Tom Lane wrote:
>
> Hiroshi Inoue <Inoue(at)tpf(dot)co(dot)jp> writes:
> > Tom Lane wrote:
> >> IIRC, the SET does *not* start a transaction,
>
> > Yes but doesn't autocommit-off mode mean that
> > it implicitly begins a transaction in suitable
> > places ? For example, 'set autocommit to off;
> > declare .. cursor ..' works though it never
> > work without BEGIN under autocommit-on mode.
>
> But the DECLARE would start a transaction --- AFAIR,
Yes it's only because the behavior is useful for us.
So isn't the problem if the warning message for
'set autocommit to off;commit'
is useful or not ?
IMHO it's rather a harmful message.
regards,
Hiroshi Inoue
http://www.geocities.jp/inocchichichi/psqlodbc/