b8adf90aa3
cmake: check again for unknown documentation commands
2023-02-09 17:20:28 -08:00
cc0fa00049
cmake: include directories of (generated) dependencies as SYSTEM
2023-02-09 17:20:28 -08:00
a9650d440a
Fixed MFI controller being opened while HIDAPI controller was being opened
...
This was the callstack:
frame #3 : 0x00000001004e1930 libSDL3.1.0.0.dylib`IOS_AddJoystickDevice(controller=0x0000600003b0c000, accelerometer=SDL_FALSE) at SDL_mfijoystick.m:528:14
frame #4 : 0x00000001004e1a54 libSDL3.1.0.0.dylib`__IOS_JoystickInit_block_invoke(.block_descriptor=0x0000000100547760, note=@"GCControllerDidConnectNotification") at SDL_mfijoystick.m:673:45
frame #5 : 0x000000018601e578 CoreFoundation`__CFNOTIFICATIONCENTER_IS_CALLING_OUT_TO_AN_OBSERVER__ + 128
frame #6 : 0x00000001860bc074 CoreFoundation`___CFXRegistrationPost_block_invoke + 88
frame #7 : 0x00000001860bbfbc CoreFoundation`_CFXRegistrationPost + 440
frame #8 : 0x0000000185fefbac CoreFoundation`_CFXNotificationPost + 708
frame #9 : 0x0000000186edc72c Foundation`-[NSNotificationCenter postNotificationName:object:userInfo:] + 88
frame #10 : 0x000000019b054a18 GameController`__60-[_GCControllerManagerAppClient _onqueue_publishController:]_block_invoke + 156
frame #11 : 0x0000000185dc19dc libdispatch.dylib`_dispatch_call_block_and_release + 32
frame #12 : 0x0000000185dc3504 libdispatch.dylib`_dispatch_client_callout + 20
frame #13 : 0x0000000185dd1d1c libdispatch.dylib`_dispatch_main_queue_drain + 928
frame #14 : 0x0000000185dd196c libdispatch.dylib`_dispatch_main_queue_callback_4CF + 44
frame #15 : 0x000000018606ad6c CoreFoundation`__CFRUNLOOP_IS_SERVICING_THE_MAIN_DISPATCH_QUEUE__ + 16
frame #16 : 0x00000001860287ec CoreFoundation`__CFRunLoopRun + 2036
frame #17 : 0x00000001860278a4 CoreFoundation`CFRunLoopRunSpecific + 612
frame #18 : 0x00000001003b1194 libSDL3.1.0.0.dylib`process_pending_events at hid.c:509:9
frame #19 : 0x00000001003aebe8 libSDL3.1.0.0.dylib`PLATFORM_hid_open_path(path="USB_054c_05c4_0x11a104290", bExclusive=0) at hid.c:823:2
frame #20 : 0x00000001003b051c libSDL3.1.0.0.dylib`SDL_hid_open_path_REAL(path="USB_054c_05c4_0x11a104290", bExclusive=0) at SDL_hidapi.c:1419:19
frame #21 : 0x00000001004dabdc libSDL3.1.0.0.dylib`HIDAPI_SetupDeviceDriver(device=0x0000600003518000, removed=0x000000016fdfee3c) at SDL_hidapijoystick.c:399:19
frame #22 : 0x00000001004da890 libSDL3.1.0.0.dylib`HIDAPI_AddDevice(info=0x000060000212c2d0, num_children=0, children=0x0000000000000000) at SDL_hidapijoystick.c:843:5
frame #23 : 0x00000001004d9148 libSDL3.1.0.0.dylib`HIDAPI_UpdateDeviceList at SDL_hidapijoystick.c:1000:21
frame #24 : 0x00000001004d9940 libSDL3.1.0.0.dylib`HIDAPI_JoystickDetect at SDL_hidapijoystick.c:1205:13
frame #25 : 0x00000001003bc6d8 libSDL3.1.0.0.dylib`SDL_UpdateJoysticks_REAL at SDL_joystick.c:1703:9
frame #26 : 0x00000001003a13a8 libSDL3.1.0.0.dylib`SDL_PumpEventsInternal(push_sentinel=SDL_FALSE) at SDL_events.c:855:9
frame #27 : 0x00000001003a1340 libSDL3.1.0.0.dylib`SDL_PumpEvents_REAL at SDL_events.c:879:5
frame #28 : 0x000000010038b380 libSDL3.1.0.0.dylib`SDL_PumpEvents at SDL_dynapi_procs.h:572:1
frame #29 : 0x0000000100004524 testgamepad`loop + 40
frame #30 : 0x00000001000063d8 testgamepad`main + 2140
2023-02-09 12:07:11 -08:00
ae91dcbcf4
fix dynapi after commit dc74a67a
2023-02-09 22:50:00 +03:00
dc74a67a57
SDL_UnlockTexture: change from 'int' to 'void'
2023-02-09 20:17:27 +01:00
cee245b6a9
SDL_*SceenSaver(): change return value to int. // add SDL_Unsupported() errors
2023-02-09 08:16:46 -08:00
e2e5e670bf
Remove returns int from Destroy/Close/Unlock functions
2023-02-09 07:57:14 -08:00
3bd737d44c
Add error returns to void functions that can fail/set errors.
...
This takes care of the last set of void functions that could
potentially be shifted to instead return an int indicating success and
setting an error in case of an error.
2023-02-09 07:23:21 -08:00
b305d9e3c0
Change return type from void to int for audio function. Eventually set invalid parameter error.
2023-02-08 17:43:44 -08:00
824b9b0a58
Removed SDL_GetDisplayDPI()
...
This function wasn't consistently correct across platforms and devices.
If you want the UI scale factor, you can use display_scale in the structure returned by SDL_GetDesktopDisplayMode(). If you need an approximate DPI, you can multiply this value times 160 on iPhone and Android, and 96 on other platforms.
2023-02-08 17:35:54 -08:00
b7c6fec10a
Make sure we synchronously re-enable USB reporting mode for Switch controllers (thanks @SortaCore!)
...
Fixes https://github.com/libsdl-org/SDL/issues/3450
2023-02-08 14:22:39 -08:00
50575ed223
Note the Google Stadia controller doesn't support rumble over Bluetooth
2023-02-08 14:16:24 -08:00
00e02de77b
Fixed mis-detecting the share button on the Xbox controller over NDIS
2023-02-08 14:16:21 -08:00
d1c72bb0bc
Dynamically update the serial number if it isn't available at first
...
This happens for Xbox One controllers using newer firmware connected over NDIS
2023-02-08 14:16:17 -08:00
c5c94a6be6
Change return type from void to int, for functions that set an error
...
(SDL_SetError(), SDL_OutOfMemory(), SDL_Unsupported(), SDL_InvalidParam())
Update prototype to forward errors to generic layer, for the functions:
MoveCursor, WarpMouse, GL_DeleteContext, GetDisplayModes.
Check invalid parameter in SDL_SetTextInputRect() generic layer.
2023-02-07 13:51:45 -08:00
6c37d5b57f
x11: Cast the dot clock value to 64-bit when calculating the refresh rate
...
The Xrandr dot clock value is declared as an unsigned long and the result when multiplying by 100 can overflow on a 32-bit system. Explicitly cast it to Sint64 to ensure that no overflow will occur.
2023-02-07 13:39:38 -05:00
20a4e31a12
Re-added the internal hint SDL_BORDERLESS_WINDOWED_STYLE
...
sdl2-compat should set this to "0"
2023-02-07 09:47:17 -08:00
fa8fba3812
Add clang-tidy config and CMake/CI support.
2023-02-06 21:02:25 -08:00
5fded632d6
Added support for the Turtle Beach REACT-R and Recon Xbox controllers
2023-02-06 20:14:12 -08:00
d603371848
video: Try to get the display for fullscreen windows via the window position
...
Attempt to retrieve the display for fullscreen windows using the window position so that the correct display ID is returned if an exclusive fullscreen window is moved to another display.
2023-02-06 17:18:35 -08:00
a357021800
Use GNU sed if available
...
Fixes update-copyright.sh on macOS, which doesn't support the -b option.
2023-02-06 15:47:16 -08:00
0208a0eeb3
Added support for the Logitech ChillStream
2023-02-06 13:28:59 -08:00
7f3036c353
make sure to not enable SDL_PASSED_BEGINTHREAD_ENDTHREAD for WinRT
2023-02-07 00:22:37 +03:00
b82a8b76b1
windows/SDL_systhread.c: remove some dead code.
2023-02-07 00:22:20 +03:00
22bafe8729
Removed debug code accidentally committed
2023-02-06 11:23:53 -08:00
f8b41919da
Fixed warning running a command queue without any vertex operations
2023-02-06 11:22:00 -08:00
375bde8a4f
[SDL3] Cleanup void functions ( #7253 )
...
Some functions that do call SDL_SetError but return void changed to instead return non-zero in case of errors.
2023-02-06 08:53:52 -08:00
9d8f98c735
Don't change presentation modes when resizing the logical target
...
Fixes https://github.com/libsdl-org/SDL/issues/7251
2023-02-06 08:51:58 -08:00
6f0acca8bd
Return the SDL_SetError value
2023-02-06 11:42:38 +01:00
a077cc8e4d
[SDL3] macOS SetCursor performance fix ( fixes #7151 ) ( #7249 )
...
* Setting the same mouse cursor twice is a no-op
* Cocoa: Call [NSCursor set] to change mouse cursor
The previous way, changing the mouse cursor was handled by invalidating
the mouse cursor rectangles and then recreating them (with the new
cursor) the next event loop. This is extremely slow; sometimes it can
take over a millisecond! With [NSCursor set] it happens instantly and
very quick performance-wise.
The downside is that it sets the cursor for the whole screen, so we
have some guards in place to change it to the system cursor if
the mouse moves outside the window or the window loses focus.
* Cocoa: Remove unneeded resetCursorRects: function
2023-02-05 17:58:33 -08:00
fbe0352764
SDL_DestroyRenderer: drop all commands first
...
and don't queue new command from SDL_DestroyTexture() (memory leak)
2023-02-05 13:35:08 -08:00
e1d79b418c
fix build error due to -Wshadow
2023-02-05 20:45:02 +03:00
653f2c4ba3
Made the render tests less verbose in the successful case
2023-02-05 09:11:00 -08:00
ec1b42557f
Fixed memory leak at shutdown (thanks @1bsyl!)
...
Clean up the textures first, as destroying the current render target will queue viewport and clip rectangle commands
2023-02-05 09:10:21 -08:00
5d1006657a
SDL_migration.cocci: remove metavariable warning
2023-02-05 17:48:18 +01:00
6f06e4bc80
Add Linux mapping for 8BitDo Pro 2 Wired Controller for Xbox
...
Without these mappings, this controller "kinda" works out of the box:
- `SDL_GameControllerMapping()` works because it will notice "Xbox" in
the name and use the default XInput mappings
- `SDL_GameControllerMappingForGUID()` will not find any mapping
lsusb:
```
ID 2dc8:2000 8BitDo 8BitDo Pro 2 Wired Controller for Xbox
```
In Linux this controller is supported by two drivers:
- `xpad` (built-in to the kernel), exposes the controller name from the
USB descriptor and the GUID starts with 03 (0x03 = BUS_USB)
- `xone` (https://github.com/medusalix/xone ), exposes a virtual
controller which is always named "Microsoft X-Box One pad" and the
GUID starts with 06 (0x06 = BUS_VIRTUAL)
This commit adds the 2 GUIDs from both drivers so mappings will always
be found and the real controller name will always be reported.
(cherry picked from commit 4266cf8504c08b17ef55be11c27f2fc7a17b6edf)
2023-02-05 08:45:24 -08:00
1c03ddefc3
Sorted controller list
2023-02-05 08:43:26 -08:00
d563f38a0d
Note that the renderer will by default scale from window coordinates to pixels
2023-02-05 08:41:35 -08:00
52cd9fcbb0
Added Elecom 8button gamepad
2023-02-05 08:41:07 -08:00
85143c28b5
SDL_migration.cocci: fix syntax
2023-02-05 09:37:37 +01:00
4a16d8d00e
Merge pull request #7238 from Kontrabant/wayland_fix_portrait
...
wayland: Transform mode values for native portrait displays
2023-02-04 11:42:43 -05:00
b2d8a1a58c
wayland: Transform mode values for native portrait displays
...
Portrait displays may have native, physical resolutions that are taller than wide. Reverse the mode dimensions when dealing with these displays as well as those rotated via software means.
2023-02-04 11:40:55 -05:00
02fbf2cf99
replaced line comments in public header.
2023-02-04 10:02:10 +03:00
2a83093b36
Use a reasonable upper bound on the number of pixels we'll try to draw when traversing a line
...
Fixes https://github.com/libsdl-org/SDL/issues/6116
2023-02-03 20:53:30 -08:00
d6fdb842b0
include: add comment documenting the change(s) to SDL_opengl_glext.h
2023-02-04 02:44:10 +01:00
4dea13e02f
Adds Stadia BT mapping
2023-02-03 15:01:49 -08:00
a34a84ba98
Rename int versions of the SDL2 render functions ( #7235 )
...
This makes it clear what the new versions are, and in the case of SDL_RenderDrawPoint() and SDL_RenderDrawLine(), the coccinelle script actually does the (float) casts for you.
2023-02-03 14:55:32 -08:00
14a4ce8b59
Fixed SDL_ScaleMode values for consistency
2023-02-03 14:20:51 -08:00
e5edce8e75
Fixed permissions on src/render/SDL_render.c ( #7232 )
2023-02-03 13:14:48 -08:00
cb6b8b0132
Simplify flags testing ( #7220 )
2023-02-03 13:08:42 -08:00
Anonymous Maarten
Sam Lantinga
Ozkan Sezer
Sylvain
Linus Probert
Frank Praznik
Pierre Wendling
Sean Ridenour
Leonardo Brondani Schenkel
Antonis Geralis
Alynne