- Jul 15, 2022
-
-
Alexandre Julliard authored
Signed-off-by: Alexandre Julliard <julliard@winehq.org>
-
There can be multiple compat modes defined, separated by commas, but also surrounded by whitespace. The highest mode in the list is picked as the document mode, with 'edge' being the highest mode available. It stops as soon as an invalid mode is found in the list and returns whatever highest mode was found until then. Signed-off-by: Gabriel Ivăncescu <gabrielopcode@gmail.com>
-
For documents exposing a <!DOCTYPE> node but no specific X-UA-Compatible, mshtml defaults to IE7 compat mode, unless the app is Internet Explorer and is in Internet URL Zone. This checking for the `iexplore.exe` app name seems hardcoded into mshtml; the FeatureControl registry keys do not affect this directly, and none are set by default for iexplore.exe that would affect this, anyway. Signed-off-by: Gabriel Ivăncescu <gabrielopcode@gmail.com>
-
-
-
Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=39987 Signed-off-by: Zhiyi Zhang <zzhang@codeweavers.com>
-
Signed-off-by: Zhiyi Zhang <zzhang@codeweavers.com>
-
Signed-off-by: Zhiyi Zhang <zzhang@codeweavers.com>
-
The third parameter is of type PCUITEMID_CHILD_ARRAY, not PCUITEMID_CHILD_ARRAY *. Signed-off-by: Zhiyi Zhang <zzhang@codeweavers.com>
-
FWF_SHOWSELALWAYS is deprecated and has no effect. Manual tests show that LVS_SHOWSELALWAYS is always used. Signed-off-by: Zhiyi Zhang <zzhang@codeweavers.com>
-
LVIS_FOCUSED has to be added to both stateMask and state to set focused state for list views. Signed-off-by: Zhiyi Zhang <zzhang@codeweavers.com>
-
The value of SVSI_EDIT is 0x3. Signed-off-by: Zhiyi Zhang <zzhang@codeweavers.com>
-
Signed-off-by: Dmitry Timoshkov <dmitry@baikal.ru> Signed-off-by: Huw Davies <huw@codeweavers.com> Signed-off-by: Alexandre Julliard <julliard@winehq.org>
-
-
Signed-off-by: Ziqing Hui <zhui@codeweavers.com>
-
Signed-off-by: Ziqing Hui <zhui@codeweavers.com>
-
Signed-off-by: Ziqing Hui <zhui@codeweavers.com>
-
Signed-off-by: Ziqing Hui <zhui@codeweavers.com>
-
Signed-off-by: Byeongsik Jeon <bsjeon@hanmail.net>
-
10 DLUs is the recommended height for check boxes on Windows. Signed-off-by: Zhiyi Zhang <zzhang@codeweavers.com>
-
10 DLUs is the recommended height for check boxes on Windows. 8 DLUs is not enough and may cause shrinking in HiDPI settings. Signed-off-by: Zhiyi Zhang <zzhang@codeweavers.com>
-
10 DLUs is the recommended height for check boxes on Windows. 8 DLUs is not enough and may cause shrinking in HiDPI settings. Signed-off-by: Zhiyi Zhang <zzhang@codeweavers.com>
-
The destination rectangle may be smaller than that of theme parts, in which case, the draw size returned is adjusted to fit the destination rectangle. Without passing a destination rectangle to GetThemePartSize(), the function assumes that there is no limit for destination. This fixes winecfg check boxes missing the upper edge in 192 DPI. You may notice that the check box bitmap is slightly shrank and so the shape is a bit off. However, this is what happens on Windows as well when check box height is set to 8 DLUs. GetThemePartSize() without destination rectangles are also used in other controls, but manual tests showed that those controls don't shrink their theme background as check boxes do so they are left unchanged. Signed-off-by: Zhiyi Zhang <zzhang@codeweavers.com>
-
-
It's already set previously in sock_poll_event(). Moreover, complete_async_connect() will not always be called.
-
-
- Jul 14, 2022
-
-
Signed-off-by: Zhiyi Zhang <zzhang@codeweavers.com>
-
Signed-off-by: Georg Lehmann <dadschoorse@gmail.com>
-
The gitlab scripts operate without access to the wine tree. Requiring a trailing slash on patterns simplifies the logic to automatically assign reviewers to MRs.
-
Signed-off-by: Zhiyi Zhang <zzhang@codeweavers.com>
-
Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=53289 Signed-off-by: Jinoh Kang <jinoh.kang.kr@gmail.com>
-
Signed-off-by: Jinoh Kang <jinoh.kang.kr@gmail.com>
-
Signed-off-by: Jinoh Kang <jinoh.kang.kr@gmail.com>
-
-
send_socket does some extra bookkeeping that's currently missing from the register_async path. Instead of adding it to sock_queue_async(), let's just centralize all send requests so that they go through send_socket.
-
recv_socket does some extra bookkeeping that's currently missing from the register_async path. Instead of adding it to sock_queue_async(), let's just centralize all recv requests so that they go through recv_socket.
-
-
Besides being unnecessary, this triggers a bug in radeonsi (since fixed upstream), which can cause failures in d3d10core and d3d11 tests.