MIME-Version: 1.0
From: Ben Horowitz <bhorowit@gmail.com>
Date: Fri, 17 Nov 2023 17:19:26 -0800
Message-ID: 
 <CAM1tkGVCRayeakBXb7ZqDTuJRFup7_kiF1S4gc8-_b99Cy0i-A@mail.gmail.com>
Subject: Mapping of timestamptz to TIMESTAMP vs. TIMESTAMP_WITH_TIMEZONE
To: pgsql-jdbc@lists.postgresql.org
Content-Type: multipart/alternative; boundary="000000000000e44673060a630b84"
Archived-At: 
 <https://www.postgresql.org/message-id/CAM1tkGVCRayeakBXb7ZqDTuJRFup7_kiF1S4gc8-_b99Cy0i-A%40mail.gmail.com>
Precedence: bulk

--000000000000e44673060a630b84
Content-Type: text/plain; charset="UTF-8"

Hi,

Apologies if this is a duplicate of an answered question. I searched in the
mailing list archives, Stack Overflow, etc., and didn't find anything.

At the line linked below in TypeInfoCache, it appears that type timestamptz
is getting mapped to the type java.sql.Types.TIMESTAMP [1] . It seems like
it should be mapped to java.sql.Types.TIMESTAMP_WITH_TIMEZONE.

TIMESTAMP_WITH_TIMEZONE was introduced in Java 8 (2014). Is the mapping of
timestamptz to TIMESTAMP intentional for reasons of backwards compatibility?

I found the behavior surprising, and also found a question on Stack
Overflow where someone else was surprised [2].

Thanks very much,
Ben

[1]
https://github.com/pgjdbc/pgjdbc/blob/65f158ed8e06ffea77ca7f3b8413847817a7831f/pgjdbc/src/main/java/org/postgresql/jdbc/TypeInfoCache.java#L106

[2]
https://stackoverflow.com/questions/60451102/for-postgresql-resultsetmetadata-getcolumntype-function-returns-same-value-ty

--000000000000e44673060a630b84
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div>Hi,</div><div><br></div><div>Apologies if this is a d=
uplicate of an answered question. I searched in the mailing list archives, =
Stack Overflow, etc., and didn&#39;t find anything.</div><div><br></div>At =
the line linked below in TypeInfoCache, it appears that type=C2=A0timestamp=
tz is getting mapped to the type java.sql.Types.TIMESTAMP [1] . It seems li=
ke it should be mapped to java.sql.Types.TIMESTAMP_WITH_TIMEZONE.<div><br><=
/div><div>TIMESTAMP_WITH_TIMEZONE was introduced in Java 8 (2014). Is the m=
apping of timestamptz to TIMESTAMP intentional for reasons of backwards com=
patibility?</div><div><br></div><div>I found the behavior surprising, and a=
lso found a question on Stack Overflow where someone else was surprised [2]=
.</div><div><br></div><div>Thanks very much,</div><div>Ben</div><div><br></=
div><div>[1]=C2=A0<a href=3D"https://github.com/pgjdbc/pgjdbc/blob/65f158ed=
8e06ffea77ca7f3b8413847817a7831f/pgjdbc/src/main/java/org/postgresql/jdbc/T=
ypeInfoCache.java#L106">https://github.com/pgjdbc/pgjdbc/blob/65f158ed8e06f=
fea77ca7f3b8413847817a7831f/pgjdbc/src/main/java/org/postgresql/jdbc/TypeIn=
foCache.java#L106</a></div><div><br></div><div>[2] <a href=3D"https://stack=
overflow.com/questions/60451102/for-postgresql-resultsetmetadata-getcolumnt=
ype-function-returns-same-value-ty">https://stackoverflow.com/questions/604=
51102/for-postgresql-resultsetmetadata-getcolumntype-function-returns-same-=
value-ty</a><br></div></div>

--000000000000e44673060a630b84--