Currently the LCP is computed naively if there is at most S*-suffix in the text. A more efficient algorithm would be preferred,