- Apr 09, 2025
-
-
This partially reverts 52146f62. As stated in that commit, a WM_PAINT will eventually be generated via window exposure. However, some applications rely on WM_PAINT being sent synchronously. As create_visible_dialog_seq shows, this happens reliably, and the behaviour is specific to dialogs. Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=35652
-
-
It confuses Unity and breaks input after alt-tab. Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=48121
-
It confuses Unity and breaks input after alt-tab. Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=48121
-
-
Otherwise, on systems with only Wine provided fonts, we end up selecting first non-vertical font ignoring other properties.
-
-
-
Alexandre Julliard authored
-
Alexandre Julliard authored
-
Alexandre Julliard authored
-
Alexandre Julliard authored
-
In Korean input, when ime is in the composition state, if a non-printable key (ENTER, HOME, etc) is pressed, it returns a result_string and passes the key to WinProc with the WM_IME_KEYDOWN message. This can be seen in ‘imm32/test/imm32.c::test_ga_na_da()’. In the current winemac.drv, if I select the “Korean 2-Set Keyboard” as the input source and type ‘r-k-ENTER’, it only returns the result string ‘0xAC00’. No WM_IME_KEYDOWN message is generated. At this point, the WINE_IME_POST_UPDATE call occurs and macdrv_ImeProcessKey returns FALSE. A FALSE return value means that the key has not been consumed by ime and the key should be passed to the WinProc as a raw WM_KEYDOWN. However, because the ime_update call have occurred, ImeProcessKey should return TRUE and a raw WM_KEYDOWN message should be generated via a different path. So a WM_IME_KEYDOWN message is required. This patch allows the macdrv_ImeProcessKey return value to be passed to ImeToAsciiEx() to determine if a WM_IME_KEYDOWN message should be generated.
-
In winemac.drv, there are multiple ime update calls occur during ImeProcessKey. These multiple ime update calls need to be properly merged into ‘data->update’. However, currently win32u/imm.c::post_ime_update() only keeps the last ime update call. Valid rules are: - The previous comp_str is discarded when the new ime update call occurs, so it is only valid if the last ime update call comp_str is not NULL. - The previous result_str is kept if the new ime update call result_str is NULL. - If both the previous and the new ime update call result_str are not NULL, then concatenate them. Test key sequences are: - "Japanese Romanji" : 'nihongo-SPACE-n' - "Korean 2-Set Keyboard" : 'r-k-s-k' - "Korean 2-Set Keyboard" : 'r-k-1" or 'r-k-SPACE'
-
-
-
Signed-off-by:
Nikolay Sivov <nsivov@codeweavers.com>
-
- Apr 08, 2025
-
-
Several window managers have a troublesome sticky maximized state bit, and it's better not to mess with it when asking for fullscreen window is enough. Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=58066
-
-
-
Fixes: 84708d10
-
-
This will later allow for optimization by limiting the area we consider to the region's bounding rectangle.
-
Signed-off-by:
Eric Pouech <epouech@codeweavers.com>
-
Fixing a couple of bugs in the process. Signed-off-by:
Eric Pouech <epouech@codeweavers.com>
-
(making use of already set values by Dwarf backend). Signed-off-by:
Eric Pouech <epouech@codeweavers.com>
-
Signed-off-by:
Eric Pouech <epouech@codeweavers.com>
-
Signed-off-by:
Eric Pouech <epouech@codeweavers.com>
-
Alexandre Julliard authored
-
Alexandre Julliard authored
-
Alexandre Julliard authored
-
Alexandre Julliard authored
-
-
Signed-off-by:
Nikolay Sivov <nsivov@codeweavers.com>
-
Signed-off-by:
Nikolay Sivov <nsivov@codeweavers.com>
-
Signed-off-by:
Nikolay Sivov <nsivov@codeweavers.com>
-
Signed-off-by:
Nikolay Sivov <nsivov@codeweavers.com>