RE: Renderkit using Freemarker templates

["Kito D. Mann" <km...@virtua.com>]

Hmmmm... I presented something similar to this at JavaOne (called
"intuition"), except that it focuses more on component creation (although
the render kit part is just a step further). I was talking to Jacob Hookom
about including it in Facelets, since it uses Facelets for the template
(although that's not required). He wasn't terribly interested, though, so I
haven't quite figured out where it should go...

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Kito D. Mann - Author, JavaServer Faces in Action
http://www.virtua.com - JSF/Java EE consulting, training, and mentoring
http://www.JSFCentral.com - JavaServer Faces FAQ, news, and info

* Sign up for the JSF Central newsletter!
http://oi.vresp.com/?fid=ac048d0e17 *


> -----Original Message-----
> From: news [mailto:news@sea.gmane.org] On Behalf Of Laurie Harper
> Sent: Friday, June 29, 2007 4:51 PM
> To: dev@myfaces.apache.org
> Subject: Re: Renderkit using Freemarker templates
> 
> The performance question is a good one, but I thought I'd throw
> something else into the discussion... I've thought about creating
> something like this a number of times, as a way to make components more
> flexible for the user.
> 
> When a component has more complex rendering, I often find I'd like to
> be
> able to tweak the generated HTML in ways that simply aren't possible
> without replacing the renderer. If the renderer was driven by a
> template
> *and* there was a supported mechanism for specifying a custom template
> (on a per-component-use basis or even just on a per-application basis),
> this would be much easier.
> 
> On the other hand, the potential for people breaking the component by
> changing the render template in unexpected ways is a little
> unsettling...
> 
> L.
> 
> Adam Winer wrote:
> > When I've thought of this, I've always imagined that
> > really complex renderers would end up requiring
> > Java code, and that really simple renderers, well,
> > aren't that hard to write in the first place, and would be
> > much slower when written that way.  At that point,
> > I gave up and worked on things I was sure would work. :)
> >
> > Templating a la Facelets has seemed like a pretty
> > good middle ground for assembling on-the-fly
> > components.
> >
> > That said, I'd be absolutely thrilled to proven wrong!
> > So I guess two questions:
> >  - Have you had a chance to benchmark the Freemarker-coded
> >    outputText against the one in MyFaces?
> >  - Any chance of tackling a complex component so we could
> >   see what it looks like?
> >
> > -- Adam
> >
> >
> >
> > On 6/28/07, Jurgen Lust <jl...@apache.org> wrote:
> >> Hi,
> >>
> >> I think most people will agree that writing Renderers for JSF
> components
> >> using the standard tools, i.e. ResponseWriter.startElement and
> >> ResponseWriter.writeAttribute etc., is not very straightforward or
> >> productive, especially when creating complex components. If you take
> a
> >> look at
> org.apache.myfaces.custom.schedule.ScheduleDetailedDayRenderer,
> >> you'll know what I mean :)
> >>
> >> One of the great things about Dojo is that you can create templates
> for
> >> your components. Inspired by this, I tried to do something similar
> for
> >> JSF renderers, using Freemarker. I put it in the sandbox, and wrote
> a
> >> replacement HtmlOutputTextRenderer based on it. I know the
> outputText is
> >> a bad example for this, but for complex stuff, this could be really
> >> useful.
> >>
> >> What do you think?
> >>
> >> Jurgen
> >>
> >