[OpenJDK 2D-Dev] RFR: [JDK-8130400] Test java/awt/image/DrawImage/IncorrectClipXorModeSurface2Surface.java fails with ClassCastException
prasanta sadhukhan
prasanta.sadhukhan at oracle.com
Mon Jul 27 19:16:34 UTC 2015
Thanks Phil for the review.
Modified webrev can be found here:
http://cr.openjdk.java.net/~psadhukhan/8130400/webrev.01/
Please let me know if this ok.
Regards
Prasanta
On 7/25/2015 4:00 AM, Phil Race wrote:
> A couple of niggles
>
> 1) " && extraAlpha == 1.0f)){"
>
>
> There should be a space between ) and {
> But preferably there { should be on the next line .. properly indented to
> the same depth as the beginning of "if {". This is a special
> formatting rule
> for multi-line conditions
>
> 2) The test does not need to catch ClassCastException or do anything
> special
> since CCE *is* a runtime exception and all you are doing is hiding the
> details
> of what actually happened. In fact I wonder why you created the new
> test to
> begin with. It seems to duplicate the failing test. All you need to do is
> add this new bug ID to the @bug line in the existing test
>
> Clemens, do you see any issue with this fix ?
>
> -phil.
>
>
>
> On 07/22/2015 12:08 AM, prasanta sadhukhan wrote:
>> Hi All,
>>
>> It seems OGL and D3D pipeline delegates XORcomposite transform to
>> DrawImage#renderImageXform which is the solution opted for XRender
>> pipeline too.
>> Can you please review this?
>>
>> Regards
>> Prasanta
>> On 7/17/2015 5:10 PM, prasanta sadhukhan wrote:
>>> Hi,
>>>
>>> Bug: https://bugs.openjdk.java.net/browse/JDK-8130400
>>> webrev: http://cr.openjdk.java.net/~psadhukhan/8130400/webrev.00/
>>>
>>> The test fails with ClassCastException
>>> java.lang.ClassCastException: sun.java2d.loops.XORComposite cannot
>>> be cast to java.awt.AlphaComposite
>>> when drawing volatimeImage image to volatimeImage in xor mode since
>>> during setting xor mode, we do a SetComposite with XORComposite object
>>> which is not taken into account during transformed blits in linux in
>>> Xrender pipeline.
>>>
>>> Regards
>>> Prasanta
>>>
>>>
>>
>
More information about the 2d-dev
mailing list