XRSession: selectstart event

Use the event name in methods like addEventListener(), or set an event handler property.

addEventListener("selectstart", (event) => {});

onselectstart = (event) => {};

An XRInputSourceEvent. Inherits from Event.

In addition to the properties listed below, properties from the parent interface, Event, are available.

frame Read only

An XRFrame object providing the needed information about the event frame during which the event occurred. This frame may have been rendered in the past rather than being a current frame. Because this is an event frame, not an animation frame, you cannot call XRFrame.getViewerPose() on it; instead, use getPose().

inputSource Read only

An XRInputSource object indicating which input source generated the input event.

Triggered when the user presses triggers or buttons, taps a touchpad, speaks a command, or performs a recognizable gesture when using a video tracking system or handheld controller with an accelerometer.

The selectstart and selectend events tell you when you might want to display something to the user indicating that the primary action is going on. This might be drawing a controller with the activated button in a new color, or showing the targeted object being grabbed and moved around, starting when selectstart arrives and stopping when selectend is received.

The select event tells your code that the user has completed the action they want to complete. This might be as simple as throwing an object or pulling the trigger of a gun in a game, or as involved as placing a dragged object in a new location.

If your primary action is a simple trigger action and you don't need to animate anything while the trigger is engaged, you can ignore the selectstart and selectend events and act on the start event.

The following example uses addEventListener() to establish handlers for the selection events: selectstart, selectend, and select. This snippet is the core of an event handler to allow the user to grab objects in the scene and move them around.

In this case, a single function is used to handle all three events, allowing them to share certain code that's the same regardless of which of the three events is received. Only after completing those tasks does the onSelectionEvent() function below dispatch the action out to a specialized function to handle things.

After checking to ensure that the received event is a tracked-pointer event (the only kind we handle here), the target ray's pose is obtained using getPose().

If the target ray pose was fetched successfully, the code then uses the value of Event property type to route control to an appropriate function to handle the event which arrived:

• For selectstart events, a myBeginTracking() function is called with the target ray pose's matrix. The myBeginTracking() function would presumably start the presentation of the object-dragging process, using the transform to perform a hit test, determining which object to pick up. myBeginTracking() returns an object representing the object the user has begun to drag.
• Upon receiving a select event, the myDropObject() function is called with the target object, and the current target ray pose transform as inputs. This places the object into its new position in the world and triggers any effects that may arise, like scheduling an animation of a splash if dropped in water, etc.
• The selectend event results in a myStopTracking() function being called with the object being dragged and the final target ray pose's transform.

xrSession.addEventListener("selectstart", onSelectionEvent);
xrSession.addEventListener("select", onSelectionEvent);
xrSession.addEventListener("selectend", onSelectionEvent);

function onSelectionEvent(event) {
  let source = event.inputSource;
  let targetObj = null;

  if (source.targetRayMode !== "tracked-pointer") {
    return;
  }

  let targetRayPose = event.frame.getPose(source.targetRaySpace, myRefSpace);
  if (!targetRayPose) {
    return;
  }

  switch (event.type) {
    case "selectstart":
      targetObj = myBeginTracking(targetRayPose.matrix);
      break;
    case "select":
      myDropObject(targetObj, targetRayPose.matrix);
      break;
    case "selectend":
      myStopTracking(targetObj, targetRayPose.matrix);
      break;
  }
}

You can also set up a handler for selectend events by setting the XRSession object's onselectend event handler property to a function that handles the event:

xrSession.onselectstart = onSelectionEvent;
xrSession.onselect = onSelectionEvent;
xrSession.onselectend = onSelectionEvent;

• select and selectend
• beforexrselect