<AWT Dev> [9] Review request for 8132123 MultiResolutionCachedImage unnecessarily creates base image to get its size

[Sergey Bylokhov]

Thanks for clarification. The fix looks fine.

On 24.07.15 19:20, Alexander Scherbatiy wrote:
> On 7/24/2015 7:00 PM, Sergey Bylokhov wrote:
>> On 24.07.15 14:10, Alexander Scherbatiy wrote:
>>>
>>> Could you review the updated fix:
>>>   http://cr.openjdk.java.net/~alexsch/8132123/webrev.01
>>>
>>>  There is one more place where AquaSplitPaneDividerUI requests 
>>> scaled smooth image. It is better to allow the native system to 
>>> scale the image.
>>>  So I updated the getScaledInstance(width, height, hints) method in 
>>> the MultiResolutionCachedImage. It also allows to avoid base image 
>>> creation.
>> Why did you change getProperty in the latest version?
>
>    ImageIcon requests comment property from the image and it also 
> leads to unnecessary base image creation.
>
>   Thanks,
>   Alexandr.
>
>>>
>>> On 7/22/2015 8:23 PM, Sergey Bylokhov wrote:
>>>> Hi, Alexander.
>>>> Why you did update of bugid in the test, but did not update the 
>>>> test itself? Did the test fail before the fix?
>>>
>>>      There are limited number of loaded native images and they are 
>>> immediately cached so it is hard to write a test which calculates 
>>> the excessive memory usage using only public API.
>>>
>>>    I added the new test that uses the internal 
>>> MultiResolutionCachedImage class directly.
>>>
>>>   Thanks,
>>>   Alexandr.
>>>>
>>>> On 22.07.15 16:44, Alexander Scherbatiy wrote:
>>>>>
>>>>> Hello,
>>>>>
>>>>> Could you review the fix:
>>>>>   bug: https://bugs.openjdk.java.net/browse/JDK-8132123
>>>>>   webrev: http://cr.openjdk.java.net/~alexsch/8132123/webrev.00
>>>>>
>>>>>   The baseImageWidth/Height and Image.UndefinedProperty are 
>>>>> returned to not trigger base image creation in the 
>>>>> MultiResolutionCachedImage.
>>>>>
>>>>>
>>>>> Thanks,
>>>>> Alexandr.
>>>>>
>>>>
>>>>
>>>
>>
>>
>


-- 
Best regards, Sergey.