the concrete frame in question - surprise, surprise :-) - it the mainframe of the SwingXSet: the layout still isn't clean enough to survive a pack. And don't intend to go for the real reason just now, still hoping the _real_ solution will drop out of the sky (which occasionally happens with layout probs). The original version in SwingSet3 side-stepped any problems by ...horror: freely setting preferredSizes and null layout.
This is a problem only in the very first run of the app, later the size/location is restored from persisted session state. A quick fix was to manually size the frame always - which wasn't a solution at all because it short-cuts the restored values. Now I came up with something looking extremely dirty: override configureWindow, pack, set a reasonable enough initial size and validate:
[code]
@Override
protected void configureWindow(Window root) {
super.configureWindow(root);
if (root == getMainFrame()) {
// PENDING JW
// brute force: prevent super from pack()'ing the mainframe
// initially - actually it's a problem of our overall
// layout, packing giving horrible results ...
// we have to do it here (instead of in show as I assumed)
// because super's injecting properties will invalidate
root.pack();
root.setBounds(0, 0, 1024, 748);
root.validate();
}
}
[/code]
Probably overlooking something obvious - any ideas, what?
Thanks
Jeanette
[Message sent by forum member 'kleopatra' (
fastegal@...)]
http://forums.java.net/jive/thread.jspa?messageID=370648---------------------------------------------------------------------
To unsubscribe, e-mail:
jdnc-unsubscribe@...
For additional commands, e-mail:
jdnc-help@...
