On 29.02.16 13:24, Alexey Ivanov wrote:
Hi Sergey,
The fix looks good.
Could you please make the frame disappear when the test completes?
Yes, the new version:
http://cr.openjdk.java.net/~serb/8147994/webrev.02
Regards,
Alexey
On 15.02.2016 20:35, Sergey Bylokhov wrote:
Hello.
Please review the fix for jdk9. It will be backported to jdk8.
This fix is a rework of the fix for JDK-8033000:
http://hg.openjdk.java.net/jdk9/jdk9/jdk/rev/c2acbd0292f3
https://bugs.openjdk.java.net/browse/JDK-8033000
After the fix for JDK-8033000 all our L&F have the next wheel logic:
1 If shift is not pressed and vertical scroll bar is visible it
should be scrolled.
2 If vertical scroll-bar is invisible then we should scroll
horizontal scroll bar if possible.
3 If shift is pressed and horizontal scroll-bar is visible it should
be scrolled, but if it is not visible then we should try vertical
scroll-bar.
Note that the step 3 was added in the JDK-8033000. This caused an
issue which previously was reproduced in Aqua L&F only. When the user
uses diagonal scroll(top+right) and the application has only vertical
scroll-bar, then we try to scroll in opposite directions:
- the top move generates the scroll down.
- scroll to the right generate the wheel+shift, and since the
horizontal scroll-bar is not visible-> scroll up is executed.
In the fix the logic of step 3 is changed:
3 If shift is pressed and horizontal scroll-bar is visible it should
be scrolled, otherwise do nothing.
Testcase for JDK-8033000 is updated.
Bug: https://bugs.openjdk.java.net/browse/JDK-8147994
Webrev can be found at:
http://cr.openjdk.java.net/~serb/8147994/webrev.01
--
Best regards, Sergey.
