users@jersey.java.net

Re: [Jersey] Webkit issue for revisions after 1.1.3-ea

From: Daniel Shaw <java_at_dshaw.com>
Date: Mon, 30 Nov 2009 13:04:49 -0500

We're using implicit views.

Actually, I first encountered the issue with 1.1.5-ea-SNAPSHOT because I
wanted to check out the new Tracing feature. I then stepped back through the
1.1.4.x release and finally had to revert to 1.1.3-ea.

Kind of swamped getting ready for a release and I don't have any show
stoppers rolling out on a previous version, but I'll definitely try to pull
something together so we can explore this further.

Thanks,

Daniel Shaw
i: http://dshaw.com
c: http://www.linkedin.com/in/dshaw


On Mon, Nov 30, 2009 at 3:45 AM, Paul Sandoz <Paul.Sandoz_at_sun.com> wrote:

> Hi Daniel,
>
> Are you using explicit or implicit views?
>
> As the other Daniel ( :-) ) says the @ ImplicitProduces (see the bookstore
> sample) might work but it still does not explain why things previously
> worked for you.
>
> Have you tried 1.1.5-ea-SNAPSHOT?
>
> I recently found a regression in 1.1.4.x that i fixed in 1.1.5-ea-SNAPSHOT.
> The regression was that if you set any headers and status code (in a built
> Response or response filter) they are ignored for JSP views (the flushing of
> headers was not occurring before the servlet forwarding to the JSP).
>
> If 1.1.5-ea-SNAPSHOT does not work then can you send me a simple test case
> and i can investigate.
>
> Paul.
>
>
> On Nov 28, 2009, at 9:43 PM, Daniel Shaw wrote:
>
> We're using MVCJ backing some JSPs and on any of the revisions after
> 1.1.3-ea, I'm getting the following error for any Webkit browser (Safari and
> Chrome, in my case tested on Mac): "Resource interpreted as document but
> transferred with MIME type application/xml." In every case, Webkit does not
> even process the page. View source shows a blank page. Firefox loads
> everything without any problems.
>
> I've been digging through the change logs to try to find what might be
> going on. However, I have yet to come up with a lead. I'm hoping someone
> else has encountered the same issue. Thanks.
>
> Daniel Shaw
> i: http://dshaw.com
> c: http://www.linkedin.com/in/dshaw
>
>
>