Hi, Manajit. Need of additional movement of a mouse means that we did not get all necessary events when the window occurred under the mouse. This test was created for JDK-7154048 which fixed similar issues: http://mail.openjdk.java.net/pipermail/awt-dev/2012-April/002613.html "1) The window does not receive the mouse entered event when it is created under the mouse"
I run this test twice on jdk 8 GA and it is passed, so it seems we have a regression in jdk8/9. I suggest to investigate what is the root issue, and change the test so it will always fail for the first run (we can run twice when the mouse inside/ouside label). ----- manajit.hal...@oracle.com wrote: > > > Hi All, > > Kindly review the fix for JDK10. > > Bug: > https://bugs.openjdk.java.net/browse/JDK-8177326/ > > Webrev: > http://cr.openjdk.java.net/~mhalder/8177326/webrev.00/ > > Issue: > [macosx] mouse event goes to button in > java/awt/Mouse/EnterExitEvents/DragWindowOutOfFrameTest.java. > > Cause: > Problem in the test case. Initial observation was that the test case used to > pass for the first time and fail second time onwards. On further debugging it > was observed that the test case used to pass every time if the initial mouse > move in the test actually happened. The test used to fail every second or > further times because after the first execution using jtreg the mouse used to > remain at the same location. > Fix: > The mouse is moved at the beginning of the test to make sure the initial > mouse move on the label always happens. > > Note: > BUTTON1_MASK is deprecated and hence is replaced with BUTTON1_DOWN_MASK. > > Regards, > Manajit
