Hi Dmitry,
Well, the fix seems correct to me. I tried to thought of any possible regressions but nothing came
to my mind (let's suppose this was really a mistake in the code).
However, wouldn't you like to do the same for swing's SortingFocusTraversalPolicy? And also, include
it into the test scenario?
(Hope you've run all the focus related regression tests).
Thanks,
Anton.
On 06.04.2015 10:14, dmitry markov wrote:
Hello,
Could you review the fix for jdk9, please?
bug: https://bugs.openjdk.java.net/browse/JDK-8073453
webrev: http://cr.openjdk.java.net/~dmarkov/8073453/jdk9/webrev.00/
Problem description:
The method ContainerOrderFocusTraversalPolicy.getLastComponent() always returns null if the last
component is a container with focus traversal policy and does not have any sub-components. In some
cases such behaviour of getLastComponent() causes failure during reverse focus transition, (i.e.
focus stays on the selected component when SHIFT+TAB is pressed).
Fix:
If the last component is a container with focus traversal policy and does not have any
sub-components, the method getLastComponent() should return a previous component instead of null.
Please note: the same approach is already implemented for
ContainerOrderFocusTraversalPolicy.getFirstComponent().
Thanks,
Dmitry
