GNOME Bugzilla – Bug 308518
Restoring a maximised window from the system tray gives window of wrong size
Last modified: 2005-08-26 22:14:35 UTC
Version details: 2.8.6-2 (FC3), libgnome 2.8.0-2 (FC3) Distribution/Version: Fedora Core 3 / FC3 I have tested this with both GAIM and amaroK. When either of these applications is maximised and then 'closed' (clicking '[X]') to the system track, and then restored (clicking the application's icon in the system tray), the application cmes back NOT minimised. In fact, it comes back /almost maximised/ - it's top-left position is correct, and the height is correct, but the window width is about 8 pixels too wide, such that the right-hand-side of the window hangs off the edge of the screen. Also, instead of the window showing a 'restore' icon in the middle icon of the top right of the window, it shows a 'maximise' icon, indicating that it's forgotten that it was previously maximused. I reported this over in KDE bugs originally, but I'm using GNOME (under FC3) so I think that I actually should report here. Sorry if it's already been reported. I couldn't find it anywhere in bugzilla here. Also, just noticed: when I start up Synaptic having previously run it maximised, this same too-large-window thing happens as well. extra keywords: maximize maximized, minimize minimized.
It's most likely an application bug. If the app unmaps the window, metacity won't track it across the unmap. The apps should probably be minimizing themselves rather than hiding/unmapping. They could set skip_taskbar if the tray icon is enabled. Alternatively the apps could unmap but record the state (including maximized) across the unmap.
FYI http://bugs.kde.org/show_bug.cgi?id=107844
Also note that Lubos Lunak thought this issue could be related to another issue with maximised applications forgetting their 'always on top' status. Could it be some strange interaction between KDE & GNOME going on? http://bugs.kde.org/show_bug.cgi?id=59572 Any thoughts?
Actually, metacity messes up on unmap with maximized windows. I'm fairly sure that this is a duplicate of bug 137185. John: Any chance you could test whether the patch in that bug fixes this problem for you?
No response, I'll assume it's a dupe of bug 137185. *** This bug has been marked as a duplicate of 137185 ***