*
   * <p><b>{@code Stream} equivalent:</b> use {@code stream.collect(ImmutableMap.toImmutableMap(k ->
   * k, valueFunction))}. {@code ImmutableMap.copyOf(stream.collect(Collectors.toMap(k -> k,
   * valueFunction)))} behaves similarly, but may not preserve the order of entries.
   *
   * @throws NullPointerException if any element of this iterable is {@code null}, or if {@code
   *     valueFunction} produces {@code null} for any key
   * @since 14.0

 *     .weakKeys()
 *     .makeMap();
 * }
 *
 * <p>These features are all optional; {@code new MapMaker().makeMap()} returns a valid concurrent
 * map that behaves similarly to a {@link ConcurrentHashMap}.
 *
 * <p>The returned map is implemented as a hash table with similar performance characteristics to
 * {@link ConcurrentHashMap}. It supports all optional operations of the {@code ConcurrentMap}
 * interface. It does not permit null keys or values.

       * AGP: https://github.com/google/guava/issues/7769.
       *
       * My impression is that an AtomicReferenceFieldUpdater in a static field is similarly fast to
       * Unsafe on modern JVMs (if perhaps not quite as fast as VarHandle?). However, I'm not sure
       * exactly what we've benchmarked, and we certainly haven't benchmarked as far back as JDK 8.

   * corresponding to each key appear in the same order as they are encountered.
   *
   * <p><b>{@code Stream} equivalent:</b> {@code stream.collect(Collectors.groupingBy(keyFunction))}
   * behaves similarly, but returns a mutable {@code Map<K, List<E>>} instead, and may not preserve
   * the order of entries.
   *
   * @param keyFunction the function used to produce the key for each value

    }

    // Test hook method when MultiThreadedHttpConnectionManager class is not available
    public void test_hook_classNotFound() {
        // Create a custom hook that simulates class not found scenario
        FessCurtainFinallyHook customHook = new FessCurtainFinallyHook() {
            @Override
            public void hook(final FwAssistantDirector assistantDirector) {

 * handling text normalization, content extraction, and similar document hash encoding/decoding.
 * It also manages document processing requests and integrates with the crawler system.
 *
 */
public class DocumentHelper {
    private static final Logger logger = LogManager.getLogger(DocumentHelper.class);

    /** Prefix used for encoded similar document hashes */
    protected static final String SIMILAR_DOC_HASH_PREFIX = "$";

   * the discussion in the {@link #addListener} documentation. All its warnings about heavyweight
   * listeners are also applicable to heavyweight functions passed to this method.
   *
   * <p>This method is similar to {@link java.util.concurrent.CompletableFuture#exceptionally}. It
   * can also serve some of the use cases of {@link java.util.concurrent.CompletableFuture#handle}

...
BUILD SUCCESSFUL in 1m 30s
63 actionable tasks: 61 executed, 2 up-to-date

```

    } finally {
      if (interrupted) {
        Thread.currentThread().interrupt();
      }
    }
  }

  /**
   * Invokes {@code future.}{@link Future#get() get()} uninterruptibly.
   *
   * <p>Similar methods:
   *
   * <ul>
   *   <li>To retrieve a result from a {@code Future} that is already done, use {@link
   *       Futures#getDone Futures.getDone}.

   *
   * - https://github.com/jspecify/jdk/commit/71d826792b8c7ef95d492c50a274deab938f2552
   */
  /*
   * TODO(cpovirk): Is the unchecked cast avoidable? Would System.arraycopy be similarly fast (if
   * likewise not type-checked)? Could our single caller do something different?
   */
  @SuppressWarnings({"nullness", "unchecked"})