The test gets spurious GetScrollInfo SB_HORZ failures if some other
windows were created before using CW_USEDEFAULT positions, for instance
the window from test_capture_4.
This happened on some Win10 VMs, and was thought to be fixed with
9b5346df86bf1a85336ed03a67a630970ce08e9f, but it then broke the test on
some other VMs. This is hopefully a better fix, and it reverts the
previous one too.
Wine-Bug: https://bugs.winehq.org//show_bug.cgi?id=52507
Signed-off-by: Rémi Bernon <rbernon@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
We build with msvcrt, so we don't link against MinGW's assert() anymore.
Signed-off-by: Zebediah Figura <zfigura@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
Can be reproduced randomly even without user32:monitor, for instance
running user32:win in a fresh Win7 VM.
Wine-Bug: https://bugs.winehq.org//show_bug.cgi?id=51392
Signed-off-by: Rémi Bernon <rbernon@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
Shows that it switches focus to some other process window and back, and
only on some older Windows versions. Win10 simply returns 0 and does
nothing.
Stop calling it elsewhere, as it makes focus go to an unpredictable
window, which may very well not be ours.
Wine-Bug: https://bugs.winehq.org//show_bug.cgi?id=51391
Wine-Bug: https://bugs.winehq.org//show_bug.cgi?id=51130
Signed-off-by: Rémi Bernon <rbernon@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
There's no conversion to do, the device name in the filter is ignored.
Signed-off-by: Rémi Bernon <rbernon@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
When an application handles WM_NCPAINT by itself, standard scroll bars are not drawn even if
WS_HSCROLL or WS_VSCROLL is used. In this case, when handling WM_NCMOUSEMOVE, DefWinProc() shouldn't
repaint standard scroll bars to reflect hovered state, otherwise, DefWinProc() draws over the custom
non-client area painted by the application. This is also the behavior on XP.
Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=51900
Signed-off-by: Zhiyi Zhang <zzhang@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
Mostly to test that standard scroll bars doesn't get painted in hovered state if they weren't
previously painted by DefWinProcA/W().
Signed-off-by: Zhiyi Zhang <zzhang@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
The DS4 controllers are sending 563 bytes HID reports by default, this
translates to WM_INPUT messages larger than the default message size.
We would otherwise need an additional server roundtrip on each message,
and these devices are also known to be very verbose and continuously
send HID reports, so we really don't want it.
Signed-off-by: Rémi Bernon <rbernon@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
When changing DPI, a "fix blurry apps" popup may appear on Windows 10. The popup may interfere with
other tests as it steals focus, causing them to fail. So set IgnorePerProcessSystemDPIToast to 1 to
temporarily disable the popup.
Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=52108
Signed-off-by: Zhiyi Zhang <zzhang@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
DefWindowProc() does not propagate the wparam; it updates it instead.
Spotted by YAL.
Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=52327
Signed-off-by: Huw Davies <huw@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
Device context bit depths other than 32-bit are emulated and the real bit depth for display DCs is
still 32-bit. Thus, a 32-bit DDB should be allowed to be selected into a display compatible DC even
if the DC reports a different bit depth.
Fix a regression from d171d11.
Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=51805
Signed-off-by: Zhiyi Zhang <zzhang@codeweavers.com>
Signed-off-by: Huw Davies <huw@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
This is a preparation for a patch that changes setupapi to return lowercased
device paths, which is the source of RawInput device name here. On Windows
RawInput returns mostly uppercase names (except the GUID) and we want to keep
that behavior.
Signed-off-by: Rémi Bernon <rbernon@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
Spotted by John Sullivan.
Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=52207
Signed-off-by: Zebediah Figura <zfigura@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
The given bitmap is in DDB format, not DIB format.
Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=51296
Signed-off-by: Zebediah Figura <zfigura@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
Up to once every 2s.
This fixes a regression from c2c78a2fe0,
which caused severe FPS drop in Samurai Shodown.
Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=51282
Signed-off-by: Rémi Bernon <rbernon@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
For icons/cursors created by CreateIconFromResource(),
GetIconInfo() still succeed after they were destroyed,
which indicates that these icons/cursors are shared.
Signed-off-by: Ziqing Hui <zhui@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
It makes sure that shared icon/cursor created by CreateIconFromResourceEx()
won't be destroyed by Destroy{Icon,Cursor}().
Signed-off-by: Ziqing Hui <zhui@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
The tests show that shared icon created by CreateIconFromResourceEx()
won't be destroyed by DestroyIcon(). However, a second call to it for the
same icon bits will return a different handle.
Signed-off-by: Ziqing Hui <zhui@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
This was accidentally removed by 318673405c.
Signed-off-by: Jacek Caban <jacek@codeweavers.com>
Signed-off-by: Huw Davies <huw@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
Currently there is no need to hook DefDlgProc(). Tests show that dialog theming shouldn't be
implemented in DefDlgProc().
Also fix a double free bug because WM_DESTROY in UXTHEME_DefDlgProc() calls CloseThemeData() even
when the theme handle is not opened when handling WM_CREATE in UXTHEME_DefDlgProc(). The bug can
be demonstrated by running comctl32/tests/propsheet.c tests with Light theme.
Signed-off-by: Zhiyi Zhang <zzhang@codeweavers.com>
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
Jacek Caban