A few notes:
- Typo in the header of the test: "Copyright (c) 20015", also please
dispose the window before "throw new RuntimeException"
- FocusEvent.java: some of the new javadoc overlap the 80 chars per
line, also remove double spaces in the text.
- I wonder about serialization, the new field "cause" was added to the
FocusEvent which implements Serializable, but the serialVersionUID was
not changed.
- Why did you add a check to the new constructor of FocusEvent? This
check should be done already in the EventObject, and executes before
your new check? Is it a typo and it should be cause? When why do not
throw an NPE? The test should verify this also.
- I think we should make getCause method fine. Do you see some benefits
to override it? In this case it will not be necessary to add an Accessor
or getCause_NonCLientCode like we usually do when the user is able to
override it.
On 11.09.15 0:15, Semyon Sadetsky wrote:
Hello,
Please review fix for JDK9:
bug: https://bugs.openjdk.java.net/browse/JDK-8080395
webrev: http://cr.openjdk.java.net/~ssadetsky/8080395/webrev.00/
This fix moves the caused property to java.awt.event.FocusEvent to make
it public. The sun.awt.CausedFocusEvent class is removed as unnecessary.
The API change was approved http://ccc.us.oracle.com/8080395.
--Semyon
--
Best regards, Sergey.
