmsg("numeric argument of jsonpath item method .%s() is out of range =
for type decimal or number",

> They seem to be suggesting that PostgreSQL has the types "decimal" and
> "number". I know of the former, but I don't think PostgreSQL has the
>  latter type. Perhaps the "number" was intended to refer to "numeric"?

Probably.  But I would write just "type numeric".  We do not generally
acknowledge "decimal" as a separate type, because for us it's only an
alias for numeric (there is not a pg_type entry for it).

Also, that leads to the thought that "numeric argument ... is out of
range for type numeric" seems either redundant or contradictory
depending on how you look at it.  So I suggest wording like

argument "...input string here..." of jsonpath item method .%s() is out of=
 range for type numeric

> (And I think it is largely helpful if the given string were shown in
> the error message, but it would be another issue.)

Agreed, so I suggest the above.

> The same commit has introduced the following set of messages:

>> %s format is not recognized: "%s"
>> date format is not recognized: "%s"
>> time format is not recognized: "%s"
>> time_tz format is not recognized: "%s"
>> timestamp format is not recognized: "%s"
>> timestamp_tz format is not recognized: "%s"

> I believe that the first line was intended to cover all the others:p

+1

			regards, tom lane