<AWT Dev> [9] Review request for 8150844 [hidpi] [macosx] should -Dsun.java2d.uiScale be taken into account for OS X?

Sergey Bylokhov Sergey.Bylokhov at oracle.com
Wed Mar 16 16:34:56 UTC 2016


HiDPIPropertiesUnixTest will be noop on OSX:
  try {
   50             UIManager.setLookAndFeel(
   51                     "com.sun.java.swing.plaf.gtk.GTKLookAndFeel");
   52         } catch (Exception e) {
   53             return;
   54         }

On 16.03.16 15:12, Alexander Scherbatiy wrote:
>
> Could you review the updated fix:
>     http://cr.openjdk.java.net/~alexsch/8150844/webrev.01/
>
>   - the is2x() method is removed from the tests
>   - imports are updated
>
> On 16/03/16 13:25, Alexander Stepanov wrote:
>> Hello Alexandr,
>>
>> just a minor remark about is2x() in tests:
>>
>> do we still need this method in MultiresolutionIconTest?
>> The following simple check can be used in doTest() instead:
>>
>> boolean is2x = "2".equals(scale);
>> Color
>>     expected = is2x ? C2X : C1X,
>>     unexpected = is2x ? C1X : C2X;
>>
>> the same, probably, for MenuMultiresolutionIconTest, but here it is a
>> matter of taste.
>>
>> It seems also that the import for java.awt.geom.AffineTransform is
>> unused for Corrupted2XImageTest (and, in case of changes, in
>> MenuMultiresolutionIconTest).
>>
>> Thanks,
>> Alexander
>>
>> P.S. MultiresolutionIconTest: "@requires (os.family != "mac")" is
>> connected to JDK-8151303, so the test will fail on OS X until it is
>> fixed (having this line removed).
>    It should be fine because bug number 8151303 is mentioned in the @bug
> tag and is counted as a known issue.
>
>    Thanks,
>    Alexandr.
>>
>>
>> On 3/16/2016 11:29 AM, Alexander Scherbatiy wrote:
>>>
>>> Hello,
>>>
>>> Could you review the fix:
>>>   bug: https://bugs.openjdk.java.net/browse/JDK-8150844
>>>   webrev: http://cr.openjdk.java.net/~alexsch/8150844/webrev.00
>>>
>>>   The sun.java2d.uiScale property is now used in CGraphicsDevice.
>>>   The requested tests are updated.
>>>
>>>   Thanks,
>>>   Alexandr.
>>
>


-- 
Best regards, Sergey.


More information about the awt-dev mailing list