Inner panics could make it hard to trouble shoot the issues and for some
users it's not desirable.
The inner panics were left only when they are used to `assert!` during
development.
This reverts commit 9f91bc413fe20618bd7090829832bb074aab15c3 which
reverted the original patch which was merged without a proper review.
Fixes: #500.
Inner panics could make it hard to trouble shoot the issues and for some
users ints not desirable.
The inner panics were left only when they are used to `assert!` during
development.
Mark it as breaking, since some clients relied on that behavior, simply
because dispatching clients queue always woke up a winit, meaning that
they won't be able to use user events for this sake.
Lifetimes don't work nicely when dealing with multithreaded environments
in the current design of the existing winit's event handling model, so
remove it in favor of `InnerSizeWriter` fences passed to client, so they
could try to update the size.
Fixes#1387.
The idea that redraw events are dispatched with a specific ordering
that makes it possible to specifically report when we have finished
dispatching redraw events isn't portable and the way in which we
dispatched RedrawEventsCleared was inconsistent across backends.
More generally speaking, there is no inherent relationship between
redrawing and event loop iterations. An event loop may wake up at any
frequency depending on what sources of input events are being listened
to but redrawing is generally throttled and in some way synchronized
with the display frequency.
Similarly there's no inherent relationship between a single event loop
iteration and the dispatching of any specific kind of "main" event.
An event loop wakes up when there are events to read (e.g. input
events or responses from a display server / compositor) and goes back
to waiting when there's nothing else to read.
There isn't really a special kind of "main" event that is dispatched
in order with respect to other events.
What we can do more portably is emit an event when the event loop
is about to block and wait for new events.
In practice this is very similar to how MainEventsCleared was
implemented except it wasn't the very last event previously since
redraw events could be dispatched afterwards.
The main backend where we don't strictly know when we're going to
wait for events is Web (since the real event loop is internal to
the browser). For now we emulate AboutToWait on Web similar to how
MainEventsCleared was dispatched.
In practice most applications almost certainly shouldn't care about
AboutToWait because the frequency of event loop iterations is
essentially arbitrary and usually irrelevant.
Considering the possibility of re-running an event loop via run_ondemand
then it's more correct to say that the loop is about to exit without
assuming it's going to be destroyed.
This renames all internal implementations of pump_events_with_timeout
to pump_events and makes them public.
Since all platforms that support pump_events support timeouts there's
no need to have a separate API.
Although we document that applications can't keep windows between
separate run_ondemand calls it's possible that the application has only
just dropped their windows and we need to flush these requests to the
server/compositor.
This fixes the window_ondemand example - by ensuring the window from
the first loop really is destroyed before waiting for 5 seconds
and starting the second loop.
This re-works the portable `run()` API that consumes the `EventLoop` and
runs the loop on the calling thread until the app exits.
This can be supported across _all_ platforms and compared to the
previous `run() -> !` API is now able to return a `Result` status on all
platforms except iOS and Web. Fixes: #2709
By moving away from `run() -> !` we stop calling `std::process::exit()`
internally as a means to kill the process without returning which means
it's possible to return an exit status and applications can return from
their `main()` function normally.
This also fixes Android support where an Activity runs in a thread but
we can't assume to have full ownership of the process (other services
could be running in separate threads).
Additionally all examples have generally been updated so that `main()`
returns a `Result` from `run()`
Fixes: #2709
Wayland:
I found the calloop abstraction a little awkward to work with while I was
trying to understand why there was surprising workaround code in the wayland
backend for manually dispatching pending events.
Investigating this further it looks like there may currently be several issues
with the calloop WaylandSource (with how prepare_read is used and with (not)
flushing writes before polling)
Considering the current minimal needs for polling in all winit backends I do
personally tend to think it would be simpler to just own the responsibility for
polling more directly, so the logic for wayland-client `prepare_read` wouldn't
be in a separate crate (and in this current situation would also be easier to fix)
I've tried to maintain the status quo with calloop + workarounds.
X11:
I found that the recent changes (4ac2006cbc) to port the X11 backend
from mio to calloop lost the ability to check for pending events before
needing to poll/dispatch. (The `has_pending` state being queried
before dispatching() was based on state that was filled in during
dispatching)
As part of the rebase this re-introduces the PeekableReceiver and
WakeSender which are small utilities on top of
`std::sync::mpsc::channel()`. This adds a calloop `PingSource`
so we can use a `Ping` as a generic event loop waker.
For taking into account false positive wake ups the X11 source now
tracks when the file descriptor is readable so after we poll via
calloop we can then specifically check if there are new X11 events
or pending redraw/user events when deciding whether to skip the
event loop iteration.
This update rewrites the winit's Wayland backend using new wayland-rs
0.30 API. This fixes long standing issue with the forward compatibility
of the wayland backend, meaning that future updates to the wayland
protocol won't break rust code anymore. like it was before when adding
new shm/enum variants into the protocol.
Fixes#2560.
Fixes#2164.
Fixes#2128.
Fixes#1760.
Fixes#725.
On Waylnad when asking for redraw before `MainEventsCleared`
would result for redraw being send on the next event loop tick,
which is not expectable given that it must be delivered on the same
event loop tick.
To be more consistent with mobile platforms this updates the Windows,
macOS, Wayland, X11 and Web backends to all emit a Resumed event
immediately after the initial `NewEvents(StartCause::Init)` event.
The documentation for Suspended and Resumed has also been updated
to provide general recommendations for how to handle Suspended and
Resumed events in portable applications as well as providing
Android and iOS specific details.
This consistency makes it possible to write applications that lazily
initialize their graphics state when the application resumes without
any platform-specific knowledge. Previously, applications that wanted
to run on Android and other systems would have to maintain two,
mutually-exclusive, initialization paths.
Note: This patch does nothing to guarantee that Suspended events will
be delivered. It's still reasonable to say that most OSs without a
formal lifecycle for applications will simply never "suspend" your
application. There are currently no known portability issues caused
by not delivering `Suspended` events consistently and technically
it's not possible to guarantee the delivery of `Suspended` events if
the OS doesn't define an application lifecycle. (app can always be
terminated without any kind of clean up notification on most
non-mobile OSs)
Fixes#2185.
Co-authored-by: Marijn Suijten <marijns95@gmail.com>
Co-authored-by: Markus Røyset <maroider@protonmail.com>
This reverts commit 78e5a395da.
It was discovered that in some cases mesa will lock the back
buffer, e.g. when making context current, leading to resize
missing. Given that applications can restructure their rendering
to account for that, and that winit isn't limited to playing
nice with mesa reverting the original commit.
This commit brings new Ime event to account for preedit state of input
method, also adding `Window::set_ime_allowed` to toggle IME input on
the particular window.
This commit implements API as designed in #1497 for desktop platforms.
Co-authored-by: Artur Kovacs <kovacs.artur.barnabas@gmail.com>
Co-authored-by: Markus Siglreithmaier <m.siglreith@gmail.com>
Co-authored-by: Murarth <murarth@gmail.com>
Co-authored-by: Yusuke Kominami <yukke.konan@gmail.com>
Co-authored-by: moko256 <koutaro.mo@gmail.com>
If you try to use `EventLoop::run_return` API in a way that you do on
demand polling of events it won't actually poll, since in such strategy
the `ControlFlow::Exit` is sent right before Wayland backend starts to
poll.
This was observed with smithay compositor Anvil which was doing this
particular thing leading to GNOME thinking that app isn't responding,
due to connection not being polled.
* Add exit code to control flow and impl on linux
* Fix examples to have an exit code
* Fix doc examples to use an exit code
* Improve documentation wording on the exit code
* Add exit code example
* Add exit code on windows
* Change i32 as exit code to u8
This avoids nasty surprises with negative numbers on some unix-alikes
due to two's complement.
* Fix android usages of ControlFlow::Exit
* Fix ios usages of ControlFlow::Exit
* Fix web usages of ControlFlow::Exit
* Add macos exit code
* Add changelog note
* Document exit code on display server disconnection
* Revert "Change i32 as exit code to u8"
This reverts commit f88fba0253b45de6a2ac0c3cbcf01f50503c9396.
* Change Exit to ExitWithCode and make an Exit const
* Revert "Add exit code example"
This reverts commit fbd3d03de9c2d7516c7a63da489c99f498b710df.
* Revert "Fix doc examples to use an exit code"
This reverts commit daabcdf9ef9e16acad715c094ae442529e39fcbc.
* Revert "Fix examples to have an exit code"
This reverts commit 0df486896b8d106acf65ba83c45cc88d60d228e1.
* Fix unix-alike to use ExitWithCode instead of Exit
* Fix windows to use ExitWithCode rather than Exit
* Silence warning about non-uppercase Exit const
* Refactor exit code handling
* Fix macos Exit usage and recover original semantic
* Fix ios to use ExitWithCode instead of Exit
* Update documentation to reflect ExitWithCode
* Fix web to use ExitWithCode when needed, not Exit
* Fix android to use ExitWithCode, not Exit
* Apply documenation nits
* Apply even more documentation nits
* Move change in CHANGELOG.md under "Unreleased"
* Try to use OS error code as exit code on wayland
This commit also drops 'Theme' trait with its support types
in favor of 'FallbackFrame' meaning that winit will use some
predefined frame for the time being, since porting 'ConceptFrame'
will require adding font rendering librarires right into winit,
which is not desired.
Fixes#1889.
* Update SCTK to 0.11.0
Updates smithay-client-toolkit to 0.11.0. The major highlight
of that updated, is update of wayland-rs to 0.27.0. Switching
to wayland-cursor, instead of using libwayland-cursor. It
also fixes the following bugs:
- Disabled repeat rate not being handled.
- Decoration buttons not working after tty switch.
- Scaling not being applied on output reenable.
- Crash when `XCURSOR_SIZE` is `0`.
- Pointer getting created in some cases without pointer capability.
- On kwin, fix space between window and decorations on startup.
- Incorrect size event when entering fullscreen when using
client side decorations.
- Client side decorations not being hided properly in fullscreen.
- Size tracking between fullscreen/tiled state changes.
- Repeat rate triggering multiple times from slow callback handler.
- Resizable attribute not being applied properly on startup.
- Not working IME
Besides those fixes it also adds a bunch of missing virtual key codes,
implements proper cursor grabbing, adds right click on decorations
to open application menu, disabled maximize button for non-resizeable
window, and fall back for cursor icon to similar ones, if the requested
is missing.
It also adds new methods to a `Theme` trait, such as:
- `title_font(&self) -> Option<(String, f32)>` - The font for a title.
- `title_color(&self, window_active: bool) -> [u8; 4]` - The color of
the text in the title.
Fixes#1680.
Fixes#1678.
Fixes#1676.
Fixes#1646.
Fixes#1614.
Fixes#1601.
Fixes#1533.
Fixes#1509.
Fixes#952.
Fixes#947.
