Fullscreen spaces and miniaturization will always give us some notification that they succeeded or failed, so the timeout isn't required. This prevents errant timeouts when live-resize is active.
Since there is no option to change it this seems like a much better default value. The stretching behaviour is really off-putting.
The new behavior shows a small black border when resizing the windows. This makes it more in line with the other backends and it's what you would expect to happen as a user.
I changed it so that the OS does the orientation change itself with a potential performance penalty.
This makes it automatically do the right thing, just like on iOS which would make the orientation change behaviour more consistent across different platforms.
But without adding an option to the user, this would disallow the user solving the problem in his in the app/shaders and saving some performance.
It's up to you to decide what to do. But I changed this in my local copy of the source for my use case.
But this also
* Update SDL_mouse.c
Stop a mouseID of SDL_PEN_MOUSEID being discarded when dispatching mouse events. I'm not sure if this enough to fix the lack of SDL_PEN_MOUSEID being emitted.
* Update SDL_mouse.c
Since we test for touch input here we also test for pen input in the same way.
* Stop duplicate synthetic touch events
If SDL_HINT_PEN_MOUSE_EVENTS and SDL_HINT_MOUSE_TOUCH_EVENTS are both enabled, the pen generated synthetic mouse event will then produce a touch event without this additional check.
This requires the previous commits in order to do anything since it needs to be able to identify those pen generated mouse events.
It looks like both macOS (15.1.1) and SDL are trying to talk to the controller at the same time, which can cause interleaved replies or even locking up the controller. Waiting a bit before talking to the controller seems to take care of this.
This can happen on Windows when the controller is turned off directly. It still shows up in the device list and you can send packets to it, but it's off and doesn't respond. We'll mark this device as broken and open the other as a single Joy-Con.
When the flushing is not able to keep up with the audio stream coming in, it
will end up flushing forever and block API clients from getting any audio.
The example program in #9706 get some audio with SDL 3, while do not get any
audio with SDL 2, which I suspect is because SDL 3 is quicker at flushing the audio.
A fix for the SDL 2 issue is available in #12378.
Additionally, add a warning about calling the frame visibility function universally during the mapping process, as the libdecor Cairo plugin has a bug that will cause a crash in this scenario.
If the pipewire thread invokes output_callback() while we're still
waiting inside PIPEWIRE_OpenDevice(), we will deadlock. The pipewire
thread owns the loop lock and is blocked on the audio device lock,
which cannot be released because pw_thread_loop_wait() needs to
reacquire the loop lock before it can return and allow
PIPEWIRE_OpenDevice() to complete and release the device lock.
isZoomed returns true if the window has the size and position that it would if it were maximized, so we need to check to see if our floating state matches that before saying we're zoomed.
This fixes calling zoom:nil on a borderless resizable window that was created with the same size as the usable desktop area, which happens to also be the maximized state.
Fixes https://github.com/libsdl-org/SDL/issues/12228
This was intended to make the API public, so SDL_hashtable.h got an extreme
documentation makeover, but for now this remains a private header.
This makes several significant interface changes to SDL_HashTable, and
improves code that makes use of it in various ways.
- The ability to make "stackable" tables is removed. Apparently this still
worked with the current implementation, but I could see a future
implementation struggle mightily to support this. It'll be better for
something external to build on top of the table if it needs it, inserting a
linked list of stacked items as the hash values and managing them separately.
There was only one place in SDL using this, unnecessarily, and that has also
been cleaned up to not need it.
- You no longer specify "buckets" when creating a table, but rather an
estimated number of items the table is meant to hold. The bucket count was
crucial to our classic hashtable implementation, but meant less once we
moved to an Open Addressing implementation anyhow, since the bucket count
isn't static (and they aren't really "buckets" anymore either). Now you
can just report how many items you think the hash will hold and SDL will
allocate a reasonable default for you...or 0 to not guess, and SDL will
start small and grow as necessary, which is often the correct thing to do.
- There's no more SDL_IterateHashTableKey because there's no more "stackable"
hash tables.
- SDL_IterateHashTable() now uses a callback, which matches other parts of SDL,
and also lets us hold the read-lock for the entire iteration and get rid of
the goofy iterator state variable.
- SDL_InsertIntoHashTable() now lets you specify whether to replace existing
keys or fail if the key already exists.
- Callbacks now use SDL conventions (userdata as the first param).
- Other naming convention fixes.
I discovered we use a lot of hash tables in SDL3 internally. :) So the bulk
of this work is fixing up that code to use the new interfaces, and simplifying
things (like checking for an item to remove it if it already exists before
inserting a replacement...just do the insert atomically, it'll do all that
for you!).
_GetWinID() doesn't work with keyboard-related BMessages, because Haiku
assumes you know what window has keyboard focus at the time, so these events
don't have a `window-id` property. So when this call failed, the key event
handler would return early.
This was probably a copy/paste error that snuck in at some point, as SDL2
doesn't have this issue.
Use the modifier state supplied with key events to track the system modifier state instead of relying on the state returned by XQueryPointer(), which can be racy when used with automated text entry.
It turns out the mapping we include doesn't work for real controllers, and they're using a generic chipset and generic name and can't be generally distinguished from other controllers.
See https://github.com/libsdl-org/SDL/issues/8644 for details.
This is the output format of stb_image for image decoding, so let's avoid a texture format conversion where possible.
Also standardized SDL_PIXELFORMAT_ARGB8888 as the default texture format for all renderers.
A window may have been maximized by dragging it to the top of another display, in which case the floating position may be out-of-date. If the window is being restored to maximized, and the maximized and floating position are on different displays, try to center the window on the maximized display for restoration, which mimics native Windows behavior.
The Wayland keyboard repeat code assumes that if we have a certain timeout then we'll wait at least that long, and generate a key repeat event on timeout. If we wait a shorter time, we won't generate a key repeat event and then return 0, even if we were supposed to wait indefinitely.
Fixes https://github.com/libsdl-org/SDL/issues/12239
The LED hint was getting registered for SDL_HINT_JOYSTICK_ENHANCED_REPORTS
instead of SDL_HINT_JOYSTICK_HIDAPI_PS5_PLAYER_LED, which results in a
use-after-free followed by a crash.
This PR removes the incorrect implementation of `SDL_cond` currently included with the 3DS port.
Pseudocode of the incorrect implementation of `SDL_CondWait` this PR removes:
* Receive an `SDL_cond` backed by a `libctru` `CondVar` and an `SDL_mutex` backed by a `libctru` `RecursiveLock`.
* Want to call `libctru` function `CondVar_Wait` which expects a `CondVar` and a `LightLock` (non-recursive lock)
* Do so by calling this function with the internal (inadequately protected) `LightLock` member of the `RecursiveLock` (`&mutex->lock.lock` on line 105), without updating any internal thread or lock count fields of the `RecursiveLock`.
Happy to discuss or test some examples. My own use case works much better with the generic cond logic, and this seems like a safe fix to me given that the generic logic is well-tested and this seems not to be.
If you like the PR I'll send another one for the SDL2 branch.
Frank Praznik
Sam Lantinga
Ryan C. Gordon
Brick
Anonymous Maarten
Ozkan Sezer
guitarfreak
Elbagast
Wim Taymans
Alexander Batalov
jsoulier
hwsmm
cosmonaut
Tomasz Pakuła
Ethan Lee
zopsicle
Caleb Cornett
Ralph Sennhauser
Petter Reinholdtsen
Petar Popovic
Cameron Gutman
williamhCode
expikr
Paper
Sam Clegg
nightmareci
Mike Kosek
ImThour
captain0xff
Chen Steenvoorden
Semphris
coffeechriph
Edu Garcia
danginsburg
Neal Gompa
Stenzek
Clint Kilmer
Guldoman
ds-sloth