<Swing Dev> Review Request for 7126823 : JInternalFrame.getNormalBounds() returns bad value after iconify/deiconify

[Prem Balakrishnan]

Hi Rajeev,

 

I have tested for all supported LAF's across all platforms(Windows, Linux and Mac)

Test updated for the same.

 

Webrev: http://cr.openjdk.java.net/~arapte/prem/7126823/webrev.01/

 

Regards,

Prem

 

From: Rajeev Chamyal 
Sent: Wednesday, February 10, 2016 2:00 PM
To: Prem Balakrishnan
Cc: Sergey Bylokhov; Semyon Sadetsky; Alexander Scherbatiy; Ambarish Rapte; swing-dev at openjdk.java.net
Subject: RE: <Swing Dev> Review Request for 7126823 : JInternalFrame.getNormalBounds() returns bad value after iconify/deiconify

 

Hello Prem,

 

Did you test this fix for other LAF's as well.

 

Regards,

Rajeev Chamyal

 

From: Prem Balakrishnan 
Sent: 09 February 2016 14:40
To: Sergey Bylokhov; Semyon Sadetsky; Alexander Scherbatiy; Ambarish Rapte; HYPERLINK "mailto:swing-dev at openjdk.java.net"swing-dev at openjdk.java.net; HYPERLINK "mailto:awt-dev at openjdk.java.net"awt-dev at openjdk.java.net
Subject: <Swing Dev> Review Request for 7126823 : JInternalFrame.getNormalBounds() returns bad value after iconify/deiconify

 

Hi,

Please review fix for JDK9,

Bug: https://bugs.openjdk.java.net/browse/JDK-7126823

Webrev: http://cr.openjdk.java.net/~arapte/prem/7126823/webrev.00/

 

 

Issue:

JInternalFrame.getNormalBounds() returns bad value after iconify/deiconify

 

Cause:

Regression: due to 4424247: DefaultDesktopManager does not handle InternalFrame state changes as expected.

https://bugs.openjdk.java.net/browse/JDK-4424247 related to https://bugs.openjdk.java.net/browse/JDK-4900778 

 

Fix:

DefaultDesktopManager.java

setNormalBounds(getBounds()); call removed from iconifyFrame(JInternalFrame f) method.

 

Justification:

This fix is not breaking any of the previous fix i.e., related to 4424247 and 4900778.
(all previous scenarios are tested and validated with the current fix)

 

In JInternalFrame.java

getNormalBounds() returns getBounds() when normalBounds is NULL, 

Hence no need to setNormaBounds externally.

 

public Rectangle getNormalBounds() {

 

      /* we used to test (!isMaximum) here, but since this

         method is used by the property listener for the

         IS_MAXIMUM_PROPERTY, it ended up getting the wrong

         answer... Since normalBounds get set to null when the

         frame is restored, this should work better */

 

      if (normalBounds != null) {

          return normalBounds;

      } else {

        return getBounds();

      }

    }

 

 

As per Javadoc: 
public Rectangle getNormalBounds() 
If the JInternalFrame is not in maximized state, returns getBounds(); otherwise, returns the bounds that the JInternalFrame would be restored to. 

In maximizeFrame() , normalBounds is set to a value which is not NULL, i.e., setNormalBounds(getBounds()); 
In minimizeFrame(), normalBounds is set to NULL, i.e., setNormalBounds(null); 
NormalBounds should NOT be set elsewhere.

 

Test:

Integrated test for current bug and regression.

 

Regards,
Prem

 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.openjdk.java.net/pipermail/swing-dev/attachments/20160210/9b52aa14/attachment.html>