RFR: 6681958: Maximization state of JInternalFrames is corrupted by WindowsDesktopManager [v2]

[Phil Race]

On Wed, 11 Oct 2023 02:40:20 GMT, Prasanta Sadhukhan <psadhukhan at openjdk.org> wrote:

>> src/java.desktop/windows/classes/com/sun/java/swing/plaf/windows/WindowsDesktopManager.java line 80:
>> 
>>> 78:                         if (f.isMaximizable()) {
>>> 79:                             if (!f.isMaximum()) {
>>> 80:                                 f.setMaximum(true);
>> 
>> So if the new frame is set to maximum=true, don't you need to set the current frame to maximum=false ?
>> The way I'm interpreting the class doc. and the comment about "transfer" is that only one frame can be in the maximised state.
>
> I was interpreting it as transfer the attribute but retain it for current frame....but it seems your interpretation is more logical...so I have updated the PR..
> Since I guess there's no JInternalFrame equivalent in native widget, it leaves scope for interpretation..
> 
> Now with this fix, in SwingSet2 JInternalFrame demo in WindowsL&F
> 
> - If Frame 0 maximised, it goes to maximised state
> - Then when another frame say, Frame 4 maximised, Frame 4 is not restored to un-maximised size but actually been maximised and Frame0 is restored to un-maximised state..

I'm still not sure everything is right.
It seems that a frame can have the "maximized" state, even if it is currently iconified.
Why ? Well because have this test "if (!currentFrame.isIcon()) {"  // (line 77)
which we can only reach "currentFrame.isMaximum()" // line 71

I'm wondering if that test at line 77 should be there ?
It prevents anything happening if the currently maximised frame is iconified. Doesn't it ?
Seems like it shouldn't matter, the new frame "f" should be set to maximum regardless 

I don't understand that 'special case' comment either.

-------------

PR Review Comment: https://git.openjdk.org/jdk/pull/16097#discussion_r1364538277