Continuous mapping theorem

If $X_n\stackrel{d}{\to }X$ and a measurable function $g$ is continuous at $X$ almost surely, then $g(X_n)\stackrel{d}{\to }g(X)$. The same principle holds for random elements of metric spaces.