Re: PGAPI_ProcedureColumns and no support for table output arguments

From: "Matej(dot)PgODBC(at)Neosys" <pgodbc(at)matej(dot)neosys(dot)si>
To: "Inoue, Hiroshi" <h-inoue(at)dream(dot)email(dot)ne(dot)jp>
Cc: <pgsql-odbc(at)postgresql(dot)org>
Subject: Re: PGAPI_ProcedureColumns and no support for table output arguments
Date: 2017-04-21 06:53:03
Message-ID: 20170421085303.354f6dc7d07a9dc333c49245@matej.neosys.si
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-odbc

On Wed, 19 Apr 2017 15:35:23 +0900
"Inoue, Hiroshi" <h-inoue(at)dream(dot)email(dot)ne(dot)jp> wrote:

> Hi,
>
> On 2017/04/13 12:08, Matej(dot)PgODBC(at)Neosys wrote:
> > Hi Everyone,
> >
> > While debugging the driver, I've noticed that there is no support for table output arguments (proargmodes = 't').
> >
> > Is this intentional or is it a simple omission or even a bug?
>
> When SQLProcedureColumns() was implemented, table output arguments did
> not exist.
> Unfortunately I've not known table output arguments.
> Is it OK to interpret them as OUT parameters?

Good morning Hiroshi,

I think the appropriate mapping for the table output columns is SQL_RESULT_COL.

I am slightly confused why an additional column with no name and type SQL_RESULT_COL is being reported with procedures returning table results. I would expect an additional SQL_RETURN_VALUE field, but what I am seeing is like a duplication of one, perhaps the last one, of the output table's columns.

We are still in the process of validating this change. The first results with our application are positive, but we haven't made any systematic tests.

Regards,
Matej.
>
> regards,
> Hiroshi Inoue

--
Matej(dot)PgODBC(at)Neosys <pgodbc(at)matej(dot)neosys(dot)si>

In response to

Responses

Browse pgsql-odbc by date

  From Date Subject
Next Message Inoue, Hiroshi 2017-04-21 13:30:14 Re: PGAPI_ProcedureColumns and no support for table output arguments
Previous Message Adrian Klaver 2017-04-20 14:28:08 Re: Ynt: Ynt: Server Closed the connection unexpectedly