[Bug 7404] A regression: Lunarcal.exe did work perfectly well with wine 0.9 .3 but does not launch anymore with wine 0.9.29.

wine-bugs at winehq.org wine-bugs at winehq.org
Thu Sep 6 08:42:44 CDT 2007


--- Comment #16 from Dmitry Timoshkov <dmitry at codeweavers.com>  2007-09-06 08:42:43 ---
Created an attachment (id=7972)
 --> (http://bugs.winehq.org/attachment.cgi?id=7972)
Make ShowWindow(SW_MINIMIZE) move a top level window to position -32000,-32000

The app creates a new dialog with CreateDialogIndirectParamA, and in its
WM_INITDIALOG handler does:

ShowWindow(hwdlg, SW_MINIMIZE); // calls SetWindowPos(SWP_SHOWWINDOW)

UpdateWindow causes a WM_PAINT, and the app crashes in WM_PAINT handler.
The app (or rather mfc42?) simply doesn't expect to receive WM_PAINT for
a minimized top level window.

Before my patch ShowWindow(hwdlg, SW_MINIMIZE) didn't set SWP_SHOWWINDOW flag
for the SetWindowPos() call, and SetWindowPos() didn't create an update region
for a window. The tests show that ShowWindow(hdlg, SW_MINIMIZE) correctly adds
the flag SWP_SHOWWINDOW to SetWindowPos(), so that part is correct.

What is not correct, is that under Windows ShowWindow(hwdlg, SW_MINIMIZE)
although adds WS_VISIBLE (Wine does as well) actually moves a top level
window to position -32000,-32000, i.e. moves it outside of the desktop.

Attached patch implements the same behaviour under Wine, but this uncovers
another Wine bug: Wine still generates a WM_PAINT message for such a window.
I briefly looked at the code in server/window.c,set_window_pos(), but I'd
leave this to Alexandre. To me it looks like a new visible region of a window
should be clipped by parent/desktop, but currently it is not.

Configure bugmail: http://bugs.winehq.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching all bug changes.

More information about the wine-bugs mailing list