// it. To preserve FIFO order and failure atomicity of rejected execution when the same
      // Runnable is executed more than once, allocate a wrapper that we know is safe to remove by
      // object identity.
      // A data structure that returned a removal handle from add() would allow eliminating this
      // allocation.
      submittedTask =
          new Runnable() {
            @Override

      // it. To preserve FIFO order and failure atomicity of rejected execution when the same
      // Runnable is executed more than once, allocate a wrapper that we know is safe to remove by
      // object identity.
      // A data structure that returned a removal handle from add() would allow eliminating this
      // allocation.
      submittedTask =
          new Runnable() {
            @Override

The main limitation is that to update or delete content of a file inside a ZIP file the entire ZIP file must be replaced.

## How to enable S3 ZIP behavior ?

  /**
   * Returns a dispatcher that queues events that are posted reentrantly on a thread that is already
   * dispatching an event, guaranteeing that all events posted on a single thread are dispatched to
   * all subscribers in the order they are posted.
   *
   * <p>When all subscribers are dispatched to using a <i>direct</i> executor (which dispatches on

   * on the fields of TimeoutFuture. This field is slightly different from the fields discussed
   * there: cancel() never reads this field, only writes to it. That makes the race here completely
   * harmless, rather than just 99.99% harmless.
   */
  @LazyInit private @Nullable List<@Nullable Present<V>> values;

  @SuppressWarnings("EmptyList") // ImmutableList doesn't support nullable element types

     * exception, and wrongly believes that its exception is new (leading it to logging it when it
     * shouldn't)
     *
     * Our solution is for threads to CAS seenExceptionsField from null to a Set populated with _the
     * initial exception_, no matter which thread does the work. This ensures that
     * seenExceptionsField always contains not just the current thread's exception but also the

   * That is, the first call to {@link #nextIndex()} will return {@code position}, and the first
   * call to {@link #next()} will return the element at that index, if available. Calls to {@link
   * #previous()} can retrieve the preceding {@code position} elements.
   *
   * @throws IndexOutOfBoundsException if {@code position} is negative or is greater than {@code
   *     size}

    }

    /**
     * Test that buffer cache handles allocation when cache is full.
     */
    @Test
    public void testBufferAllocationWhenCacheFull() {
        // Given - Allocate many buffers to potentially fill the cache
        List<byte[]> buffers = new ArrayList<>();

        // Allocate more buffers than cache can hold
        for (int i = 0; i < 20; i++) {

	dynamicTimeoutLogSize              = 16
	maxDuration                        = math.MaxInt64
	maxDynamicTimeout                  = 24 * time.Hour // Never set timeout bigger than this.
)

// timeouts that are dynamically adapted based on actual usage results
type dynamicTimeout struct {
	timeout       int64
	minimum       int64
	entries       int64
	log           [dynamicTimeoutLogSize]time.Duration

      // https://github.com/openjdk/jdk/commit/3e393047e12147a81e2899784b943923fc34da8e, a bug
      // meant that sometimes a too-large threshold is calculated. However, this new threshold is
      // independent of the initial capacity, except that it won't be lower than the threshold
      // computed from that capacity. Because the internal table is only allocated on the first