MIME-Version: 1.0
References: <CA+FnnTygjVvSmzTHnvteuL4KEC=0P_+mF8Kn84JtNGZWUGxpUA@mail.gmail.com>
 <2547700.1765940917@sss.pgh.pa.us> <CA+FnnTxO6kam9=mRonvsnUA1OxJxdDdQsj+ubdgFNNb9b3WMPQ@mail.gmail.com>
 <9a0ce31a49f149d8d982d2b359095560655f781e.camel@cybertec.at>
 <CA+FnnTzE34VJWR7xkymaKEwXiiQ8j3o03m0+qcQsRr2JkRQLwQ@mail.gmail.com>
 <CAKFQuwZaf2quGaqsGAj4+rVVCebOwU3JjbMUJPxxj6NeF_O_gA@mail.gmail.com>
 <CA+FnnTwZUo-ZRZ4if_ngXXkYgtOCsmMp1uFpsMFkG5AOmRxX4Q@mail.gmail.com>
 <CAKFQuwYSM2xMxX+v8LAhJFG3kB1O-VBtV3eo3fj0ag3rGJ=ehQ@mail.gmail.com>
 <CA+FnnTzqQbBMnMyX3fz44tL0bYykDU=qcUpvT=vUy+cp8Q6vfw@mail.gmail.com>
 <699b6649-6841-48a7-8182-b92d28a70ff7@aklaver.com> <138367.1766337436@sss.pgh.pa.us>
In-Reply-To: <138367.1766337436@sss.pgh.pa.us>
From: Igor Korot <ikorot01@gmail.com>
Date: Sun, 21 Dec 2025 11:44:49 -0800
Message-ID: <CA+FnnTxCCoA+ju+-HAde8t8kmQt6Qm58oP5z2U2eZGHr7qAf6Q@mail.gmail.com>
Subject: Re: PQexecPrepared() question
To: Tom Lane <tgl@sss.pgh.pa.us>
Cc: Adrian Klaver <adrian.klaver@aklaver.com>, 
	"David G. Johnston" <david.g.johnston@gmail.com>, Laurenz Albe <laurenz.albe@cybertec.at>, 
	"pgsql-generallists.postgresql.org" <pgsql-general@lists.postgresql.org>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Archived-At: <https://www.postgresql.org/message-id/CA%2BFnnTxCCoA%2Bju%2B-HAde8t8kmQt6Qm58oP5z2U2eZGHr7qAf6Q%40mail.gmail.com>
Precedence: bulk

Hi, Tom,

On Sun, Dec 21, 2025 at 9:17=E2=80=AFAM Tom Lane <tgl@sss.pgh.pa.us> wrote:
>
> Adrian Klaver <adrian.klaver@aklaver.com> writes:
> > It is not the same setup, in your original question you are hand writin=
g
> > code not running a script through the ODBC driver. Best bet is the ODBC
> > driver is picking up the client(Windows) encoding and passing it to the
> > server.
>
> It might be the other way around.  psql absolutely will try to infer
> an encoding from its environment, but perhaps the ODBC driver does
> not, or does it differently.
>
> In any case, "SHOW client_encoding" in both the working and
> non-working contexts would yield useful information.

Adding following code:

        SQLHSTMT stmt;
        SQLWCHAR value[25];
        ret =3D SQLAllocHandle( SQL_HANDLE_STMT, m_hdbc, &stmt );
        ret =3D SQLExecDirect( stmt, L"SHOW client_encoding", SQL_NTS );
        ret =3D SQLBindCol( stmt, 1, SQL_C_WCHAR, &value, 25, 0 );
        ret =3D SQLFetch( stmt );
        ret =3D SQLFreeHandle( SQL_HANDLE_STMT, stmt );

the value of the "value" variable is still "UTF8".

Thank you.

>
>                         regards, tom lane