On Wed, 2015-01-28 at 17:14 +0300, Alexander Scherbatiy wrote: > On 1/27/2015 11:02 PM, Mario Torre wrote: > > Hi all, > > > > I've just created this bug: > > > > https://bugs.openjdk.java.net/browse/JDK-8071705 > > > > Basically, there's a problem when using menus with a dual (or more) > > monitor configuration and the screens are stacked one on top of the > > other. > > > > In this case, the JMenu code that calculates the position for the popup > > does not account the screen bounds relative y coordinate, resulting in > > application that have popups on the bottom side of the second screen to > > have their popup growing upward even if there is space downward. > > > > A proposed fix for the issue is here: > > > > http://cr.openjdk.java.net/~neugens/8071705/webrev.00/ > > > > I need to create an automated test case, however I'm not entirely sure > > how to do that (since it needs a dual screen setup). Perhaps a manual > > test will do? > You can check number of screen devices and pass the test in case > there is only one screen.
Hi Alexander, The test is a bit more complicated than what I wanted, but most of it is just because of setup and preparation: http://cr.openjdk.java.net/~neugens/8071705/webrev.02/jdk.patch It basically tests if there would be enough screen real estate to allow for a menu to popup downward, and then check if this is indeed the case. I let the test pass if there is only one screen or the screens are not vertically stacked, or if there is no space for a downward facing popup. Cheers, Mario
