Mailing-List: contact dev-help@cocoon.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@cocoon.apache.org
Date: Mon, 8 Dec 2003 16:25:44 +0100
From: simon.mieth@t-online.de (Simon Mieth)
To: dev@cocoon.apache.org
Subject: Re: [CocoonBean] recycle?
Message-Id: <20031208162544.4e27707c.simon.mieth@t-online.de>
In-Reply-To: <3FD491F2.3050409@upaya.co.uk>
References: <20031208153849.4dd9b390.simon.mieth@t-online.de>
	<3FD491F2.3050409@upaya.co.uk>
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit

On Mon, 08 Dec 2003 15:00:02 +0000
Upayavira <uv@upaya.co.uk> wrote:

> Are you really sure that this.crawler = new Crawler() is
> all that is needed? If so, then that's great.
> 

In my case, i'm sure. By processing the Crawler iterates
through the targets, remove from stillNotVisited and put
they to visitedAlready. 
If i will now rebuild the pipeline i add the URI as a new
target to CocoonBean, but the Crawler ignores this, because
the uri is in "visitedAlready". 
So if  i will rebuild the pipeline i need a empty Crawler,
which now reports correct the generated pages for this
process-session.

What i do, is:

addTarget(uri1)
addTarget(uri2)
process()


//change the documents

recycle()
addTarget(uri1)
addTarget(uri2)
process()

and it works now with recycle() without create a new
cocoon-instance, as before.

In my case it looks, as it works, but not i'm sure, if it
works for all ;-) .


> Note, this would not be an implementation of an interface,
> as the CocoonBean is not an Avalon component. It would
> just be a convenience method.
> 
> I'm happy to add it. Any objections anyone?
> 
> Regards, Upayavira
> 
> >BTW: the Checksum and page-reporting stuff works fine,
> >thanks.
> >  
> >
> Great. Maybe I should document it!
> 
> Regards, Upayavira

With the page/link-reporting a have build a progressbar, so
everybody can see the page-building, thats really fine,
thanks again.


Best regards,

Simon