webrtc-java

Appearance

These docs describe the unreleased development version 0.21.0-SNAPSHOT. See the docs of the latest release, 0.20.0.

Screen Capturer ​

The ScreenCapturer enumerates and captures full desktop screens/monitors. Use it to list available screens and to drive screen-source selection for desktop capture workflows.

API: dev.onvoid.webrtc.media.video.desktop.ScreenCapturer

Overview ​

  • Purpose: Discover and capture full screens (monitors).
  • Pairs with: VideoDesktopSource to produce a capturable video track from a selected screen ID.
  • Common ops: list screens, select one, configure capture parameters on DesktopCapturer, start capture via callback, or pass the ID to VideoDesktopSource.
  • Resource management: Call dispose() when finished to free native resources.

Key methods (inherited from DesktopCapturer):

  • List<DesktopSource> getDesktopSources() – enumerate available screens
  • void selectSource(DesktopSource source) – choose the screen to capture
  • void setFocusSelectedSource(boolean focus) – try to focus the selected source during capture
  • void setMaxFrameRate(int maxFrameRate) – cap the capture FPS
  • void start(DesktopCaptureCallback callback) – begin capturing; frames are delivered via callback
  • void captureFrame() – request a single frame (manual capture)
  • void dispose() – release resources

Typical usage: enumerate screens ​

java
import dev.onvoid.webrtc.media.video.desktop.DesktopSource;
import dev.onvoid.webrtc.media.video.desktop.ScreenCapturer;
import java.util.List;

ScreenCapturer screenCapturer = new ScreenCapturer();
try {
    List<DesktopSource> screens = screenCapturer.getDesktopSources();
    for (DesktopSource s : screens) {
        System.out.printf("Screen: %s (ID: %d)%n", s.title, s.id);
    }
}
finally {
    screenCapturer.dispose();
}

Selecting a screen and starting capture with a callback ​

java
import dev.onvoid.webrtc.media.video.desktop.*;
import dev.onvoid.webrtc.media.video.VideoFrame;

ScreenCapturer capturer = new ScreenCapturer();
try {
    // pick the first screen
    DesktopSource screen = capturer.getDesktopSources().stream().findFirst()
        .orElseThrow(() -> new IllegalStateException("No screens found"));

    capturer.selectSource(screen);
    capturer.setMaxFrameRate(30);
    capturer.setFocusSelectedSource(false);

    DesktopCaptureCallback callback = (result, frame) -> {
        if (result == DesktopCapturer.Result.SUCCESS && frame != null) {
            // process VideoFrame
        }
    };

    capturer.start(callback);

    // Optionally trigger ad-hoc capture
    capturer.captureFrame();
}
finally {
    capturer.dispose();
}

Using with VideoDesktopSource to create a VideoTrack ​

For most WebRTC pipelines, you will create a VideoDesktopSource and set the selected screen ID:

java
import dev.onvoid.webrtc.media.video.VideoDesktopSource;
import dev.onvoid.webrtc.media.video.desktop.*;

ScreenCapturer sc = new ScreenCapturer();
DesktopSource screen = sc.getDesktopSources().get(0);

VideoDesktopSource vds = new VideoDesktopSource();
vds.setFrameRate(30);
vds.setMaxFrameSize(1920, 1080);

// Select the screen (isWindow = false)
vds.setSourceId(screen.id, false);

// Start the source and use it to create a VideoTrack in your PeerConnection
vds.start();

// ...

// cleanup
vds.stop();
vds.dispose();
sc.dispose();

Integration tips ​

  • If screens can change (hot‑plug monitors), refresh getDesktopSources() as needed.
  • Use setMaxFrameRate to limit capture load on the system.
  • Prefer using VideoDesktopSource for WebRTC pipelines; keep ScreenCapturer for discovery and advanced control.