|
808c3ae6
|
2021-07-24T10:36:56
|
|
Allow mapping Android controllers that just have a D-PAD and face buttons
This fixes bug https://github.com/libsdl-org/SDL/issues/4327
|
|
856c87fc
|
2021-07-24T17:32:05
|
|
Sync wiki -> header
|
|
|
d89f4b3a
|
2021-07-24T10:29:34
|
|
Note that mouse warping doesn't work over Microsoft Remote Desktop
Fixes the documentation portion of bug https://github.com/libsdl-org/SDL/issues/4206
|
|
|
94b7a876
|
2021-07-24T09:10:18
|
|
Added SDL_GameControllerType enumerations for the Amazon Luna and Google Stadia controllers
Fixes bug https://github.com/libsdl-org/SDL/issues/4019
|
|
6956f4aa
|
2021-07-24T11:06:40
|
|
cmake: use CMAKE_INSTALL_LIBDIR instead of hardcoding "lib", etc.
These came from HaikuPorts, specifically this patchset:
https://github.com/haikuports/haikuports/blob/462947dd4f285a885d8011cade7cd0f7151cfe74/media-libs/libsdl2/patches/libsdl2-2.0.14.patchset
This is just the part that is _not_ Haiku-specific. I wanted this in a
separate commit though, since I'm not really a CMake expert; if this causes
problems and we need to bisect to it, we won't be confused about it being a
Haiku issue when we get here.
Fixes #4092.
(and probably others.)
|
|
|
403afdb5
|
2021-07-24T11:02:27
|
|
cmake: Haiku fixes.
These came from HaikuPorts, specifically this patchset:
https://github.com/haikuports/haikuports/blob/462947dd4f285a885d8011cade7cd0f7151cfe74/media-libs/libsdl2/patches/libsdl2-2.0.14.patchset
This is just the Haiku-specific bits; the more general parts about install
dirs will be split into a different commit here.
Reference issue #4092.
|
|
|
7d8f34cb
|
2021-07-24T01:19:58
|
|
Updated .gitignore from @Zangetsu38 - thanks!
|
|
e6e8fe6d
|
2021-07-23T10:45:03
|
|
hidapi_switch: Implement analog rumble
|
|
|
669714ca
|
2021-07-23T23:51:35
|
|
Added the ThrustMaster T.16000M as a flightstick
|
|
|
384d86b5
|
2021-07-23T23:48:23
|
|
Fixed compile warnings with Visual Studio
|
|
|
adb75d64
|
2021-07-23T23:45:18
|
|
Fixed hotplug detection not working on UWP or when SDL isn't pumping Windows events
This fixes bugs:
https://github.com/libsdl-org/SDL/issues/4321
https://github.com/libsdl-org/SDL/issues/4147
Thanks to DJm00n for the suggestion!
|
|
7b0092db
|
2021-07-23T18:58:25
|
|
vita: touch: Mark the back panel as an indirect absolute touch device
It's not a touchscreen, but an additional touch panel on the back of
the device.
|
|
|
422b053b
|
2021-07-23T16:32:35
|
|
Added support for the full line of current PowerA Xbox Series X controllers
|
|
|
7edffc57
|
2021-07-23T18:29:04
|
|
Sync wiki -> header
|
|
|
9632fb33
|
2021-07-23T12:48:28
|
|
README-dynapi.md: Cleaned up markdown formatting.
|
|
|
c06448fe
|
2021-07-23T16:39:05
|
|
Sync wiki -> header
|
|
|
48993a7f
|
2021-07-23T16:14:49
|
|
Sync wiki -> header
|
|
15ed6653
|
2021-07-22T20:50:02
|
|
Fixes to kmsdrm dynamic loading support.
Fixes: https://github.com/libsdl-org/SDL/issues/4520
|
|
48bbe31b
|
2021-07-20T21:45:21
|
|
Add PreSwapCallback for updating SceCommonDialog
|
|
|
9c1306bc
|
2021-07-20T21:44:16
|
|
Add elgext.h header
|
|
4ada14a2
|
2021-07-21T13:39:10
|
|
Replace libc functions to SDL equivalent in wayland video subsystem
Wayland video subsystem uses a mix of libc and SDL function.
This patch switches libc functions to SDL ones and fixes a mismatch in memory
allocation/dealoccation of SDL_Cursor in SDL_waylandmouse.c (calloc on line 201
and SDL_free on line 313) which caused memory corruption if custom memory
allocator where provided to SDL.
|
|
f806064e
|
2021-07-19T13:12:10
|
|
Remove 'support' for rgb/bgr textures, that was causing issues with them
|
|
d6641d71
|
2021-07-19T11:52:57
|
|
kmsdrm: Explicitly set the surface transform and alpha mode
This is required by the specification. Set these to sensible defaults.
Fixes a validation layer error that occurs otherwise.
|
|
|
76007427
|
2021-07-19T11:26:54
|
|
kmsdrm: Properly zero-initialize CreateInfo structs
As written, these contain undefined stack contents, which in practice
causes crashes/hangs and/or triggers the validation layers (they
complain about `pNext` and `flags` not being NULL).
|
|
9170c0c7
|
2021-07-15T18:35:20
|
|
CI: Add Windows ARM64 builds
|
|
|
37d4f003
|
2021-07-14T17:09:05
|
|
doxygen: Removed some cruft that accumulated during wiki migration.
|
|
|
c07e5e99
|
2021-07-14T17:07:41
|
|
wikiheaders.pl: Add --copy-to-header option.
...because I keep mistyping it without the 's' on the end. :/
|
|
|
c88eb7a8
|
2021-07-14T17:07:04
|
|
Sync wiki -> header.
|
|
|
f8c1fc49
|
2021-07-14T14:15:30
|
|
doxygen: Fix all the "\returns" so they work as part of complete sentences.
|
|
|
8ec9fbdd
|
2021-07-14T11:58:57
|
|
wikiheaders.pl: pull in full sections from headers, fix more whitespace.
|
|
|
17b4a67a
|
2021-07-14T11:45:06
|
|
doxygen: We use "\returns" as part of the sentence.
|
|
|
bab0eb71
|
2021-07-14T10:03:31
|
|
wikiheaders.pl: Fix some wordwrap whitespace issues.
|
|
|
04196a46
|
2021-07-14T09:59:52
|
|
doxygen: wrap some game controller code examples in "```c" blocks.
|
|
|
7390bc0e
|
2021-07-14T09:36:07
|
|
wikiheaders.pl: Fix \sa tags that (incorrectly!) have "()" appended.
|
|
|
d115857c
|
2021-07-14T08:11:18
|
|
wikiheaders.pl: A bunch of work on correct wordwrap and (de)wikify.
|
|
|
0ac13dc2
|
2021-07-14T08:01:55
|
|
doxygen: Tweak SDL_FlashWindow brief description.
So "e.g." doesn't look like the end of the sentence to wikiheaders.pl.
|
|
|
1f17bc54
|
2021-07-13T12:25:23
|
|
wikiheaders: look for "c++" before "c" in regular expression.
|
|
|
269d0773
|
2021-07-13T11:11:55
|
|
wikiheaders: Ignore doxygen comments that aren't "/**" on a line by itself.
Makes this ignore /** single line comments */
|
|
|
8130950e
|
2021-07-13T11:11:33
|
|
wikiheaders: trim whitespace at edges after wordwrapping.
|
|
0f107945
|
2021-07-13T23:07:54
|
|
Remove outdated comment about SDL_SoftStretch
|
|
|
817976da
|
2021-07-13T13:45:28
|
|
VITA: Rewrite and fix RenderCopyEx rotation
|
|
|
dc6f0443
|
2021-07-12T17:57:21
|
|
Added mapping for the Xbox Series X controller on tvOS
|
|
|
947ad6e1
|
2021-07-12T17:57:21
|
|
Added support for the Xbox Series X controller on iOS and tvOS
|
|
|
297ce148
|
2021-07-11T17:38:33
|
|
CI: Add Android builds to GitHub Actions
|
|
7fe6dbaf
|
2021-07-11T15:25:24
|
|
Reset rawInputEnableCount when RawInput is not supported
|
|
32e736d2
|
2021-07-10T11:39:51
|
|
CI: add github actions for windows x86
|
|
|
5ac81c75
|
2021-07-10T10:05:40
|
|
fixed typo in previous commit
|
|
|
a809d626
|
2021-07-10T10:01:40
|
|
cmake: set SDL_JOYSTICK_RAWINPUT for windows builds
Fixes: https://github.com/libsdl-org/SDL/issues/4412
|
|
|
24059a19
|
2021-07-09T18:11:42
|
|
The RAWINPUT driver is no longer tied to HIDAPI in any way
|
|
e52733af
|
2021-07-09T13:39:05
|
|
Use CMAKE_CURRENT_BINARY_DIR instead of CMAKE_BINARY_DIR
This allows for building with cmake's add_subdirectory() function.
|
|
|
f93e9047
|
2021-07-08T18:44:50
|
|
Implemented the timeout in hid_read_timeout() on Android
This fixes detecting PS5 controllers connected over USB
|
|
e2234ee9
|
2021-05-16T06:33:29
|
|
CMake: Fix SOVERSION/SONAME of non-release versions
When building SDL2 from git with CMake, you got libSDL2-2.0.so.1
instead of .0 (as it's the case when building with autotools).
This was caused by using LT_REVISION instead of LT_MAJOR for SOVERSION.
fixes #4310
|
|
|
c20207d7
|
2021-05-04T10:00:16
|
|
SDL2Config.cmake: Clean it up a bit, remove gcc-style SDL2_LIBRARIES
The GCC-style SDL2_LIBRARIES were lacking the rpath on Linux, which
seems to be implicitly set when linking path/to/libSDL2-2.0.so.0.*
and is explicitly set in the SDL2_LIBRARIES in sdl2-config.cmake
(from some autotools variable), so I removed that hack and the format
remains sth like "path/to/libSDL2main.a;path/to/libSDL2-2.0.so.0.14.1".
It's still in the revision history in case it turns out that some
platform really needs the "-L/path/to/bla/lib -lSDL2main -lSDL2" format
|
|
|
501fe465
|
2021-04-21T06:34:46
|
|
sdl2-config.cmake.in: Make SDL2::SDL2 work with MinGW, fixes #3665
It didn't work at all because shared libs defined in CMake with
add_library() need something like IMPORTED_IMPLIB (pointing to a .dll.a
or .lib for th DLLs) set to link on Windows.
But even with that it didn't work because the order of the libs is very
important: it must be -lmingw32 -lSDL2main -lSDL -mwindows - but
with normal add_library(SDL2::SDL2 SHARED IMPORTED) libs, SDL2 itself
is always linked first.
So I use an "INTERFACE" library (usually used for header-only libs), which
doesn't implicitly/automatically link anything so I can specify the whole
order of the linked libs.
(SDL2::SDL2-static is completely untested)
|
|
|
fac6e9f1
|
2021-04-21T06:27:41
|
|
sdl2-config.cmake.in: Use more CMake variables and less autotools ones
this makes it easier to create a portable sdl2-config.cmake that doesn't
hardcode its path (by replacing the hardcoded prefix with something
like "${CMAKE_CURRENT_LIST_DIR}/../../..")
|
|
|
c3426961
|
2021-04-18T17:15:34
|
|
Enhanced SDL2Config.cmake to also work on non-Windows
tested Linux; I assume most other Unices are similar; I don't have a Mac
|
|
|
3ea11f9a
|
2021-04-18T06:23:20
|
|
Make SDL2Config.cmake export SDL2_LIBRARIES etc (on Windows)
still WIP and other platforms than windows still TODO and open
questions etc
see https://github.com/libsdl-org/SDL/issues/4004
|
|
f9bf1aeb
|
2021-07-09T00:51:16
|
|
Latest Unifont with OFL is now bundled. (#4268)
* * Latest Unifont with OFL is now bundled.
* * Added a Unifont license file.
Co-authored-by: Vladislav Dmitrievich Turbanov <vladislav@turbanov.ru>
|
|
557d0f82
|
2021-03-11T13:36:56
|
|
Commenting a false alarm in WIN_GLES_SetupWindow.
When hint SDL_HINT_OPENGL_ES_DRIVER is set to "1" (e.g. for ANGLE support), assertion due to !_this->gl_config.driver_loaded can be causes while EGL is available.
|
|
|
d135c076
|
2021-07-08T13:22:41
|
|
Added SDL_GameControllerSendEffect() and SDL_JoystickSendEffect() to allow applications to send custom effects to the PS4 and PS5 controllers
See testgamecontroller.c for an example of a custom PS5 trigger effect
|
|
|
073cbc40
|
2021-07-08T21:20:02
|
|
SDL_platform.h: define TARGET_OS_SIMULATOR as 0 if not already defined.
c.f.: https://github.com/libsdl-org/SDL/issues/4475
|
|
|
be41cf2d
|
2021-07-08T21:20:02
|
|
CMake: don't blacklist TARGET_OS_SIMULATOR for Metal configuration
https://github.com/libsdl-org/SDL/commit/7fa5e95b6284f2cc7855022c9f0c2b086dcee43f enabled it.
|
|
|
97d4fe94
|
2021-07-08T20:55:10
|
|
removed extra whitespace.
|
|
|
3da6d2cd
|
2021-07-08T20:33:50
|
|
WIN_SetErrorFromHRESULT: kill CR/LF that FormatMessage sticks at the end
Fixes: https://github.com/libsdl-org/SDL_mixer/issues/320
|
|
8fb95034
|
2021-04-02T12:52:21
|
|
Avoid warping the cursor to the center of the window when not in relative_warp_mode
When relative mode is enabled and not using warp mode, the cursor is
being clipped to the window. Therefore there is no reason to restore the
cursor position to the center.
Avoiding the warp to center simplifies mouse position event flow, as we
are no longer potentially receiving mouse events for the automated
movement of the cursor and can be (mostly) assured that an incoming
event from the windowing system is that of external means.
|
|
|
4359a47b
|
2021-04-02T12:48:16
|
|
Simplify ClipCursor behaviour when RelativeMouseMode is enabled
The implementation of clip logic for relative mode seemed to
unnecessarily limit the usable area to the middle of the window, in a
2x2 pixel region. This has the adverse side effect of moving the
operating system cursor to that location, even if it is in a valid
location in the window.
While in most scenarios this is handled correctly (by storing the
original position of the cursor in the window and restoring when leaving
relative mode), there are edge cases where this clip operation can cause
WM_MOUSEMOVE to fire at a point in time where it counts as a relative
delta from SDL's perspective.
|
|
|
3720e254
|
2021-04-02T12:48:02
|
|
Trim some stray whitespace
|
|
|
17092427
|
2021-07-08T09:24:54
|
|
Fixed https://github.com/libsdl-org/SDL/issues/4475
-Wundef errors from clang-11.1 when targeting macOS
Targeting i386 against 10.8 SDK:
In file included from src/SDL_assert.c:21:
In file included from src/./SDL_internal.h:52:
In file included from include/SDL_config.h:33:
include/SDL_platform.h:73:5: error: 'TARGET_OS_TV' is not defined, evaluates to 0 [-Werror,-Wundef-prefix=TARGET_OS_]
^
1 error generated.
src/joystick/iphoneos/SDL_mfijoystick.m:38:5: error: 'TARGET_OS_IOS' is not defined, evaluates to 0 [-Werror,-Wundef-prefix=TARGET_OS_]
^
src/joystick/iphoneos/SDL_mfijoystick.m:460:5: error: 'TARGET_OS_TV' is not defined, evaluates to 0 [-Werror,-Wundef-prefix=TARGET_OS_]
^
2 errors generated.
src/filesystem/cocoa/SDL_sysfilesystem.m:83:6: error: 'TARGET_OS_TV' is not defined, evaluates to 0 [-Werror,-Wundef-prefix=TARGET_OS_]
^
1 error generated.
Targeting x86_64 against 10.12 SDK:
src/video/SDL_video.c:1492:25: error: 'TARGET_OS_MACCATALYST' is not defined, evaluates to 0 [-Werror,-Wundef-prefix=TARGET_OS_]
^
1 error generated.
|
|
16e3bfe8
|
2021-06-28T11:29:16
|
|
SetDisplayMode: Call XRRSetScreenSize before setting CRTC config
X11_SetDisplayMode currently calls X11_XRRSetCrtcConfig alone. This results
in the monitor's viewport getting changed, but the underlying screen dimensions
stay the same.
The spec indicates that RRSetCrtcConfig only changes the crtc mode and has no effect
on the screen dimensions, only mentioning that the new crtc must fit entirely within the
screen size. For the size to change, RRSetScreenSize also needs to be called.
This affects Metro Exodus on Linux, when changing the resolution in the in-game settings
Metro gets stuck in a loop waiting for the size of its vulkan surface to change. Because
XRRSetScreenSize is not called the screen size is never changed, the vulkan surface dimensions
do not change, and Metro hangs forever watching for a surface size update that will
never come.
This change disables the CRTC, calls XRRSetScreenSize, and then updates the
CRTC configuration. This fixes changing the resolution from the Metro settings.
Tested with:
Metro Exodus, Portal 2
|
|
|
118480e5
|
2021-07-08T17:56:14
|
|
fix permissions
|
|
|
4a20dd3e
|
2021-07-08T07:38:18
|
|
Removed unnecessary comment, the mic button is in byte 16 on all firmware versions
|
|
|
46f19c31
|
2021-07-08T07:23:29
|
|
Implemented mouse relative mode for iOS 14.1 and newer
|
|
3433f3c4
|
2021-07-08T01:18:01
|
|
CMake: Small improvement for MinGW
This is currently a small patch we have in our Conan package for SDL https://github.com/bincrafters/community/blob/4dc894cc2c4df43aabf39c212aa0a31047d183e1/recipes/sdl2/all/conanfile.py#L184
Introduced in https://github.com/bincrafters/conan-sdl2/commit/9fb93a60446914da2a0fafc692cbd175c2d5db50
|
|
|
6f0865fe
|
2021-07-07T23:51:32
|
|
updated os2 makefile.
|
|
|
599001d4
|
2021-07-07T13:30:48
|
|
Revert "Removed reference to ant project."
This reverts commit 0cad302346f95da770e5e1dc818f585c4ab4530d.
The ant directory still exists in source control
|
|
|
75a9f8e2
|
2021-07-07T13:06:15
|
|
Added SDL_hidapi_luna.c to the Xcode project
|
|
|
430bbcb8
|
2021-07-07T09:49:33
|
|
Added HIDAPI support for the Amazon Luna Controller connected over USB in DirectInput mode
|
|
|
863f5669
|
2021-07-07T08:55:16
|
|
Fixed reading thumbstick axes and mic button on the Amazon Luna Controller firmware version 305164320
|
|
17ed8d80
|
2021-07-07T16:05:35
|
|
Added HIDAPI backend for Amazon Luna Controller Model T28B69 connected via Bluetooth LE (VID:0171, PID:0419).
To enter Bluetooth pairing mode hold B and Action (button with circle) buttons for 3 seconds.
It works via usual HIDAPI if special filter driver is not installed:
https://www.amazon.com/gp/help/customer/display.html?nodeId=GZCT4CTFHXLHEB9T
With that driver installed it mimics Xbox One controller and works via XInput under Windows.
Under DInput this controller is not usable at all.
|
|
|
00d67620
|
2021-07-06T11:32:11
|
|
Added the SteelSeries manufacturer to the Nimbus+ entry on macOS
|
|
|
0f00f6dd
|
2021-07-06T10:01:31
|
|
Ignore checkkeysthreads
|
|
|
24e836b0
|
2021-07-06T10:00:09
|
|
Added controller mapping for the Nimbus+ controller on macOS when MFi support is not available
|
|
|
6b21a6cc
|
2021-07-05T20:02:02
|
|
Update config.guess and config.sub from mainstream.
|
|
|
0994a758
|
2021-07-01T20:36:49
|
|
Don't hardcode message length
|
|
|
3b6e9992
|
2021-06-23T14:57:04
|
|
Vita: remove unused variable and allow Razor perf analysis
|
|
|
656eb7df
|
2021-06-23T14:35:36
|
|
Vite: return system installed memory
|
|
|
89015b9c
|
2021-06-22T00:09:06
|
|
Vita: default window size for tests
|
|
|
b55ee12f
|
2021-06-21T23:56:04
|
|
Vita: fix clip rectangle
|
|
|
e41d3e61
|
2021-06-21T23:15:37
|
|
VIta: fix render clearing
|
|
|
6b122805
|
2021-06-21T23:14:56
|
|
Vita: proper syntax
|
|
|
32deb6f7
|
2021-06-21T23:14:40
|
|
Vita: fix point size
|
|
|
9d452719
|
2021-06-21T15:42:47
|
|
Vita: append messagebox title
|
|
|
6460151c
|
2021-06-21T15:03:31
|
|
Vita: add missing stub
|
|
|
39302c92
|
2021-06-02T14:28:33
|
|
Add Steam Virtual Gamepad and Xbox 360 Wired Controller defines instead of magic values
See https://partner.steamgames.com/doc/features/steam_controller/steam_input_gamepad_emulation_bestpractices for details on what is Steam Virtual Gamepad.
|
|
|
0c4ac33a
|
2021-06-28T14:31:27
|
|
Added support for the Razer Wolverine Tournament Edition controller
|
|
|
2143534a
|
2021-06-28T02:00:24
|
|
configure.ac: manually check immintrin.h with AC_COMPILE_IFELSE().
AC_CHECK_HEADER() emits warnings when configuring for non-x86, because
the preprocessor check is OK but the compile check is not:
configure: WARNING: immintrin.h: present but cannot be compiled
configure: WARNING: immintrin.h: check for missing prerequisite headers?
configure: WARNING: immintrin.h: see the Autoconf documentation
configure: WARNING: immintrin.h: section "Present But Cannot Be Compiled"
configure: WARNING: immintrin.h: proceeding with the compiler's result
|
|
ca383599
|
2021-06-27T01:49:31
|
|
Fix Anne Pro II keyboard showing up as a joystick
Having used this for a couple years, I can definitely confirm that it is not a joystick.
|
|
|
c90bd3bf
|
2021-06-27T03:13:47
|
|
Add missing PS5 DualSense GUIDs
Fixes my DualSense controller not being recognised as a SDL GameController which breaks Steam Input filtering causing SDL to fight for input.
|
|
|
bfdea605
|
2021-06-26T05:56:10
|
|
avoid -Wundef warning from SDL_config_windows.h
|
|
|
205b951b
|
2021-06-25T14:20:08
|
|
SDL_DINPUT_JoystickPresent() needs to do the full device enumeration
It is called from WGI before the normal joystick detection has been run, so it needs to actually enumerate currently connected devices.
We can skip the logic checking for other drivers also supporting this device, because that logic is duplicated from the call site.
|