* for use in test code.
 *
 * <p><b>Do not use this interface</b> for public or protected declarations: it is a fig leaf for
 * bad design, and it does not prevent anyone from using the declaration---and experience has shown
 * that they will. If the method breaks the encapsulation of its class, then its internal
 * representation will be hard to change. Instead, use <a

   *
   * @since 20.0
   */
  public static final String FORWARDED = "Forwarded";
  /**
   * The HTTP {@code Follow-Only-When-Prerender-Shown} header field name.
   *
   * @since 17.0
   */
  public static final String FOLLOW_ONLY_WHEN_PRERENDER_SHOWN = "Follow-Only-When-Prerender-Shown";
  /** The HTTP {@code Host} header field name. */
  public static final String HOST = "Host";
  /**

 * for use in test code.
 *
 * <p><b>Do not use this interface</b> for public or protected declarations: it is a fig leaf for
 * bad design, and it does not prevent anyone from using the declaration---and experience has shown
 * that they will. If the method breaks the encapsulation of its class, then its internal
 * representation will be hard to change. Instead, use <a

 * </ul>
 *
 * <p><b>Important:</b> although the first interpretation resembles how most multimaps are
 * <i>implemented</i>, the design of the {@code Multimap} API is based on the <i>second</i> form.
 * So, using the multimap shown above as an example, the {@link #size} is {@code 3}, not {@code 2},
 * and the {@link #values} collection is {@code [1, 2, 3]}, not {@code [[1, 2], [3]]}. For those

   * EndpointPair}s, we keep track of the nodes that we have visited. When processing endpoint
   * pairs, we skip if the "other node" is in the visited set, as shown below:
   *
   * <pre>
   * Nodes = {N1, N2, N3, N4}
   *    N2           __
   *   /  \         |  |
   * N1----N3      N4__|
   *
   * Visited Nodes = {}
   * EndpointPair [N1, N2] - return

    return this;
  }

  private long elapsedNanos() {
    return isRunning ? ticker.read() - startTick + elapsedNanos : elapsedNanos;
  }

  /**
   * Returns the current elapsed time shown on this stopwatch, expressed in the desired time unit,
   * with any fraction rounded down.
   *
   * <p><b>Note:</b> the overhead of measurement can be more than a microsecond, so it is generally

 *       }
 *     };
 * iteratorTester.test();
 * iteratorTester.testForEachRemaining();
 * }</pre>
 *
 * <p><b>Note</b>: It is necessary to use {@code IteratorTester.KnownOrder} as shown above, rather
 * than {@code KnownOrder} directly, because otherwise the code cannot be compiled.
 *
 * @author Kevin Bourrillion
 * @author Chris Povirk
 */
@GwtCompatible
@ElementTypesAreNonnullByDefault

     * into {@code horizon} between calls to {@code visitNext()}. This causes them to receive
     * additional values interleaved with those shown above.)
     */
    @CheckForNull
    abstract N visitNext(Deque<Iterator<? extends N>> horizon);
  }

  /** Poor man's method reference for {@code Deque::addFirst} and {@code Deque::addLast}. */

 * later RFCs with the apparent aim of making IPv4-to-IPv6 transition simpler.
 *
 * <p>Technically one <i>can</i> create a 128bit IPv6 address with the wire format of a "mapped"
 * address, as shown above, and transmit it in an IPv6 packet header. However, Java's InetAddress
 * creation methods appear to adhere doggedly to the original intent of the "mapped" address: all
 * "mapped" addresses return {@link Inet4Address} objects.

 *       }
 *     };
 * iteratorTester.test();
 * iteratorTester.testForEachRemaining();
 * }</pre>
 *
 * <p><b>Note</b>: It is necessary to use {@code IteratorTester.KnownOrder} as shown above, rather
 * than {@code KnownOrder} directly, because otherwise the code cannot be compiled.
 *
 * @author Kevin Bourrillion
 * @author Chris Povirk
 */
@GwtCompatible
@ElementTypesAreNonnullByDefault