Thank you Sergey,
On 12/03/2014 04:23 PM, Sergey Bylokhov wrote:
Hi, Yuri.
It will be good to add setAutoDelay(50) to these tests, because they
press keys w/o delays:
- KeyCharTest
- DragMouseOutAndRelease
- GetSizeTest
- PopupPosTest
- UnfocusableCB_ERR
- NativeInLightShow
- ConsumeForModalDialogTest
- ConsumeNextMnemonicKeyTypedTest
- EnqueueWithDialogButtonTest
- MenuDragEvents
I suggest to file a separate issue for
SwingUtilities2.getSystemMnemonicKeyMask()
Are you sure that ResizeAutoClosesChoice.java is not a java bug?
Actually there is a bug, and even the original bug: dropdown does not
close on resize -- but this test in its current state cannot
demonstrate that. It tries to check a color of background expecting
Color.red -- but on OS X setBackground() is not honored.
Tomorrow morning I'll file a bug and provide a snippet but for
the future fix we would need a regtest written on a different principle.
-yan
On 02.12.2014 19:35, Yuri Nesterenko wrote:
Added more useful
http://cr.openjdk.java.net/~yan/8063104/webrev.diff.00_1
diff between old closed and new open versions.
Thanks,
-yan
On 12/02/2014 03:26 PM, Yuri Nesterenko wrote:
Colleagues,
please review this test fix.
http://cr.openjdk.java.net/~yan/8063104/webrev.00
pre-move state of the closed tests:
http://cr.openjdk.java.net/~yan/8063104/webrev.diff.00
There are modularization-related changes in (plus certain cleanup of)
some 66 awt tests and some helper classes.
Some of the tests moved from the closed part.
Everywhere eliminated references to SunToolkit.realSync()
and some other internal API.
Partially fixed 8065960 and 7090905.
I hope that fixed as well 7020056.
One of the helpers, java/awt/regtesthelpers/Util.java, split in two.
Rationale is, we would need to allow internal access to small part
of the tests using new UtilInternal and not to all using Util.
Piggybacked, one swing test javax/swing/JLabel/6596966/bug6596966.java
and a correspondent change in javax/swing/regtesthelpers/Util.java.
Thanks,
-yan
