That works fine, until your components render inside a loop, at which
point they need new ids for each rendering (since ids names must be
unique).  This is even worse in the Portlet world, since the HTML page
will be made up of many portlets that could easily provide conflicting
ids for elements.

The framework operates, by design, in a kind of "tunnel vision", where
it can't rely on just about anything outside of the one component
actively rendering.

On 5/12/05, Preston L. Bannister <[EMAIL PROTECTED]> wrote:
> This sure feels like a mistake.
> 
> In the DOM and CSS worlds nodes can be selected by tag, class, and id.
> The Tapestry domain invents/requires yet another selector.
> 
> Personally I find this a source of much duplication.  If you have a need
> to identify an element, better to use one "id" for all worlds - CSS,
> Javascript, and Tapestry.  For elements where the id is needed only for
> Tapestry, a simple name convention could suffice to identify id's that
> should be stripped from the generated HTML.
> 
> My $0.02.
> 
> 
> Howard Lewis Ship wrote:
> > Yes, to avoid the class.
> >
> > 4.0 lets you override which attribute to look for, so you can use id
> > instead of jwcid (but then you can't set the id of your rendered
> > elements).
> 
> >>-----Original Message-----
> >>From: Gregg D Bolinger [mailto:[EMAIL PROTECTED]
> >>Sent: Tuesday, May 10, 2005 5:48 PM
> >>
> >>Why does Tapesty use jwcid= instead of just looking for the id attribute and
> >>then verfying that against the .page specification as it does for the jwcid.
> >>Would it be harder to catch missing components early on? Just curious.
> 
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [EMAIL PROTECTED]
> For additional commands, e-mail: [EMAIL PROTECTED]
> 
> 


-- 
Howard M. Lewis Ship
Independent J2EE / Open-Source Java Consultant
Creator, Jakarta Tapestry
Creator, Jakarta HiveMind

Professional Tapestry training, mentoring, support
and project work.  http://howardlewisship.com

---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to