## Version 2.4.0

_2015-05-22_

 *  **Forbid response bodies on HTTP 204 and 205 responses.** Webservers that
    return such malformed responses will now trigger a `ProtocolException` in
    the client.

 *  **WebSocketListener has incompatible changes.** The `onOpen()` method is now
    called on the reader thread, so implementations must return before further

  public <T> void testVariableTypeTokenNotAllowed() {
    /*
     * We'd use assertThrows here, but that causes no exception to be thrown under Java 8,
     * presumably because the ThrowingRunnable lambda triggers some kind of bug in Java 8's
     * reflection implementation.
     */
    try {
      new TypeToken<T>() {};
      fail();
    } catch (IllegalStateException expected) {
      // Type variables aren't allowed.

    return new SingletonImmutableSet<>(e1);
  }

  /*
   * TODO: b/315526394 - Skip the Builder entirely for the of(...) methods, since we don't need to
   * worry that we might trigger the fallback to the JDK-backed implementation? (The varargs one
   * _could_, so we could keep it as it is. Or we could convince ourselves that hash flooding is
   * unlikely in practice there, too.)
   */

  /**

  /**
   * This is a non-standard media type, but is commonly used in serving hosted binary files as it is
   * <a href="http://code.google.com/p/browsersec/wiki/Part2#Survey_of_content_sniffing_behaviors">
   * known not to trigger content sniffing in current browsers</a>. It <i>should not</i> be used in
   * other situations as it is not specified by any RFC and does not appear in the <a

func auditLogInternal(ctx context.Context, opts AuditLogOptions) {
	if len(logger.AuditTargets()) == 0 {
		return
	}

	entry := audit.NewEntry(globalDeploymentID())
	entry.Trigger = opts.Event
	entry.Event = opts.Event
	entry.Error = opts.Error
	entry.API.Name = opts.APIName
	entry.API.Bucket = opts.Bucket

 * function doing a good job of distributing the elements to the buckets to a distribution not far
 * from uniform), and amortized since some operations can trigger a hash table resize.
 *
 * <p>Unlike {@code java.util.HashMap}, iteration is only proportional to the actual {@code size()},
 * which is optimal, and <i>not</i> the size of the internal hashtable, which could be much larger

build:nonccl --define=no_nccl_support=true

# Modular TF build options
build:dynamic_kernels --define=dynamic_loaded_kernels=true
build:dynamic_kernels --copt=-DAUTOLOAD_DYNAMIC_KERNELS

# Don't trigger --config=<host platform> when cross-compiling.
build:android --noenable_platform_specific_config
build:ios --noenable_platform_specific_config

# Suppress all C++ compiler warnings, otherwise build logs become 10s of MBs.

If this happens to you, you can wait at most 10 minutes for the replication controller to start a resync, the extra pods will then be deleted. Or, you can manually trigger a resync by change the replicas in the spec of the replication controller.

### kubectl delete: < v1.4.0 client vs >=v1.4.0 cluster

   * #finishToValueAndCloser(ValueAndCloserConsumer, Executor)}, or any other derivation method on
   * the original {@code ClosingFuture} instance.
   *
   * @param exceptionType the exception type that triggers use of {@code fallback}. The exception
   *     type is matched against this step's exception. "This step's exception" means the cause of
   *     the {@link ExecutionException} thrown by {@link Future#get()} on the {@link Future}

  /**
   * This is a non-standard media type, but is commonly used in serving hosted binary files as it is
   * <a href="http://code.google.com/p/browsersec/wiki/Part2#Survey_of_content_sniffing_behaviors">
   * known not to trigger content sniffing in current browsers</a>. It <i>should not</i> be used in
   * other situations as it is not specified by any RFC and does not appear in the <a