Thread: Patch for handling "autocommit=false" in postgresql.conf

Hello,
   This is patch for handle JDBC for autocommit=false in postgresql.conf

Patch set autocommit mode to true (on) for JDBC compliant.
--- AbstractJdbc1Connection.java.orig   Tue Sep 17 02:23:49 2002
+++ AbstractJdbc1Connection.java        Tue Sep 17 02:27:52 2002
@@ -366,6 +366,11 @@
                // Initialise object handling
                initObjectTypes();

+               // handle autocommit=false in postgresql.conf
+                if (haveMinimumServerVersion("7.3")) {
+                               ExecSQL("set autocommit to on; commit;");
+               }
+
                // Mark the connection as ok, and cleanup
                PG_STATUS = CONNECTION_OK;
        }
regards
Haris Peco

snpe <snpe@snpe.co.yu> writes:
> +               // handle autocommit=false in postgresql.conf
> +                if (haveMinimumServerVersion("7.3")) {
> +                               ExecSQL("set autocommit to on; commit;");
> +               }

The above will fill people's logs with
    WARNING:  COMMIT: no transaction in progress
if they don't have autocommit off.

Use
    begin; set autocommit to on; commit;
instead.

I would recommend holding off on this patch altogether, actually,
until we decide whether SET will be a transaction-initiating
command or not.  I would still like to persuade the hackers community
that it should not be.

            regards, tom lane

On Tuesday 17 September 2002 04:26 pm, Tom Lane wrote:
> snpe <snpe@snpe.co.yu> writes:
> > +               // handle autocommit=false in postgresql.conf
> > +                if (haveMinimumServerVersion("7.3")) {
> > +                               ExecSQL("set autocommit to on; commit;");
> > +               }
>
> The above will fill people's logs with
>     WARNING:  COMMIT: no transaction in progress
> if they don't have autocommit off.
>
> Use
>     begin; set autocommit to on; commit;
> instead.
>
> I would recommend holding off on this patch altogether, actually,
> until we decide whether SET will be a transaction-initiating
> command or not.  I would still like to persuade the hackers community
> that it should not be.

if set 'begin; setautocommit to on; commit;'
this is work no matter autocommit on or off and whether SET initiate
transaction or not.
We can yet query database for transaction (on or off), but this is not
important (we must autocommit=true for jdbc spec).

regards
haris peco

Tom Lane wrote:
> snpe <snpe@snpe.co.yu> writes:
> > +               // handle autocommit=false in postgresql.conf
> > +                if (haveMinimumServerVersion("7.3")) {
> > +                               ExecSQL("set autocommit to on; commit;");
> > +               }
>
> The above will fill people's logs with
>     WARNING:  COMMIT: no transaction in progress
> if they don't have autocommit off.
>
> Use
>     begin; set autocommit to on; commit;
> instead.
>
> I would recommend holding off on this patch altogether, actually,
> until we decide whether SET will be a transaction-initiating
> command or not.  I would still like to persuade the hackers community
> that it should not be.

Sorry to be blocking that change but I can see no reason to special case
SET to not start a new transaction, especially since we added SET to be
rollbackable in 7.3.  If we hadn't done that, you easily could make a
case for it because the SET would behave without transaction semantics.

--
  Bruce Momjian                        |  http://candle.pha.pa.us
  pgman@candle.pha.pa.us               |  (610) 359-1001
  +  If your life is a hard drive,     |  13 Roberts Road
  +  Christ can be your backup.        |  Newtown Square, Pennsylvania 19073