<AWT Dev> [12]RFR: JDK-8215921: There is no change when select different Foreground and Background by mouse.

[Sergey Bylokhov]

On 23/01/2019 14:00, Philip Race wrote:
> Do you mean that the test may be showing up a problem on macos ?

The fix does not affect the macOS, but the test does not
check the usecase which it is supposed to verify.

> 
> -phil.
> 
> On 1/23/19, 1:43 PM, Sergey Bylokhov wrote:
>> Hi, Krishna.
>>
>> I tried a new version of the fix on macOS, and looks
>> like the second click is never success, the test always
>> fails to open the list of items and as a result clicked
>> on the desktop instead of the same element in the list.
>>
>> On 23/01/2019 09:33, Phil Race wrote:
>>> This looks OK so long as you can confirm that you re-tested all related tests, not
>>> just your regression test after making the source code update ...
>>>
>>> -phil.
>>>
>>> On 1/21/19 5:00 AM, Krishna Addepalli wrote:
>>>> Hi Sergey,
>>>>
>>>> I have modified the test as you suggested. Here is the new webrev: http://cr.openjdk.java.net/~kaddepalli/8215921/webrev03/
>>>>
>>>> Thanks,
>>>> Krishna
>>>>
>>>>> On 19-Jan-2019, at 3:39 AM, Sergey Bylokhov <Sergey.Bylokhov at oracle.com <mailto:Sergey.Bylokhov at oracle.com>> wrote:
>>>>>
>>>>> Hi, Krishna.
>>>>>
>>>>> Can you, please, in the test add a check after each mouse click, that
>>>>> the new/same value is selected. This will doublecheck this comments:
>>>>> =====
>>>>> 112         //make sure that the mouse moves to a different item, so that
>>>>> 113         //itemStateChanged is called.
>>>>> =====
>>>>> 119         //now click on the same item and make sure that item event is
>>>>> 120         //not generated.
>>>>> =====
>>>>>
>>>>> On 18/01/2019 04:08, krishna.addepalli at oracle.com <mailto:krishna.addepalli at oracle.com> wrote:
>>>>>> Hi Sergey, Phil,
>>>>>> I have tested the new test case in Windows, Mac, Linux, and found that it runs on all 3 platforms without any problems.
>>>>>> I have also tested all the regressions that were raised, and all of them passed. In addition, I ran all the awt/Choice jtreg tests, and found no new failures.
>>>>>> Thanks,
>>>>>> Krishna
>>>>>> *From:*Krishna Addepalli
>>>>>> *Sent:* Thursday, January 17, 2019 8:34 PM
>>>>>> *To:* Sergey Bylokhov <sergey.bylokhov at oracle.com <mailto:sergey.bylokhov at oracle.com> <mailto:sergey.bylokhov at oracle.com>>
>>>>>> *Cc:* awt-dev at openjdk.java.net <mailto:awt-dev at openjdk.java.net> <mailto:awt-dev at openjdk.java.net>
>>>>>> *Subject:* Re: <AWT Dev> [12]RFR: JDK-8215921: There is no change when select different Foreground and Background by mouse.
>>>>>> Hi Sergey,
>>>>>> The test case was leading my reasoning astray! Yes you are right that we post event even for the same item selected again.
>>>>>> I have updated the fix and the test case well, which checks if an ItemEvent is generated when a new item is selected and then it is not generated when the same item is selected again.
>>>>>> Here is the updated webrev: http://cr.openjdk.java.net/~kaddepalli/8215921/webrev02
>>>>>> Thanks,
>>>>>> Krishna
>>>>>>    On 17-Jan-2019, at 1:30 AM, Sergey Bylokhov <Sergey.Bylokhov at oracle.com <mailto:Sergey.Bylokhov at oracle.com><mailto:Sergey.Bylokhov at oracle.com>> wrote:
>>>>>>    On 16/01/2019 09:59, Krishna Addepalli wrote:
>>>>>>        Hi Sergey,
>>>>>>        I checked the code, by putting logs and can confirm that an event is not generated even if the same is reselected.
>>>>>>        I have updated the test case to reflect this fact, so here is the new webrev: http://cr.openjdk.java.net/~kaddepalli/8215921/webrev01/
>>>>>>    But why it is not generated? How (!firstPress || wasDragged ) prevents it to be posted?
>>>>>>     From the code it looks like when "wasDragged==true" or "firstPress==false" we will post an event.
>>>>>>    --     Best regards, Sergey.
>>>>>
>>>>>
>>>>> -- 
>>>>> Best regards, Sergey.
>>>>
>>>
>>
>>


-- 
Best regards, Sergey.