Is ControlFlow supposed to be sticky? (eventloop-2.0)

[aloucks]

Based on the doc comment, I would believe that it is intended to be sticky. However, the x11 and wayland implementations are resetting it per loop iteration. It doesn't look like Windows is doing this.

winit/src/event_loop.rs

Lines 65 to 69 in 0df4369

	/// ## Persistency
	/// Almost every change is persistent between multiple calls to the event loop closure within a
	/// given run loop. The only exception to this is `Exit` which, once set, cannot be unset. Changes
	/// are **not** persistent between multiple calls to `run_return` - issuing a new call will reset
	/// the control flow to `Poll`.

winit/src/platform_impl/linux/x11/mod.rs

Lines 283 to 286 in 0df4369

	ControlFlow::Poll => {
	// non-blocking dispatch
	self.inner_loop.dispatch(Some(::std::time::Duration::from_millis(0)), &mut ()).unwrap();
	control_flow = ControlFlow::default();

winit/src/platform_impl/linux/x11/mod.rs

Lines 289 to 291 in 0df4369

	ControlFlow::Wait => {
	self.inner_loop.dispatch(None, &mut ()).unwrap();
	control_flow = ControlFlow::default();

winit/src/platform_impl/linux/x11/mod.rs

Lines 301 to 306 in 0df4369

	ControlFlow::WaitUntil(deadline) => {
	let start = ::std::time::Instant::now();
	// compute the blocking duration
	let duration = deadline.duration_since(::std::cmp::max(deadline, start));
	self.inner_loop.dispatch(Some(duration), &mut ()).unwrap();
	control_flow = ControlFlow::default();

[Osspial]

It is indeed intended to be sticky.

[elinorbgr]

Ah, ok. I assumed it was only supposed to be sticky during a run of the event loop, rather than forever.

[aloucks]

I assumed it was only supposed to be sticky during a run of the event loop, rather than forever.

Yes, that's how I understand it; Control flow is sticky for the duration of run_return and reset in between calls to that. The current implementation resets the control flow on each tick of the loop.

This part is correct:

winit/src/platform_impl/linux/x11/mod.rs

Lines 226 to 229 in 0df4369

	pub fn run_return<F>(&mut self, mut callback: F)
	where F: FnMut(Event<T>, &RootELW<T>, &mut ControlFlow)
	{
	let mut control_flow = ControlFlow::default();

[elinorbgr]

Well, the fix is simple enough, I just don't have the time to work on it for now.