Proposed Solution for Screen and Window Recording

[CoolSlimbo]

What is the intended way that screen recording will/would want to be implemented into the library?

Currently thinking of the following would work: (plans to self-implement once implementation details for such are done)

use std::time::Instant;
use xcap::Monitor;

fn normalized(filename: &str) -> String {
    filename
        .replace("|", "")
        .replace("\\", "")
        .replace(":", "")
        .replace("/", "")
}

fn main() {
    let start = Instant::now();
    let monitors = Monitor::all().unwrap();

    for monitor in monitors {
        let record = monitor.record_screen().unwrap();
        record.set_fps(60);
        record.on_frame(|image| {
            image
                .save(format!("target/monitor-{}.png", normalized(monitor.name()))) // No clue on if monitor would be able to passed, it's theory not practical directly this way
                .unwrap();
                
            // Depending on how it would be implemented, maybe pass an object that can signal when to stop recording, among other things?
        });
        std::thread::sleep(std::Duration::from(from_seconds(30));
        record.stop();
    }

    println!("运行耗时: {:?}", start.elapsed());
}

This is all theoretical, the practical implementation would likely be different.