[rt.cpan.org #18387] Apparent transaction state corruption leads to silent failures

       Queue: DBD-Pg
 Ticket <URL: http://rt.cpan.org/Ticket/Display.html?id=18387 >

On Tue Mar 28 09:32:42 2006, JSIRACUSA wrote:

> I also think you're misreading the meaning of AutoCommit with respect to
> begin_work().  From the DBI docs:
> 
> http://search.cpan.org/dist/DBI/DBI.pm#begin_work
> 
> > $rc  = $dbh->begin_work   or die $dbh->errstr;
> > 
> > Enable transactions (by turning AutoCommit off) until the next call
> > to commit or rollback. After the next commit or rollback, AutoCommit
> > will automatically be turned on again.
> > 
> > If AutoCommit is already off when begin_work is called then it does
> > nothing except return an error. If the driver does not support
> > transactions then when begin_work attempts to set AutoCommit off the
> > driver will trigger a fatal error.
> 
> The whole point of begin_work() is that it begins a transaction when
> AutoCommit is on.  In light of this, look again at the example and it
> should become clear that it demonstrates obviously incorrect behavior. 

Huh. Yeah, the DBI docs make more sense there. But maybe I was
misunderstanding the DBD::Pg docs on the subject. They mention "BEGIN"
et al. (SQL statements), and not the begin_work method, which are two
different things.

If AutoCommit is on, and one does a begin_work, some other statements,
and a commit or rollback, is everything after that point auto-committed
again until the next begin_work call?