Date: Mon, 18 Nov 2002 16:39:08 +0000
From: Tim Bunce <Tim.Bunce@pobox.com>
To: Tom Lane <tgl@sss.pgh.pa.us>
Cc: David Wheeler <david@wheeler.net>, dbi-dev@perl.org,
	pgsql-interfaces@postgresql.org
Subject: Re: DBD::PostgreSQL
Message-ID: <20021118163908.GD1444@dansat.data-plan.com>
References: <0907BA9A-FB11-11D6-93B3-0003931A964A@wheeler.net>
	<29660.1037636365@sss.pgh.pa.us>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <29660.1037636365@sss.pgh.pa.us>
User-Agent: Mutt/1.4i

On Mon, Nov 18, 2002 at 11:19:25AM -0500, Tom Lane wrote:
> David Wheeler <david@wheeler.net> writes:
> > On Sunday, November 17, 2002, at 10:15  PM, Tom Lane wrote:
> >> Both of these seem pretty bogus to me.  Ideally the driver should not
> >> issue a "begin" until the application issues the first command of the
> >> new transaction.  Otherwise you get into scenarios where idle
> >> connections are holding open transactions, and ain't nobody gonna be
> >> happy with that.
> 
> > Okay. I think I'll use a flag in the driver to track when it's in a 
> > transaction, and do the right thing in the begin and rollback functions.
> 
> I think someone else said that the DBD framework already includes such a
> flag ("BegunWork"?) --- if so, you should surely use that one.

BegunWork _only_ relates to the begin_work method. It's not used unless
that method is used, so it's not appropriate for your use here.

Just add a flag to the drivers private structure.

Tim.