case 0:
        return ImmutableSet.of();
      case 1:
        return ImmutableSet.of(getOnlyElement(set));
      default:
        return new ImmutableEnumSet<>(set);
    }
  }

  /*
   * Notes on EnumSet and <E extends Enum<E>>:
   *
   * This class isn't an arbitrary ForwardingImmutableSet because we need to
   * know that calling {@code clone()} during deserialization will return an

 *   <li><a href="https://github.com/google/guava/wiki/GraphsExplained#synchronization">
 *       Synchronization policy</a>
 *   <li><a href="https://github.com/google/guava/wiki/GraphsExplained#notes-for-implementors">Notes
 *       for implementors</a>
 * </ul>
 *
 * @author James Sexton
 * @author Joshua O'Madadhain
 * @param <N> Node parameter type
 * @param <E> Edge parameter type
 * @since 20.0
 */

   * all are provided (an empty list), the resulting Cartesian product has one element, an empty
   * list (counter-intuitive, but mathematically consistent).
   *
   * <p><i>Performance notes:</i> while the cartesian product of sets of size {@code m, n, p} is a
   * set of size {@code m x n x p}, its actual memory consumption is much smaller. When the

   * at all are provided (an empty list), the resulting Cartesian product has one element, an empty
   * list (counter-intuitive, but mathematically consistent).
   *
   * <p><i>Performance notes:</i> while the cartesian product of lists of size {@code m, n, p} is a
   * list of size {@code m x n x p}, its actual memory consumption is much smaller. When the

 *       mutable objects as elements at all, as many users may expect your "immutable" object to be
 *       <i>deeply</i> immutable.
 * </ul>
 *
 * <h4>Performance notes</h4>
 *
 * <ul>
 *   <li>Implementations can be generally assumed to prioritize memory efficiency, then speed of
 *       access, and lastly speed of creation.

 *       mutable objects as elements at all, as many users may expect your "immutable" object to be
 *       <i>deeply</i> immutable.
 * </ul>
 *
 * <h4>Performance notes</h4>
 *
 * <ul>
 *   <li>Implementations can be generally assumed to prioritize memory efficiency, then speed of
 *       access, and lastly speed of creation.

    - [Server Binaries](#server-binaries-4)
    - [Node Binaries](#node-binaries-4)
    - [Container Images](#container-images-4)
  - [Changelog since v1.32.0](#changelog-since-v1320)
  - [Urgent Upgrade Notes](#urgent-upgrade-notes)
    - [(No, really, you MUST read this before you upgrade)](#no-really-you-must-read-this-before-you-upgrade)
  - [Changes by Kind](#changes-by-kind-4)
    - [Deprecation](#deprecation)

 * for comparisons. This class is not being deprecated, but we gently encourage you to migrate to
 * streams.
 *
 * <p><i>Performance notes:</i> Unless otherwise noted, all of the iterables produced in this class
 * are <i>lazy</i>, which means that their iterators only advance the backing iteration when
 * absolutely necessary.
 *

    - [Server Binaries](#server-binaries-8)
    - [Node Binaries](#node-binaries-8)
    - [Container Images](#container-images-8)
  - [Changelog since v1.31.0](#changelog-since-v1310)
  - [Urgent Upgrade Notes](#urgent-upgrade-notes)
  - [Changes by Kind](#changes-by-kind-8)
    - [Deprecation](#deprecation)
    - [API Change](#api-change-2)
    - [Feature](#feature-5)
    - [Documentation](#documentation)

    - [Server Binaries](#server-binaries-12)
    - [Node Binaries](#node-binaries-12)
    - [Container Images](#container-images-12)
  - [Changelog since v1.30.0](#changelog-since-v1300)
  - [Urgent Upgrade Notes](#urgent-upgrade-notes)
    - [(No, really, you MUST read this before you upgrade)](#no-really-you-must-read-this-before-you-upgrade)
  - [Changes by Kind](#changes-by-kind-12)
    - [Deprecation](#deprecation-1)