*
   *  * Invalid percent-encoded sequences like `%xx` will be encoded like `%25xx`.
   *
   *  * Whitespace and control characters in the fragment will be stripped.
   *
   * These differences may have a significant consequence when the URI is interpreted by a
   * web server. For this reason the [URI class][URI] and this method should be avoided.
   */
  @JvmName("uri")
  fun toUri(): URI {

        return set1.contains(element) ^ set2.contains(element);
      }

      @Override
      int minSize() {
        int difference = minSize(set1) - maxSize(set2);
        return difference >= 0 ? difference : max(minSize(set2) - maxSize(set1), 0);
      }

      @Override
      int maxSize() {
        return saturatedAdd(maxSize(set1), maxSize(set2));
      }

     * Another way to express this signature would be to bound <V> by @NonNull and accept
     * LF<? extends @Nullable V>. That might be better: There's currently no difference between the
     * outputs users get when calling this with <Foo> and calling it with <@Nullable Foo>. The only
     * difference is that calling it with <Foo> won't work when an input Future has a @Nullable
     * type. So why even make that error possible by giving callers the choice?

   *
   * <p>As we iterate from {@code 0} to {@code nThreads}, threads with an even index will call
   * {@code getUnchecked}, and threads with an odd index will call {@code get}. If the cache throws
   * exceptions, this difference may be visible in the returned List.
   */
  @SuppressWarnings("ThreadPriorityCheck") // TODO: b/175898629 - Consider onSpinWait.
  private static <K> List<Object> doConcurrentGet(