Message-ID: <237d58a65d887376ded17a67e7b99af5936fa9a3.camel@cybertec.at>
Subject: Re: Request for official clarification on SQL parameter parsing
 changes in PostgreSQL 15 and 16
From: Laurenz Albe <laurenz.albe@cybertec.at>
To: =?UTF-8?Q?=E7=8E=8B=E3=80=80=E5=8D=9A?= <bo.wang@infortech.co.jp>, 
 "pgsql-general@lists.postgresql.org"
	 <pgsql-general@lists.postgresql.org>
Cc: =?UTF-8?Q?=E6=9D=8E=E3=80=80=E6=B5=A9?= <hao.li@infortech.co.jp>
Date: Thu, 17 Apr 2025 15:41:19 +0200
In-Reply-To: <TYCP286MB3620D4D19E3793CFA106CCBFB6BC2@TYCP286MB3620.JPNP286.PROD.OUTLOOK.COM>
References: 
	<TYCP286MB3620D4D19E3793CFA106CCBFB6BC2@TYCP286MB3620.JPNP286.PROD.OUTLOOK.COM>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
User-Agent: Evolution 3.54.3 (3.54.3-1.fc41) 
MIME-Version: 1.0
Archived-At: <https://www.postgresql.org/message-id/237d58a65d887376ded17a67e7b99af5936fa9a3.camel%40cybertec.at>
Precedence: bulk

On Thu, 2025-04-17 at 05:17 +0000, =E7=8E=8B=E3=80=80=E5=8D=9A wrote:
> We have encountered unexpected SQL parsing errors during application upgr=
ades from
> PostgreSQL 14 to 15 and 16, related to the use of JDBC-style parameter
> placeholders (`?`) in our legacy applications.
> =C2=A0
> Specifically:
> =C2=A0
> 1. In PostgreSQL 15 and later:
> =C2=A0=C2=A0 The following SQL causes a syntax error unless a space is ad=
ded after the `?`:
> =C2=A0=C2=A0=C2=A0=C2=A0 SELECT * FROM table WHERE a =3D ?AND b =3D 123;
> =C2=A0=C2=A0 =E2=86=92 Adding a space (`? AND`) resolves the issue.
> =C2=A0=C2=A0 This did not cause an error in PostgreSQL 14.
> =C2=A0
> 2. In PostgreSQL 16:
> =C2=A0=C2=A0 The following SQL causes a syntax error:
> =C2=A0=C2=A0=C2=A0=C2=A0 SELECT * FROM table WHERE a =3D ?ORDER BY b;
> =C2=A0=C2=A0 =E2=86=92 This query runs in PostgreSQL 15, but causes an er=
ror in 16 unless a space is added after `?`.
> =C2=A0
> We understand that this behavior may stem from changes to the parser to m=
ake it more
> strict and SQL-compliant. However, we were unable to find any explicit me=
ntion of these
> changes in the official release notes of PostgreSQL 15 or 16.
> =C2=A0
> Our client is asking for official documentation or a statement confirming=
 whether this
> behavior change was intentional and when it was introduced.
> =C2=A0
> Could anyone kindly point us to:
> - Relevant release note sections, commits, or discussions on this change
> - Confirmation that these are intentional parser changes (and not regress=
ions)

I'd say it is this change:

https://postgr.es/c/2549f0661bd28571d7200d6f82f752a7ee5d47e1

The mailing list discussion is here:

https://www.postgresql.org/message-id/flat/b239564c-cad0-b23e-c57e-166d883c=
b97d@enterprisedb.com

Yours,
Laurenz Albe