Mailing-List: contact user-help@commons.apache.org; run by ezmlm
Precedence: bulk
Reply-To: "Commons Users List" <user@commons.apache.org>
Received-SPF: pass (athena.apache.org: local policy)
DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws;
  s=s1024; d=yahoo.com;
  h=Message-ID:X-YMail-OSG:Received:X-Mailer:Date:From:Subject:To:In-Reply-To:MIME-Version:Content-Type:Content-Transfer-Encoding;
  b=2nm6aBKztE1dEOl94tfzVd6v4/Ko84sq9x09becc7CaKATKgenJ0/2AjSyhHJdcuq+MxPe2K9QET3YQZfKVccciLQVamCr5pmC+GPJmkZANU6j0blGceq5Yn0UkTpeNOJ3SPzx26BT8jLs9Zmzqbsyl+TJBurk5097uqwZETzpM=;
Message-ID: <810797.68101.qm@web63104.mail.re1.yahoo.com>
Date: Tue, 30 Nov 2010 05:56:43 -0800 (PST)
From: Adrian Crum <adrian.crum@yahoo.com>
Subject: Re: [digester] keeping parsing logic out of the data model
To: Commons Users List <user@commons.apache.org>
In-Reply-To: <AANLkTi=ZaZo5akrra2GEhkxuq1-bzuRHtdjdss_MPAOU@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: quoted-printable

You could put the parsing helper code in a separate class:=0A=0Apublic clas=
s ParserUtil {=0A=0A  public static boolean isFullDuplex(String hardwareDep=
endentString) {=0A    return hardwareDependentString.equalsIgnoreCase("FULL=
-DUPLEX");=0A  }=0A}=0A=0A-Adrian=0A=0A--- On Tue, 11/30/10, Pau Minoves <p=
au.minoves@i2cat.net> wrote:=0A> Hi all,=0A> =0A> We have some doubts here =
on how apache digester (2.1) is=0A> meant to be=0A> used. I don't think we =
are on a strange scenario so there=0A> is probably=0A> something we are mis=
sing.=0A> =0A> In our case, we have a (bean) data model that we need to=0A>=
 create from a=0A> series of xmls. This data model represents network=0A> c=
onfiguration and=0A> we have xml's that come from different network devices=
.=0A> Each network=0A> device produces different XMLs so we use a set of di=
gester=0A> rules to=0A> parse that into the data model.=0A> =0A> So digeste=
r rules are device specific while our data model=0A> (wants to=0A> be) is d=
evice agnostic.=0A> =0A> The problem we see is that the digester forces us =
(via=0A> rules as=0A> CallMethodRule) to include parsing logic in methods=
=0A> *inside* the data=0A> model and that kind of breaks our hardware abstr=
ation=0A> layer.=0A> =0A> I can illustrate this with an example. In one of =
the=0A> devices, a string=0A> comes stating if an interface is "HALF-DUPLEX=
" or=0A> "FULL-DUPLEX", in=0A> our data model, we have a boolean for that. =
We have this=0A> logic:=0A> =0A> void bool isFullDuplex(String hardwareDepe=
ndentString)=0A> {=0A> =A0=A0=A0if(=0A> hardwareDependentString.equalsIgnor=
eCase("FULL-DUPLEX")=0A> =A0 =A0 =A0=A0=A0return true;=0A> =A0=A0=A0else=0A=
> =A0 =A0 =A0=A0=A0return false;=0A> }=0A> =0A> Using basic digester rules =
we have to create this method in=0A> the data=0A> model while i would like =
to keep it next to the parser, as=0A> it is where=0A> this kind of logic be=
longs.=0A> =0A> Are we missing something? This is something that is=0A> hap=
pening all the=0A> time to us (XMLs are quite hardware dependent) so we are=
=0A> open to=0A> proposals :)=0A> =0A> Any help is very much appreciated. T=
hanks,=0A> Pau=0A=0A=0A=0A      

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