Hi Sergey, The fix looks fine. Shall we add ‘headful’ key to the test?
Thanks, Dmitry > On 29 Apr 2019, at 11:27, Sergey Bylokhov <[email protected]> wrote: > > Hello. > Please review the fix for JDK 13. > > Bug: https://bugs.openjdk.java.net/browse/JDK-8144125 > Fix: http://cr.openjdk.java.net/~serb/8144125/webrev.00 > > One of the tests fails on macOS because we send two events for the component > is resized/moved. > When the LWComponentPeer was implemented I was not aware that the Component > class can send such events: > http://hg.openjdk.java.net/jdk/client/file/6b1bac001aaf/src/java.desktop/share/classes/java/awt/Component.java#l2387 > > Currently, LWComponentPeer sends these events even if the target Component > should not be updated. As a fix I suggest to post these events by the peer > only if the target should be updated as well, which will mean that we got > this change from some other code, not the component(and the component will > not fire these events). > > -- > Best regards, Sergey.
