On Tue Sep 16 12:48:37 2025 +0000, Nikolay Sivov wrote:
> Here it's fine to remove it, because original text does not have it.
The “fuzzy” translations block the translation after the fact. It would be better to delete the fuzzy translations and do a side-by-side translation so that everything is taken into account.
For example, in “winecfg,” under “integration with the desktop,” you will see what I mean.
--
https://gitlab.winehq.org/wine/wine/-/merge_requests/8857#note_115961
This fixes Trials Fusion often crashing when disconnecting a controller while there are more still connected.
--
v17: ntoskrnl/tests: Use the 'Nt' version of the CancelIo APIs.
ntoskrnl/tests: Test the thread ID the cancellation routine runs from.
ntoskrnl/tests: Add more cancellation tests.
ntoskrnl/tests: Fix tests on current Windows 10 / 11.
ntdll/tests: Test IOSB values of the cancel operation.
ntdll/tests: Add more NtCancelIoFile[Ex]() tests.
ntdll: Wait for all asyncs to handle cancel in NtCancelIoFile().
server: Factor out a cancel_async() function.
server: Introduce a find_async_from_user helper.
ntdll: Factor out a cancel_io() function.
https://gitlab.winehq.org/wine/wine/-/merge_requests/7797
When we were able to load it at least.
EGL has some advantages, like not requiring a specific pixel format on the windows, it will share most of its problems with winewayland (and wineandroid although we may also assume GL doesn't work anymore there at this point) so anything fixed with it could benefit all. It will also be required for shared resources and compositing when Vulkan is unavailable.
This might be a bit early still, I'm not sure it is on par when it comes to child window rendering synchronization. However this doesn't drop GLX either, and instead just changes the default. I think it would be useful to switch early though, at least to have more time to settle (and eventually decide to switch back during code freeze if it proves a bad idea).
--
https://gitlab.winehq.org/wine/wine/-/merge_requests/8977
Current implementation of system.fill_battery_state only reads "charge\__" and "current\__" attributes. Some batteries expose different attributes, namely "power\__" and "energy\__". Added the required logic to support both types.
Further, when rewriting the function I also added support for systems with multiple batteries. Windows documentation was not clear but from what I gathered struct SYSTEM_BATTERY_STATE sets Charging and Discharging if ANY battery is charging/discharging. So, to check if the overall system is charging/discharging only Rate can be used.
Also cleaned the code related to "AC". There was a "fixme" there which I didn't fully undertand as SYSTEM_BATTERY_STATE.AcOnLine == True if any AC is present, so the proper logic was already implemented
--
v15: ntdll: improve readibility of linux_battery capacity units
ntdll: Support multi-battery systems and batteries that expose energy_* and power_*
ntdll: Add enum, struct for battery status, information
ntdll: Fix test with proper battery EstimatedTime expected value
ntdll: Update tests to use just Rate for charging logic
https://gitlab.winehq.org/wine/wine/-/merge_requests/8460
Signed-off-by: Nikolay Sivov <nsivov(a)codeweavers.com>
--
v2: dwrite/opentype: Use mask shifts only for non-zero masks.
dwrite/layout: Do not shadow output parameter.
dwrite/layout: Always initialize effective run bounding box.
dwrite: Always initialize 'contours' flag.
https://gitlab.winehq.org/wine/wine/-/merge_requests/8981