Mailing-List: contact commons-user-help@jakarta.apache.org; run by ezmlm
Precedence: bulk
Reply-To: "Jakarta Commons Users List" <commons-user@jakarta.apache.org>
From: "James Carman" <james@carmanconsulting.com>
To: "'Jakarta Commons Users List'" <commons-user@jakarta.apache.org>
Subject: RE: Digester and Threads Question
Date: Wed, 24 Mar 2004 12:27:04 -0500
Organization: Carman Consulting, Inc.
Message-ID: <000d01c411c5$3d917970$7153058f@carmani600m>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
In-Reply-To: <4061C2C5.8050309@ignos.com>
Importance: Normal

Or you could just use a pool (Jakarta Commons Pool).  You would have to
implement your own PoolableObjectFactory which basically creates/manages
your Digester instances.

-----Original Message-----
From: Adrian Perez Jorge [mailto:aperez@ignos.com]=20
Sent: Wednesday, March 24, 2004 12:18 PM
To: Jakarta Commons Users List
Subject: Re: Digester and Threads Question


Hi,

  What would be the nice way to use a digester per thread?  Suppose the=20
number of threads is bounded, so a digester could be created to serve=20
each thread. The idea is to have a multiton and get from it the right=20
digester for the running thread.

  To build the multiton load once the rules for the digester (well..I=20
use xmlrules, so this process is somewhat expensive), then clone the=20
digester in a lazy way (i.e. as threads are created).

  Will just a Hashtable do the job where the keys are Threads objects?=20
What do you think?

Cheers,

Adrian P.J.

James Carman wrote:

>As long as you use a new Digester each time (like as a local variable
inside
>a method), you'll be fine.  What that means is that you CANNOT use one
>single digester instance in multiple threads at the same time.
>
>-----Original Message-----
>From: Alan Pocklington [mailto:ajp@pocko.freeserve.co.uk]=20
>Sent: Wednesday, March 24, 2004 11:24 AM
>To: commons-user@jakarta.apache.org
>Subject: Digester and Threads Question
>
>Hi,
>
>I want to use the Digester in a web application to parse XML documents =
on=20
>the fly.  Basically, the user logs in and his corresponding XML file is =

>loaded from the database, parsed using Digester and displayed to the =
user=20
>for manipulation.
>
>I'm concerned about the following line in the JavaDoc:
>
>"A single Digester instance may only be used within the context of a =
single

>thread at a time, and a call to parse() must be completed before =
another=20
>can be initiated even from the same thread."
>
>I'm not sure if I fully understand what this is saying. Will this mean =
I=20
>will face problems using the Digester in the above way in a web=20
>application?  Is it saying that I can use only one digester instance to =

>service all requests, or will it be okay to put Digester code in my =
servlet
>
>which will create a Digester for each request?
>
>Please help, I'm confused.
>
>Thanks in advance,
>
>Alan.
>
> =20
>

---------------------------------------------------------------------
To unsubscribe, e-mail: commons-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-user-help@jakarta.apache.org


---------------------------------------------------------------------
To unsubscribe, e-mail: commons-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-user-help@jakarta.apache.org