On 18.11.16 14:51, Semyon Sadetsky wrote:
On 11/18/2016 1:20 PM, Sergey Bylokhov wrote:
On 18.11.16 12:40, Semyon Sadetsky wrote:
On 15.11.2016 17:06, Sergey Bylokhov wrote:
What test did you use to verify the bug? It seems that the test in
closed ws was reworked by Manajit and moved to the open ws. But closed
version was existed till today. (Note probably the fix for
both tests passed.
And both failed before this fix?
It depends on the selected multiple screen configuration. But I agree
that the test was moved with a mistake in 8163270.
Isn't the open test always passed before/after the fix?
JDK-8145784 should be reworked because it return behavior which was
before JDK-8013116).
could you clarify?
The change which was added in the JDK-8145784(as well as the changes
in this fix) looks quite similar to the code which was removed in
JDK-8013116.
Apparently that was incorrect.
Why do you think so? After reading description of those bug I guess that
it was intended decision to use screen coordinate space in jdk8, in case
of windows we share the same coordinates space between screens, so it is
unimportant on what screen the robot was created.
On 11.11.16 17:06, Semyon Sadetsky wrote:
Hello,
Please review fix for JDK9:
bug: https://bugs.openjdk.java.net/browse/JDK-8169133
webrev: http://cr.openjdk.java.net/~ssadetsky/8169133/webrev.00/
Screen specific AWT robot was not implemented on Windows platform. The
fix proposes the missed implementation.
--Semyon
--
Best regards, Sergey.
