}
              });
      thread.start();
      // we want to wait until each thread is WAITING - one thread waiting inside CacheLoader.load
      // (in startSignal.await()), and the others waiting for that thread's result.
      while (thread.isAlive() && thread.getState() != Thread.State.WAITING) {
        Thread.yield();
      }
    }
    gettersStartedSignal.countDown();

            SingletonLaContainerFactory.init();

            final Thread shutdownCallback = new Thread("ShutdownHook") {
                @Override
                public void run() {
                    destroyContainer();
                }

            };
            Runtime.getRuntime().addShutdownHook(shutdownCallback);

            commandThread = new Thread(() -> {

	buf := make([]byte, 1)
	if _, err := b.Read(nil); err != nil {
		t.Fatalf("1st nil Read = %v; want nil", err)
	}
	if _, err := b.Read(buf); err != errFake {
		t.Fatalf("1st Read = %v; want errFake", err)
	}
	if _, err := b.Read(nil); err != nil {
		t.Fatalf("2nd nil Read = %v; want nil", err)
	}
	if _, err := b.Read(buf); err != nil {
		t.Fatalf("3rd Read with buffer = %v; want nil", err)
	}

```

</div>

<details markdown="1">
<summary>À propos de la commande <code>uvicorn main:app --reload</code> ...</summary>

La commande `uvicorn main:app` fait référence à :

* `main` : le fichier `main.py` (le "module" Python).
* `app` : l'objet créé à l'intérieur de `main.py` avec la ligne `app = FastAPI()`.
* `--reload` : fait redémarrer le serveur après des changements de code. À n'utiliser que pour le développement.

```Python
{!../../docs_src/first_steps/tutorial001.py!}
```

Copia eso a un archivo `main.py`.

Corre el servidor en vivo:

<div class="termy">

```console
$ uvicorn main:app --reload

<span style="color: green;">INFO</span>:     Uvicorn running on http://127.0.0.1:8000 (Press CTRL+C to quit)
<span style="color: green;">INFO</span>:     Started reloader process [28720]

   * CacheLoader#reload}.
   *
   * <p>As the default implementation of {@link CacheLoader#reload} is synchronous, it is
   * recommended that users of this method override {@link CacheLoader#reload} with an asynchronous
   * implementation; otherwise refreshes will be performed during unrelated cache read and write
   * operations.
   *

  @Footprint(exclude = {Runnable.class, Executor.class, Thread.class, Exception.class})
  public Object measureSize() {
    for (Thread thread : blockedThreads) {
      thread.interrupt();
    }
    blockedThreads.clear();
    final Facade<Object> f = impl.newFacade();
    for (int i = 0; i < numThreads; i++) {
      Thread thread =
          new Thread() {
            @Override
            public void run() {

   * CacheLoader#reload}.
   *
   * <p>As the default implementation of {@link CacheLoader#reload} is synchronous, it is
   * recommended that users of this method override {@link CacheLoader#reload} with an asynchronous
   * implementation; otherwise refreshes will be performed during unrelated cache read and write
   * operations.
   *

  Thread newStartedThread(Runnable runnable) {
    Thread t = new Thread(runnable);
    t.setDaemon(true);
    t.start();
    return t;
  }

  /**
   * Waits for the specified time (in milliseconds) for the thread to terminate (using {@link
   * Thread#join(long)}), else interrupts the thread (in the hope that it may terminate later) and
   * fails.
   */

    @Override
    public boolean offer(E o) {
      assertTrue(Thread.holdsLock(mutex));
      return delegate.offer(o);
    }

    @Override
    public @Nullable E poll() {
      assertTrue(Thread.holdsLock(mutex));
      return delegate.poll();
    }

    @Override
    public E remove() {
      assertTrue(Thread.holdsLock(mutex));
      return delegate.remove();
    }

    @Override