RFR: Hole for alternative plugin implementation

[David Holmes]

Hi Omair,

I don't think supporting the ability to extend sun.* classes publicly is 
a good solution for this (or any) RFE. Further I would not expect it to 
even be possible later in JDK 9 due to the expected modularity controls.

Just my personal opinion.

David

On 24/01/2014 9:38 AM, Omair Majid wrote:
> Hi,
>
> Over in the IcedTea project, we have an alternative (and Open Source)
> implementation of javaws/plugin called IcedTea-Web [1]. IcedTea-Web has
> always been built and run against IcedTea. However, that has required
> IcedTea to carry patches that are not part of OpenJDK.
>
> It would be much better if these patches were part of OpenJDK, I think.
> It would mean OpenJDK could work with an Open Source/Free Software
> plugin, IcedTea-Web wouldn't need a patched IcedTea and IcedTea wouldn't
> have to carry this patch locally.
>
> The webrev is at:
> http://cr.openjdk.java.net/~omajid/webrevs/applet-hole/00/
>
> This is a very small patch and it only relaxes visibility for methods
> and members declared in two sun.applet.* classes. I think it's a very
> safe change.
>
> The change has been quite stable, we have been using this patch in
> IcedTea 7 (without any modifications) since 2010.  The history of the
> patch before that was slightly more complex. Andrew Hughes added it to
> IcedTea 7 but it was taken from IcedTea6. The log for
> icedtea6/patches/applet_hole.patch shows the following people have
> contributed to it:
>
> Andrew John Hughes
> Deepak Bhole
> Omair Majid
> Thomas Fitzsimmons
>
> Since all are/were Red Hat employees, the patch is covered by the CLA.
>
> So, what do you all think about this? Are there any concerns or
> suggestions? I will be happy to rework the patch as needed.
>
> Thanks,
> Omair
>
> [1] http://icedtea.classpath.org/wiki/IcedTea-Web
>