Right, but in general that demonstrates how leaky this abstraction is.
Again, I look forward to Max's response on this because really I'm just
getting this conversation started. He seems to have thought through all the
ramifications quite extensively. I think the bigger issue is actually the
CDATA, but it is really linked.
-Dan
On Fri, Dec 11, 2009 at 6:28 PM, Ken Paulsen <Ken.Paulsen(a)sun.com> wrote:
I'm sure this is already known by most of you, but a work-a-round is to
wrap the outer-most content that you want sent to the browser in a
<ui:composition> tag. Not the cleanest solution, but it does the trick:
<?xml junk...>
...
<ui:composition>
Real content here
</ui:composition>
...
+1 to making this configurable option.
Ken
Dan Allen wrote:
> Max (Oracle) brought up a critical point at JSF Summit (and earlier in
> private discussions). Facelets is passing through the XML declaration in a
> template to the browser. This is problematic for certain browsers, to the
> point that it can change the rendering behavior of the browser. There needs
> to be a way to suppress this. A stopgap solution is to introduce a context
> param. A more long term approach is to add an attribute to f:view that
> indicates whether the XML declaration should be sent to the client (you may
> want it for an atom feed for example).
>
> I'll let Andy chime in on Max's behalf for follow up, since he has some
> use cases he can cite and perhaps further suggestion.
>
> -Dan
>
> --
> Dan Allen
> Senior Software Engineer, Red Hat | Author of Seam in Action
> Registered Linux User #231597
>
>
http://mojavelinux.com
>
http://mojavelinux.com/seaminaction
>
http://www.google.com/profiles/dan.j.allen
>
--
Dan Allen
Senior Software Engineer, Red Hat | Author of Seam in Action
Registered Linux User #231597
http://mojavelinux.com
http://mojavelinux.com/seaminaction
http://www.google.com/profiles/dan.j.allen