Mailing-List: contact dev-help@cocoon.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@cocoon.apache.org
Received-SPF: pass (asf.osuosl.org: local policy)
From: Tilman Rassy <rassy@math.TU-Berlin.DE>
Organization: TU Berlin
To: dev@cocoon.apache.org
Subject: Re: setup() called twice in one request
Date: Fri, 20 Jan 2006 14:34:02 +0100
User-Agent: KMail/1.5.4
References: <200601191901.58102.rassy@math.tu-berlin.de>
 <200601191928.19109.rassy@math.tu-berlin.de> <43D0AEC1.6030906@apache.org>
In-Reply-To: <43D0AEC1.6030906@apache.org>
MIME-Version: 1.0
Content-Type: text/plain;
  charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
Content-Disposition: inline
Message-Id: <200601201434.02941.rassy@math.tu-berlin.de>

On Friday 20 January 2006 10:34, Carsten Ziegeler wrote:
> Tilman Rassy wrote:
> > I think we can change our code so that multiple setup() calls are not a
> > problem any longer. The reasons why I posted anyway are: If there is a
> > bug, the Cocoon developers might want to know about it. Secondly, we do a
> > lot of Cocoon programming in our project, and it would be easier for my
> > collegues and me if we knew the exact rules that apply to the setup()
> > calls.
>
> Yes, you're right: now, the contract is that setup() is only called once
> per request per component instance in the pipeline. So, if you have a
> pipeline with one generator, one transformer and one serializer, setup()
> is called once on the generator, once on the transformer and once on the
> serializer (per request for this pipeline).
> Now, if you happen to have the same transformer twice in the pipeline,
> like applying two xslts, then of course setup is called twice for the
> xslt transformer but on two different instances/objects.

Thank you very mutch for these explanations.

Unfortunately, I'm quite sure that the two setup() calls are from the same 
instance and in the same request. We can see that very clearly in the logs. I 
included an excerpt from the logs in my reply to the mail from Joerg Heinicke 
in this matter. You will find further explanations there, too. If you are 
interested in my problem further on, please look there.

Best regards,
Tilman Rassy