Hello Sergey.
There's a second version of fix with a test:
http://cr.openjdk.java.net/~mcherkas/8077409v9/webrev.01/
I found out that the problem is in layout method, it should not use insets,
ViewportSize is already calculated with considering about insets.
During testing jdk without the fix I faced with rendering problems like
described in jbs bug, also
I faced with duplication of very first pixel of scroll pane.
With http://cr.openjdk.java.net/~mcherkas/8077409v9/webrev.01/
everything works fine. I don't see any rendering issues or problems with
scrolling.
I also run test from test/(closed)/java/awt/ScrollPane, all tests are
passed.
Thanks,
Mikhail.
On 5/5/2015 7:24 PM, Sergey Bylokhov wrote:
Hi, Mikhail.
Is it possible to write a test for this issue based on the code in the
bug report?
On 30.04.15 22:44, mikhail cherkasov wrote:
Hi there,
Could you please review the fix:
bug: https://bugs.openjdk.java.net/browse/JDK-8077409
webrev: http://cr.openjdk.java.net/~mcherkas/8077409v9/webrev.00/
In ScrollPane, move() is called without considering the quota of
Insets when invoking ScrollPane$PeerFixer.adjustmentValueChanged(). When
validate() is called, drawing deviates to coordinate with Insets
interanlly.
Thanks,
Mikhail.
