* custom, stateful strategy we may define (e.g. any kind of strategy that would depend on user
     * input).
     */
    int ordinal();
  }

  /** The bit set of the BloomFilter (not necessarily power of 2!) */
  private final LockFreeBitArray bits;

  /** Number of hashes per element */
  private final int numHashFunctions;

  /** The funnel to translate Ts to bytes */

   * situations for compatibility.
   *
   * @since 17.0
   */
  public static final MediaType WOFF = createConstant(APPLICATION_TYPE, "font-woff");

  /**
   * <a href="https://tools.ietf.org/html/rfc8081">RFC 8081</a> declares {@link #FONT_WOFF2
   * font/woff2} to be the correct media type for WOFF2, but this may be necessary in certain
   * situations for compatibility.

 * return hex.decode("deadbeef"); // throws an IllegalArgumentException
 * }
 *
 * <p>It is guaranteed that {@code encoding.decode(encoding.encode(x))} is always equal to {@code
 * x}, but the reverse does not necessarily hold.
 *
 * <table>
 * <caption>Encodings</caption>
 * <tr>
 * <th>Encoding
 * <th>Alphabet
 * <th>{@code char:byte} ratio
 * <th>Default padding
 * <th>Comments
 * <tr>
 * <td>{@link #base16()}

 *       deadlock situation, and the appropriate Policy is executed.
 * </ul>
 *
 * <p>Note that detection of potential deadlock does not necessarily indicate that deadlock will
 * happen, as it is possible that higher level application logic prevents the cyclic lock
 * acquisition from occurring. One example of a false positive is:
 *
 * <pre>

   * table is internally partitioned to try to permit the indicated number of concurrent updates
   * without contention. Because assignment of entries to these partitions is not necessarily
   * uniform, the actual concurrency observed may vary. Ideally, you should choose a value to
   * accommodate as many threads as will ever concurrently modify the table. Using a significantly

   *       complex cases may require using thread pools or making deeper changes.)
   *   <li>If an exception propagates out of a {@code Runnable}, it is not necessarily seen by any
   *       {@code UncaughtExceptionHandler} for the thread. For example, if the callback passed to
   *       {@link Futures#addCallback} throws an exception, that exception will be typically be

   * that this method <i>also</i> throws if this future is not done.
   *
   * <p>This method computes its result based on the internal state of {@link AbstractFuture}, so it
   * does not necessarily return the same result as {@link #get()} if {@link #get()} has been
   * overridden. Thus, it should be called only on instances of {@link Trusted} or from within
   * {@link #get()} itself.
   */
  @ParametricNullness

   * types are parameterized with proper type arguments.
   *
   * <p>Subtypes are always listed before supertypes. But the reverse is not true. A type isn't
   * necessarily a subtype of all the types following. Order between types without subtype
   * relationship is arbitrary and not guaranteed.
   *

   * table is internally partitioned to try to permit the indicated number of concurrent updates
   * without contention. Because assignment of entries to these partitions is not necessarily
   * uniform, the actual concurrency observed may vary. Ideally, you should choose a value to
   * accommodate as many threads as will ever concurrently modify the table. Using a significantly

<code>(R1)</code> (register indirect),
<code>4(R1)</code> (register indirect with offset), and
<code>$foo(SB)</code> (absolute address).
The assembler also supports some (not necessarily all) addressing modes
specific to each architecture.
The sections below list these.
</p>

<p>