From: Tom Lane <tgl@sss.pgh.pa.us>
To: =?utf-8?B?5aec5qKm5rSL?= <minesunny@icloud.com>
cc: pgsql-bugs@lists.postgresql.org
Subject: =?UTF-8?Q?Re:_the_same_time_value_return_different_values?=
         =?UTF-8?Q?=EF=BC=8Cis_it_a_problem?=
In-reply-to: <9831C223-91ED-43FC-8223-B90C46BA77C4@icloud.com>
References: <9831C223-91ED-43FC-8223-B90C46BA77C4@icloud.com>
Comments: In-reply-to =?utf-8?B?5aec5qKm5rSL?= <minesunny@icloud.com>
	message dated "Wed, 13 Sep 2023 15:24:14 +0800"
MIME-Version: 1.0
Content-Type: text/plain; charset="UTF-8"
Content-ID: <2961694.1694614648.1@sss.pgh.pa.us>
Content-Transfer-Encoding: quoted-printable
Date: Wed, 13 Sep 2023 10:17:28 -0400
Message-ID: <2961695.1694614648@sss.pgh.pa.us>
Archived-At: 
 <https://www.postgresql.org/message-id/2961695.1694614648%40sss.pgh.pa.us>
Precedence: bulk

=3D?utf-8?B?5aec5qKm5rSL?=3D <minesunny@icloud.com> writes:
> 	I find the time type accuracy of the database. Using PreparedStatement,=
 the accuracy found out for the sixth time is different from the first fiv=
e times. This number is exactly the same as the threshold of the PreparedS=
tatement. I can understand that the database can optimize PreparedStatemen=
ts and reduce repeated parsing and other unnecessary operations. but I don=
=E2=80=99t quite understand how such processing will affect the execution =
results(Or maybe I don=E2=80=99t have a deep understanding of PreparedStat=
ement.).

If you're using JDBC, this is probably the same effect discussed at

https://www.postgresql.org/message-id/flat/CAKyg5Sj%3D85niw8%3DZhX-quHaCQT=
PhG40REi1s%2BHYF6Snz4Dfi7Q%40mail.gmail.com

It's got something to do with retrieving the query results in binary
mode instead of text.  The query result at the server is the same,
but the JDBC driver is rendering it differently.

			regards, tom lane