<AWT Dev> <Awt Dev> [9] Review Request for 8067470: Examine if the replacement for sun.awt.SunToolkit.setLWRequestStatus should be provided

[Sergey Bylokhov]

On 05.10.15 15:03, Semyon Sadetsky wrote:
> Sergey, any other thoughts?

I suggest to postpone the fix as discussed in private(latest massage was 
added to the CR).


>
>
> On 9/28/2015 12:43 PM, Semyon Sadetsky wrote:
>>
>>
>> On 9/27/2015 1:13 AM, Sergey Bylokhov wrote:
>>> My understanding according to the new javadoc:
>>>
>>> "* If the window does transfer focus to other
>>>  * heavyweight containers and the synchronous lightweight focus requests
>>>  * are enabled to it then further focus behavior is unspecified."
>>>
>>> behavior of this feature is undefined on most(all?) of our platforms,
>>> because it is hard to imagine that we(or application developer) can
>>> guarantee that the focus will never be transfered to other
>>> heavyweight containers. Right?
>> Actuallty KFM tries to use the sync focus transfer and may fail if the
>> requested container is not the current. If this happens the KFM
>> fallbacks to the async mode. Probably we could remove this warning I
>> just copied it from the original spec. The aim of the bug is to
>> provide access to an internal method which is being used by one app
>> now. I did not change the original spec a lot.
>>>
>>> On 26.09.15 0:04, Phil Race wrote:
>>>> Hmm. I thought Anton Tarasov was of the opinion this was not
>>>> supportable
>>>> anyway
>>>> because of platform limitations. In other words we should not do this.
>>>>
>>>> -phil.
>>>>
>>>> On 09/25/2015 01:51 PM, Mandy Chung wrote:
>>>>>
>>>>>
>>>>> On 09/25/2015 04:32 AM, Semyon Sadetsky wrote:
>>>>>> Hello,
>>>>>>
>>>>>> Please review fix for JDK9:
>>>>>> bug: https://bugs.openjdk.java.net/browse/JDK-8067470
>>>>>> webrev: http://cr.openjdk.java.net/~ssadetsky/8067470/webrev.00/
>>>>>>
>>>>>> The corresponding java.awt.Window method is proposed. The method is
>>>>>> protected by a newly introduced property "awt.enableSyncLWFocus".
>>>>>
>>>>> Can you explain why this setter method needs to be protected by a
>>>>> system property?  By default this method is a nop?
>>>>>
>>>>> Mandy
>>>>
>>>
>>>
>>
>


-- 
Best regards, Sergey.