On Mon, 24 Apr 2023 11:25:28 GMT, Jose Pereda <jper...@openjdk.org> wrote:
>> Now I see that the call will actually get delayed. >> >> `javafx.stage.Window` will do the calls when the window will be shown(). On >> line 1162 `WindowStage.setScene` is called: >> >> >> peer.setScene(SceneHelper.getPeer(getScene())); >> >> >> but this is only called on line 1186: >> >> >> peer.setVisible(true); >> >> >> The configuration of Window properties (iconified, maximized, etc) will >> start on line 1212: >> >> >> WindowHelper.visibleChanged(Window.this, newVisible); >> >> >> which will fall into `Stage` line 1178 `doVisibleChanged`. >> >> I think this is the right place to call `requestFocus` if the window is not >> minimized/iconified. >> >> Calling `requestFocus` on `setScene` will cause an initially iconified >> window to be shown (actually pop on the screen) and then be iconified. > > @tsayao Unrelated to this PR, but I see that your master is [38 commits > ahead](https://github.com/tsayao/jfx/compare/openjdk:jfx:master...master) of > openjdk:master. > This is causing that for every new branch and PR that you create, those > commits (the first 30 ones date from 3 years ago!) are always added to it. > Could you reset your local master branch to the upstream version? > @jperedadnr This should now be ready for you to resume your review. When you do, can you confirm whether [JDK-8304734](https://bugs.openjdk.org/browse/JDK-8304734) and [JDK-8304476](https://bugs.openjdk.org/browse/JDK-8304476) are also fixed? If so, we can close them as a duplicate of [JDK-8306121](https://bugs.openjdk.org/browse/JDK-8306121). ------------- PR Comment: https://git.openjdk.org/jfx/pull/1110#issuecomment-1589777558
