*
 * <p>Subclasses are guaranteed that the life cycle methods ({@link #runOneIteration}, {@link
 * #startUp} and {@link #shutDown}) will never run concurrently. Notably, if any execution of {@link
 * #runOneIteration} takes longer than its schedule defines, then subsequent executions may start
 * late. Also, all life cycle methods are executed with a lock held, so subclasses can safely modify

 * a {@code Map<R, Map<C, V>>}. A method call like {@code column(columnKey).get(rowKey)} still runs
 * quickly, since the row key is provided. However, {@code column(columnKey).size()} takes longer,
 * since an iteration across all row keys occurs.
 *
 * <p>Because a {@code TreeBasedTable} has unique sorted values for a given row, both {@code

 *     return map.isEmpty();
 *   }
 * });
 * }</pre>
 *
 * <p>Even if your non-test code does not use finalization, you can use this class to test for
 * leaks, by ensuring that objects are no longer strongly referenced:
 *
 * <pre>{@code
 * // Helper function keeps victim stack-unreachable.
 * private WeakReference<Foo> fooWeakRef() {
 *   Foo x = ....;
 *   WeakReference<Foo> weakRef = new WeakReference<>(x);

   * until a nonzero result is found; imposes "dictionary order." If the end of one iterable is
   * reached, but not the other, the shorter iterable is considered to be less than the longer one.
   * For example, a lexicographical natural ordering over integers considers {@code [] < [1] < [1,
   * 1] < [1, 2] < [2]}.
   *

        for (int i = 0; i < array.length; i++) {
          array[i] = counter.getAndIncrement();
        }
        builder.addAll(ImmutableIntArray.copyOf(array));
      }
    },
    ADD_LARGER_ARRAY {
      @Override
      void doIt(ImmutableIntArray.Builder builder, AtomicInteger counter) {
        int[] array = new int[RANDOM.nextInt(200) + 200];
        for (int i = 0; i < array.length; i++) {

 * Stats}, {@link PairedStatsAccumulator}, and {@link PairedStats}.
 *
 * @author Pete Gillin
 */
class StatsTesting {
  // TODO(cpovirk): Convince myself that this larger error makes sense.
  static final double ALLOWED_ERROR = isAndroid() ? .25 : 1e-10;

  // Inputs and their statistics:

  static final double ONE_VALUE = 12.34;

  static final double OTHER_ONE_VALUE = -56.78;

   * "e"]}. The last piece can be smaller than {@code length} but will never be empty.
   *
   * <p><b>Note:</b> if {@link #fixedLength} is used in conjunction with {@link #limit}, the final
   * split piece <i>may be longer than the specified fixed length</i>. This is because the splitter
   * will <i>stop splitting when the limit is reached</i>, and just return the final piece as-is.
   *

 * find it easier to use a framework. Frameworks automate the process, often adding features like
 * monitoring, debugging, and cancellation. Examples of frameworks include:
 *
 * <ul>
 *   <li><a href="https://dagger.dev/producers.html">Dagger Producers</a>
 * </ul>
 *
 * <p>If you do chain your operations manually, you may want to use {@link FluentFuture}.
 *
 * @author Kevin Bourrillion
 * @author Nishant Thakkar
 * @author Sven Mawson

 * done by determining whether the domain ended with a {@linkplain #isPublicSuffix() public suffix}
 * but was not itself a public suffix. However, this test is no longer accurate. There are many
 * domains which are both public suffixes and addressable as hosts; {@code "uk.com"} is one example.
 * Using the subset of public suffixes that are {@linkplain #isRegistrySuffix() registry suffixes},

   *
   * @param value any value in the range of the {@code int} type
   * @return the {@code int} value that equals {@code value}
   * @throws IllegalArgumentException if {@code value} is greater than {@link Integer#MAX_VALUE} or
   *     less than {@link Integer#MIN_VALUE}
   */
  public static int checkedCast(long value) {
    int result = (int) value;
    checkArgument(result == value, "Out of range: %s", value);
    return result;
  }