Hi, Petr.
As far I understand we never check the ownedWindowList to null, since we
initialize it even before the constructor call. The problem exists only
in deserialisation code before we init this field. So from my point of
view the first version of the fix was better.
On 28.10.2013 12:26, Petr Pchelko wrote:
Hello, AWT Team.
Sorry for this mess with reviews, could you please review yet another version
of this fix:
The bug: http://bugs.openjdk.java.net/browse/JDK-8027152
Th fix: http://cr.openjdk.java.net/~pchelko/8027152/webrev.02/
The alwaysOnTop property would be applied to the owned windows when they would
be initialized.
With best regards. Petr.
On 25.10.2013, at 13:06, Petr Pchelko <[email protected]> wrote:
Hello, AWT Team.
Please review the fix for the issue:
http://bugs.openjdk.java.net/browse/JDK-8027152
The fix is available at:
http://cr.openjdk.java.net/~pchelko/8027152/webrev.01/
The problem:
After the fix for JDK-7081594 the setAlwaysOnTop needs the ownedWindowList to
propagate the alwaysOnTop property to owned windows. However, after the
deserialization we get an NPE. In the fix I've moved the
ownedWindowList serialization and deserialization earlier, before we apply the
alwaysOnTop. The test shows that owned windows are serialized and deserialized
properly.
With best regards. Petr.
--
Best regards, Sergey.
