bjaspan@ATHENA.MIT.EDU (Barr3y Jaspan) (07/17/90)
As I understand it, if a program wants to be *sure* that a window is actually visible on the screen after an XMapWindow request, the program should block until it gets an Expose (not MapNotify) event for that window. Is there a similar mechanism for unmapping? Does an UnmapNotify actually guarantee that window is not on the screen or does it just mean the server will consider removing it sometime in the future? Barr3y Jaspan, bjaspan@athena.mit.edu Student Information Processing Board (SIPB)
mouse@LARRY.MCRCIM.MCGILL.EDU (der Mouse) (07/18/90)
> As I understand it, if a program wants to be *sure* that a window is > actually visible on the screen after an XMapWindow request, the > program should block until it gets an Expose (not MapNotify) event > for that window. That guarantees only that some part of the window is visible (and not even that; the server may be maintaining backing-store). If you want to find out about visibility, how about selecting for VisibilityChangeMask and watching for VisibilityNotify events? That is, after all, what they're for. > Is there a similar mechanism for unmapping? Does an UnmapNotify > actually guarantee that window is not on the screen or does it just > mean the server will consider removing it sometime in the future? The former. While being mapped does not guarantee that the window is visible, being unmapped does guarantee that it's not. der Mouse old: mcgill-vision!mouse new: mouse@larry.mcrcim.mcgill.edu