XRInputSource: targetRaySpace property

Limited availability

This feature is not Baseline because it does not work in some of the most widely-used browsers.

Secure context: This feature is available only in secure contexts (HTTPS), in some or all supporting browsers.

The read-only XRInputSource property targetRaySpace returns an XRSpace (typically an XRReferenceSpace) representing the position and orientation of the target ray in the virtual space. Its native origin tracks the position of the origin point of the target ray, and its orientation indicates the orientation of the controller device itself. These values, interpreted in the context of the input source's targetRayMode, can be used both to fully interpret the device as an input source.

To obtain an XRSpace representing the input controller's position and orientation in virtual space, use the gripSpace property.

Value

An XRSpace object—typically an XRReferenceSpace or XRBoundedReferenceSpace—which represents the position and orientation of the input controller's target ray in virtual space.

The native origin of the returned XRSpace is located at the point from which the target ray is emitted, and the orientation of the space indicates the direction in which the target ray is pointing.

Usage notes

All input sources—regardless of their targetRayMode—have a valid targetRaySpace. The exact meaning of this space varies, however, depending on the mode:

Every gaze input (targetRayMode value of gaze), shares the same XRSpace object as their target ray space, since the gaze input comes from the viewer's head. This shared space represents the same location as the space returned by the XRSession method requestReferenceSpace(), but is maintained as a different object to allow for future enhancements to the API.
The target ray space reported by tracked pointer inputs (targetRayMode of tracked-pointer) is actually based upon the true spatial position and orientation of the input device.

To determine the position and orientation of the target ray while rendering a frame, pass it into the XRFrame method getPose() method, then use the returned XRPose object's transform to gather the spatial information you need.

Examples

This fragment of code shows part of a function to be called once every frame. It looks for inputs which have a non-null targetRaySpace. Inputs which have a value for this property represent inputs that project a target ray outward from the user.

For each such input, this example looks for inputs whose targetRayMode is tracked-pointer, indicating that the input is in fact intended to represent a targeting device rather than a gazing device, screen tap, or mouse click. For tracked pointers, a function myRenderTargetRayAsBeam() is called to render a beam from the input controller's virtual position outward in the direction it's pointing.

The code should continue to perform tasks such as drawing controllers or any objects representative of the user's hands' positions in the virtual space, as well as any other input-related tasks.

js
function updateInputSources(session, frame, refSpace) {
  for (const source of session.getInputSources()) {
    const targetRayPose = frame.getPose(inputSource.targetRaySpace, refSpace);

    if (targetRayPose) {
      if (source.targetRayMode === "tracked-pointer") {
        myRenderTargetRayAsBeam(targetRayPose);
      }
    }

    // …
  }
}