The implementation of URI.toURL() currently doesn't leverage the parsing result of URI, and instead calls URL to parse a second time. This needs to be improved.