* only under concurrent modification).
     *
     * To access these fields when you know that they are not null, call the pred() and succ()
     * methods, which perform null checks before returning the fields.
     */
    private @Nullable AvlNode<E> pred;
    private @Nullable AvlNode<E> succ;

    AvlNode(@ParametricNullness E elem, int elemCount) {

   *   <li>Use sendfile(2) or equivalent. Requires that both the input channel and the output
   *       channel have their own file descriptors. Generally this only happens when both channels
   *       are files or sockets. This performs zero copies - the bytes never enter userspace.
   *   <li>Use mmap(2) or equivalent. Requires that either the input channel or the output channel

   * to be a view, but it means that the function will be applied many times for bulk operations
   * like {@link List#contains} and {@link List#hashCode}. For this to perform well, {@code
   * function} should be fast. To avoid lazy evaluation when the returned list doesn't need to be a
   * view, copy the returned list into a new list of your choosing.
   *

   * ListenableFuture} listeners, should take care not to do so while holding a lock. Additionally,
   * as a further line of defense, prefer not to perform any locking inside a task that will be run
   * under {@code directExecutor}: Not only might the wait for a lock be long, but if the running
   * thread was holding a lock, the listener may deadlock or break lock isolation.
   *

    is invoked, then you must make a good faith effort to ensure that,
    in the event an application does not supply such function or
    table, the facility still operates, and performs whatever part of
    its purpose remains meaningful.

    (For example, a function in a library to compute square roots has
    a purpose that is entirely well-defined independent of the

   *
   * <p>If {@code expectedSize} is exactly the number of distinct elements added to the builder
   * before {@link Builder#build} is called, the builder is likely to perform better than an unsized
   * {@link #builder()} would have.
   *
   * <p>It is not specified if any performance benefits apply if {@code expectedSize} is close to,

      return getImpl();
    }
  }

  private static class QueueSupplier extends CountingSupplier<Queue<Integer>> {
    @Override
    /*
     * We need a Queue that implements equals() for the equality tests we perform after
     * reserializing the multimap.
     */
    @SuppressWarnings("JdkObsolete")
    public Queue<Integer> getImpl() {
      return new LinkedList<>();
    }

     * only under concurrent modification).
     *
     * To access these fields when you know that they are not null, call the pred() and succ()
     * methods, which perform null checks before returning the fields.
     */
    private @Nullable AvlNode<E> pred;
    private @Nullable AvlNode<E> succ;

    AvlNode(@ParametricNullness E elem, int elemCount) {

            }
        }
        if (sb.length() > 0) {
            return "failure in bulk execution:" + sb.toString();
        }
        return StringUtil.EMPTY;
    }

    /**
     * Performs the actual bulk insertion of documents into the OpenSearch index.
     *
     * @param <T> The type of objects to insert.
     * @param list The list of objects to insert.

   * desiredType}. The returned iterable's iterator does not support {@code remove()}.
   *
   * <p><b>{@code Stream} equivalent:</b> {@code stream.filter(type::isInstance).map(type::cast)}.
   * This does perform a little more work than necessary, so another option is to insert an
   * unchecked cast at some later point:
   *
   * {@snippet :
   * @SuppressWarnings("unchecked") // safe because of ::isInstance check