`watcher.SetUseCodedPhy(true)` breaks scanning on Windows

[Finomnis]

maybe related: #472

I just upgraded from btleplug 0.12.0 to 0.13.0.

Now, scanning no longer yields any events on my Windows System.

use btleplug::api::{Central, Manager as _, ScanFilter};
use btleplug::platform::Manager;
use futures_util::StreamExt as _;
use std::error::Error;
use std::time::Duration;

#[tokio::main]
async fn main() -> Result<(), Box<dyn Error>> {
    let manager = Manager::new().await.unwrap();

    // get the first bluetooth adapter
    let adapters = manager.adapters().await?;
    let central = adapters.into_iter().nth(0).unwrap();

    let mut events = central.events().await?;

    central.start_scan(ScanFilter::default()).await?;
    let _ = tokio::time::timeout(Duration::from_secs(2), async {
        loop {
            println!("{:?}", events.next().await.unwrap());
        }
    })
    .await;
    let _ = central.stop_scan().await;

    Ok(())
}

Expected behavior
A lot of printed lines containing discovery events of the BLE devices around me.

Actual behavior
Absolutely nothing.

Additional context
When I comment out the line let _ = self.watcher.SetUseCodedPhy(true); in src\winrtble\ble\watcher.rs:84, it works as expected.

System informations

I'm on Windows 11.
My BLE adapter is a tp-link UB400.

[userepo]

@Finomnis, thanks for reporting this issue.

Could you test the fix in #474 on your UB400? Point your test program at the branch:

btleplug = { git = "https://github.com/Poholos/btleplug", branch = "fix/gate-coded-phy-on-adapter-support" }

Scan events coming back would confirm it on the failing side; I only have a Coded-capable adapter here.

[Finomnis]

I can confirm that your [patch.crates-io] section fixes my problem, while it is still broken on

btleplug = { git = "https://github.com/deviceplug/btleplug", branch = "master" }

(just to confirm that nothing else between 0.13.0 and master already fixed it)

[Finomnis]

Do you plan to release this as a 0.13.1 hot fix within the next days by any chance?

[userepo]

Do you plan to release this as a 0.13.1 hot fix within the next days by any chance?

@qdot and/or @qwandor might be able to answer that.

[qdot]

0.13.1 will probably land today. I was about to hit release on my software w/ this dependency, of which 70% of the userbase is windows on who knows what kind of adapter, so I'd rather be safe than sorry. :)

[Finomnis]

0.13.1 will probably land today.

You might want to yank 0.13.0 afterwards :)

[qdot]

@Finomnis 0.13.1 is out, 0.13.0 yanked.