[jsr-314-open] [Behavior] Why is HtmlOutputText not a ClientBehaviorHolder ?

[Matthias Wessendorf]

On Tue, Jan 12, 2010 at 9:07 PM, Ed Burns <Ed.Burns at sun.com> wrote:
>>>>>> On Tue, 12 Jan 2010 11:03:34 -0500, Andy Schwartz <andy.g.schwartz at gmail.com> said:
>
> AS> So strange.  Matthias and Lincoln's responses didn't make it to my
> AS> Oracle account, but did make it to gmail.  Apparently I've got some
> AS> mail delivery issues here.  Yay.
>
> AS> On Tue, Jan 12, 2010 at 10:45 AM, Lincoln Baxter, III
> AS> <lincolnbaxter at gmail.com> wrote:
>>>
>>> What are your performance concerns, if I have not inferred correctly?
>>>
>
> AS> My concern is that there will be increased overhead during
> AS> h:outputText rendering due to an increased # of attribute lookups.  It
> AS> is possible that overhead is nominal, but not sure.  The use case
> AS> where this overhead would be most noticeable would be the stamping
> AS> case, where we may end up stamping out h:outputText components many
> AS> times (once per row/column of a data table).
>
> I agree with Andy.  I oppose doing anything to h:outputText until we
> have some concrete performance numbers on the impact of the proposed
> change.  Also, I feel the benefit to the user of doing this change is
> marginal at best.

I totally understand that. For me (personal requirement) the Trinidad2
<tr:outputText>
is all I need and making a custom (or extensional) outputText
ClientBehaviorHolder
is pretty trivial :-)

So, I think we can close the ticket as h:outputText was never having "DOM apis"
and adding them could be worse for performance; Do you want me to close it ?

-Matthias

>
> Ed
>
> --
> | ed.burns at sun.com  | office: 408 884 9519 OR x31640
> | homepage:         | http://ridingthecrest.com/
>



-- 
Matthias Wessendorf

blog: http://matthiaswessendorf.wordpress.com/
sessions: http://www.slideshare.net/mwessendorf
twitter: http://twitter.com/mwessendorf