Test demonstrating the need for the callback is: PSwingTest.testHidingComponentHidesPSwing
I believe the callback was introduced as an optimization (though it's been a while since the change). I will investigate further. On 27 January 2010 22:36, cmal...@pixelzoom.com <cmal...@pixelzoom.com>wrote: > Looking at PSwing in 1.3-rc1 ... > > The constructor adds a ComponentAdapter that changes the visibility of > the PSwing node as the Component is shown/hidden. Does anyone recall > the motivation for adding this? > > Looking at the PSwing.setVisible override, this ComponentAdapter seems > unnecessary, perhaps even a bit dangerous. If someone calls > setVisible directly on the Component, will this change the visibility > of the node? (The specifics of exactly when componentShow and > componentHidden are called has always eluded me.) > > One of my applications is experiencing an endless circularity related > to this. The calls look like: > > setVisible true > componentHidden > setVisible false > componentShown > setVisible true > // ... etc forever > > I'm still trying to track this down, and it may be a client-code > issue. But it wasn't an issue with earlier versions of PSwing, and > demonstrates (I think) the potential danger of this ComponentAdapter. > > -- > Piccolo2D Developers Group: > http://groups.google.com/group/piccolo2d-dev?hl=en -- Piccolo2D Developers Group: http://groups.google.com/group/piccolo2d-dev?hl=en
