Mailing-List: contact users-help@myfaces.apache.org; run by ezmlm
Precedence: bulk
Reply-To: "MyFaces Discussion" <users@myfaces.apache.org>
Received-SPF: pass (asf.osuosl.org: local policy)
Content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Subject: RE: Help me understand component lifecycle please.
Date: Thu, 27 Apr 2006 14:47:20 -0400
Message-ID: 
 <981D1FCECE43C7428583255757761182098A8A@infoterra-m2003.Infoterra.com>
Thread-Topic: Help me understand component lifecycle please.
Thread-Index: AcZqKpqffOqY7/bvTICdgZLprMnHBAAAB1Aw
From: "Rhys Parry" <rparry@Infoterra.com>
To: "MyFaces Discussion" <users@myfaces.apache.org>

Ben,

If you wouold like to take this off-line I can share my knowledge on the =
subject.  In the end I created a bunch of aspects ( compile time aspectJ =
(although runtime would be preferable)) that should not be a problem to =
use.

If you would like email me at rparry@infoterra.com


Rhys Parry


-----Original Message-----
From: Neuman, Ben J., A&M IRM [mailto:bjneuman@hq.afis.osd.mil]
Sent: April 27, 2006 2:43 PM
To: 'MyFaces Discussion'
Subject: RE: Help me understand component lifecycle please.


Rhys, I think you hit the nail on the head. You can't get the component
during the initial render response phase. The workarounds seem like an =
awful
lot of work for my needs.=20

-----Original Message-----
From: Rhys Parry [mailto:rparry@Infoterra.com]
Sent: Thursday, April 27, 2006 2:40 PM
To: MyFaces Discussion
Subject: RE: Help me understand component lifecycle please.


All,

I know I am joining this discussion late, however, I just went through =
the
painful process of writing my own component library because I could not =
get
the components during the render response phase.  My idea was that I =
should
be able to get the component id and based on that do some additional
security checking.  If it fails then set rendered =3D false.  It would =
be
clean.  However. . . no go.

>>However, it's also possible to configure it by using a binding
>>attribute -- you bind the attribute to a backing bean, and then,
>>depending on whether you use set or get, you can either modify the
>>preconstructed component, or create your own version of the component
>>yourself.
Did that and it is more work than I had hoped.

Also cluttering my code with=20
<sometag value=3D"..." isRendered=3D"{bean.method}"/>

and then=20
class SomeBean
{
	public boolean isMethod()
	{
		//do some boilerplate stuff
	}
}

seems like a lot of replicated redundant code.

A thought I had was that it would be nice if we could set up a
JSFRenderCallbackHandler.  This object would be configured in the
faces-config.xml file and would be called during the isRendered phase of =
the
component lifecycle passing in the component as its arg. . . or the =
id(?).
This would remove the boilerplate and not force developers to write a
component library.  Also a JSFDisabledCallbackHandler would be nice.

My 2 cents,

Rhys


-----Original Message-----
From: Neuman, Ben J., A&M IRM [mailto:bjneuman@hq.afis.osd.mil]
Sent: April 27, 2006 2:15 PM
To: 'MyFaces Discussion'
Subject: RE: Help me understand component lilfecycle please.


Not sure I understand. Are you referring to the binding of a component's
attribute to a backing bean? Or the binding of the component itself to a
backing bean?

-----Original Message-----
From: Mike Kienenberger [mailto:mkienenb@gmail.com]
Sent: Thursday, April 27, 2006 12:24 PM
To: MyFaces Discussion
Subject: Re: Help me understand component lilfecycle please.


On 4/27/06, Neuman, Ben J., A&M IRM <bjneuman@hq.afis.osd.mil> wrote:
> Got it. It makes sense to me to "disregard" unrendered components =
during
> phase-processing code. I guess I have issues with the inability to =
modify
> components before the initial rendering. Still feel this is a spec
weakness.

Well, you "configure" a component by specifying attributes.

However, it's also possible to configure it by using a binding
attribute -- you bind the attribute to a backing bean, and then,
depending on whether you use set or get, you can either modify the
preconstructed component, or create your own version of the component
yourself.

Sorry I didn't point this out earlier as this might be what you wanted.