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.
This reverts commit 0825d07a43.
It turns out that resizing while hidden is fine, the real problem in https://github.com/libsdl-org/sdl2-compat/issues/268 is that SDL2 did not send an initial resize event and SDL3 does, which we're fixing in a better way in sdl2-compat.
The device string indicates RED on the one I have and is PIDs 0003 and some other posts online say 0001 for the normal model, so I'll try 0001-0003 to get all 3 color variants
We'll switch to the global mouse ID just once we are ready to deliver events.
This makes sure that any button events that come in for a specific mouse ID maintain that state if we switch to relative mode and start using that mouse ID for events.
Fixes https://github.com/libsdl-org/sdl2-compat/issues/263
When the mouse is grabbed, the X server sends mouse events only to the grabbing client, and XInput2 events for the master device are not delivered.
We should consider using the window mouse rect confinement instead of a true X server grab for SDL mouse grab functionality.
src/gpu/vulkan/SDL_gpu_vulkan.c:763: error: redefinition of typedef 'VulkanUniformBuffer'
src/gpu/vulkan/SDL_gpu_vulkan.c:482: note: previous declaration of 'VulkanUniformBuffer' was here
The comment in the source wasn't true; PipeWire doesn't _have_ to work in
float format. It presumably does if it has to mix, but if a game is the only
thing making noise on the system--a common scenario--then it might be able to
pass, say, Sint16 data straight through to the hardware without conversion.
Fixes#12129.
You can end up with a NULL scratch buffer, which is otherwise not needed on
this path, then ConvertAudio will end up needing that scratch space to move
to float32 to apply gain.
Fixes#12091.
(I assume.)
Audio streams used to accept audio with a src or dest frequency between
4000Hz and 384000Hz. It was arbitrary (or perhaps a relic of older
resampler revisions), and testing shows unnecessary, so remove it.
Fixes#12098.
An undefined position means that the window manager can handle placement, so SDL shouldn't override that by forcing a position when showing a window.
Allows for removing a fair bit of now-unnecessary code as well.
It used to check for "software" in the hint, but now it has to parse a
comma-separated list.
If it tries to use the software renderer here, you'll end up in an infinite
recursion.
Provious code wrongly assumed that direction is not an important part
of conditional effect. Moreover, if there's need to hardcode polar
direction, the default should be 0x4000 (north).
For one axis affects, a direction of 0 means complete lack of force, if
a FFB-enabled device takes direction into force calculation. A sine function
graph can be used to represent the resulting forces where X is the input
direction and Y is the force multiplier (360 degrees equals to 1).
This fixes conditional effect playback on Moza Racing devices, which do
not ignore direction field.
This doesn't affect latency much, but it makes the system usable if the system
drops you down from the bluetooth a2dp profile (headphones) to the handsfree
(I think...?) profile because the bluetooth audio device is also recording,
which would be extremely common in a VoIP app, but also if you're talking
in a different app while also playing audio.
Fixes#8192.
This needs to be preserved while in fullscreen, or leaving fullscreen for the maximized state can cause the taskbar to disappear with borderless windows.
Always restore the base floating size of a window before possibly entering the maximized state, as base size can be lost during the fullscreen transition, resulting in the window de-maximizing to the wrong size.
When going into a fullscreen space, update mouseConfinementRect once the
transition is done. When going exclusive fullscreen, force a window sync so
the transition is complete before we do the update.
Fixes#9088.
The API states that the related functions must return NULL if the function
called (get the parent tray, or get the parent entry) is invalid for this
menu. Initialising the fields to NULL makes that API correct for Windows.
The test/testtray program would crash on Windows when adding any item and then removing it, because a submenu's parent_entry field was not set.
Additionally, I noticed that some extraneous code copied from the {G,S}etTrayEntryChecked made {G,S}etTrayEntryEnabled work only for checkboxes, which is not the desired behavior.
Both issues were fixed in this commit.
Update the mouse state before entering the hit test, in case the global state is queried, or the system menu opened, while in the client hit testing callback.
MonitorFromWindow can fail if called on a window being de-minimized, so fall back to using the monitor from the last window coordinates if initial retrieval fails.
We can't directly set the mouse focus since we may get spammed by entered/exited events,
but we can process the current focus later in the mouseMoved handler in line with the
mouse motion event sequence.
Fixes https://github.com/libsdl-org/SDL/issues/8188
Returning anything other an HTCLIENT result will cause windows to eat the button press, so ensure that non-left presses are passed through to the client over draggable areas.
The existing algorithm works well for min-max ratios, but didn't allow edge expansion of fixed aspect ratio windows. Use NSWindow setContentAspectRatio instead.
This can happen if all the windows shown have the SDL_WINDOW_NOT_FOCUSABLE flag. We'll still accept modifier state changes though, so you can do Control-click actions.
Corrected the calculation of the vertical hotspot position in the `GetCachedCursor` function.
Changed the variable from `data->hot_x` to `data->hot_y` to ensure the correct vertical position of the cursor's hotspot is used when scaling.
We require at least Xcode 12.2 and macOS SDK 11 to build. We support deploying to macOS 10.13, iOS 11.0, and tvOS 11.0.
This cleans up the code significantly
SDL_CreateTray now respects SDL_HINT_WINDOWS_INTRESOURCE_ICON_SMALL
and SDL_HINT_WINDOWS_INTRESOURCE_ICON hints and uses the specified icon
as the tray icon.
Adds SDL_HINT_VIDEO_MAC_FULLSCREEN_MENU_VISIBILITY to control whether or not the menu can be accessed when the cursor is moved to the top of the screen when a window is in fullscreen spaces mode.
The three values are true, false, and 'auto' (default), with auto resulting in a hidden menu if fullscreen was toggled programmatically, and the menu being accessible if fullscreen was toggled via the button on the window title bar, so the user has an easy way back out of fullscreen if the client app/game doesn't have a readily available option to toggle it.
This was originally intended to make sure that nativeAllowRecreateActivity() could be called from another thread safely, but the hint system is now thread-safe, so we don't need to use a callback here.
Fixes https://github.com/libsdl-org/SDL/issues/11938
This is just enough to delay MOUSE_LEAVE events until a drag ending outside
the canvas is released, and make sure the MOUSE_UP event has the canvas's
SDL_WindowID, which it wouldn't when mouse focus is was lost by this point.
Fixes#10033.
Any requested swap interval will be saved and set on the first PumpEvents.
Once PumpEvents has been called at least once, swap intervals are set
immediately.
This assumes that events won't be pumped until after an Emscripten main loop
has been defined, and so prevents a warning on the javascript console:
"emscripten_set_main_loop_timing: Cannot set timing mode for main loop since
a main loop does not exist! Call emscripten_set_main_loop first to set one
up."
Fixes#9969.
Add SDL keycodes for keys found commonly found in the default Xkb layout, such as left tab and compose, and keys frequently used for custom modifiers such as Meta, Hyper, and Level5 Shift.
As these keys aren't Unicode code points and don't have associated scancodes (at least on modern keyboards), they are placed in the new extended key code space, with bit 30 set as a flag.
Adds support for Mod3, which is usually Level 5 shift, as well as not altering the functionality of the more esoteric modifier keys, such as meta and hyper.
Also use the system modifier state instead of setting them based on key presses, which may be incorrect due to remapping, or toggled in some other manner.
Adds support for Mod3, which is usually Level 5 shift, but can vary, as well as not altering the functionality of the more esoteric modifier keys, such as meta and hyper.
Sam Lantinga
Petar Popovic
Ryan C. Gordon
Cameron Gutman
Frank Praznik
Chen Steenvoorden
ImThour
Semphris
coffeechriph
Edu Garcia
Mike Kosek
danginsburg
Neal Gompa
Stenzek
Clint Kilmer
Guldoman
ds-sloth
Evan Hemsley
Jean-Philip Desjardins
WinterSquire
Ozkan Sezer
Em
John Kvalevog
Timothee Besset
John Alanbrook
Aki
Seth Anderson
Adam Kewley
Carl Åstholm
nightmareci
hspir404
Katelyn Gadd
Tomasz Pakuła
William Hou
ROllerozxa
Ethan Lee
Ramez Ragaa
cosmonaut
Green Sky
Scrooge86x
expikr
Xander
Anonymous Maarten