* @param executor the executor in which the listeners callback methods will be run. For fast,
   *     lightweight listeners that would be safe to execute in any thread, consider {@link
   *     MoreExecutors#directExecutor}.
   * @since 13.0
   */
  void addListener(Listener listener, Executor executor);

  /**
   * The lifecycle states of a service.
   *

  virtual void EndStep() = 0;

  // Return the Eager Executor for current thread. Please note that Eager
  // Executor is only used in current TF but not in TFRT.
  virtual EagerExecutor& Executor() = 0;
  // Update the Eager Executor for current thread.
  virtual void SetExecutorForThread(EagerExecutor* executor) = 0;

  // Return a list of local tensorflow::Device*.

      assertEquals(Service.State.STOPPING, state());
      shutdownCalled++;
    }

    @Override
    protected Executor executor() {
      return executor;
    }

    @Override
    public void tearDown() throws Exception {
      executor.shutdown();
    }
  }

      Function<? super Throwable, ? extends V> fallback,
      Executor executor) {
    return (FluentFuture<V>) Futures.catching(this, exceptionType, fallback, executor);
  }

  public final FluentFuture<V> catchingAsync(
      Class<Throwable> exceptionType,
      AsyncFunction<? super Throwable, ? extends V> fallback,
      Executor executor) {

      Executor executor) {
    return AbstractCatchingFuture.create(input, exceptionType, fallback, executor);
  }

  public static <V extends @Nullable Object> ListenableFuture<V> catchingAsync(
      ListenableFuture<? extends V> input,
      Class<Throwable> exceptionType,
      AsyncFunction<? super Throwable, ? extends V> fallback,
      Executor executor) {

Directory for DBFlute client

manage.bat(sh) => 21 (jdbc):
A execution command of JDBC task
which gets your schema info and saves it to SchemaXML
located to the "schema" directory.
This task should be executed after ReplaceSchema task
and before other tasks(e.g. Generate, Document task).

manage.bat(sh) => 22 (doc):
A execution command of Document task
which creates documents, for example, SchemaHTML, HistoryHTML
to the "output/doc" directory.

    }
  }

  private static final class Listener {
    final Runnable command;
    final Executor executor;

    Listener(Runnable command, Executor executor) {
      this.command = checkNotNull(command);
      this.executor = checkNotNull(executor);
    }

    void execute() {
      try {
        executor.execute(command);
      } catch (RuntimeException e) {
        log.log(
            Level.SEVERE,

  private static AsyncCallable<Integer> asyncAdd(
      final ListenableFuture<Integer> future, final int delta, final Executor executor) {
    return new AsyncCallable<Integer>() {
      @Override
      public ListenableFuture<Integer> call() throws Exception {
        return Futures.transform(future, add(delta), executor);
      }
    };
  }

  private static final class LongHolder {
    long count;
  }

* **Starlette**:
    * Tendrá el siguiente mejor desempeño, después de Uvicorn. De hecho, Starlette usa Uvicorn para correr. Por lo tanto, probablemente sólo pueda volverse "más lento" que Uvicorn al tener que ejecutar más código.

programs that TensorFlow executes. TensorFlow programs are encoded as
computation
[**graphs**](https://developers.google.com/machine-learning/glossary/#graph).
Since models are practically programs that TensorFlow executes, using untrusted
models or graphs is equivalent to running untrusted code.

If you need to run untrusted models, execute them inside a