Similar to #254. Another unclear point is in the unlock() aborts a pending lock request across documents test in unlock.html. The test expects the lock to be allowed when the parent document is in fullscreen. I am not sure if this is a test bug, or an intentional behavior.
If this is intentional behavior, it would be good if we could define this more clearly in the spec. And perhaps we should only allow it when the parent document is same-origin; otherwise, the fullscreen state could be leaked to the cross-origin iframe? Also we would also need to define how to handle the lock when the parent document exits fullscreen.
Similar to #254. Another unclear point is in the
unlock() aborts a pending lock request across documentstest in unlock.html. The test expects the lock to be allowed when the parent document is in fullscreen. I am not sure if this is a test bug, or an intentional behavior.If this is intentional behavior, it would be good if we could define this more clearly in the spec. And perhaps we should only allow it when the parent document is same-origin; otherwise, the fullscreen state could be leaked to the cross-origin iframe? Also we would also need to define how to handle the lock when the parent document exits fullscreen.