{* ../../docs_src/background_tasks/tutorial002_an_py310.py hl[13,15,22,25] *}

En este ejemplo, los mensajes se escribirán en el archivo `log.txt` *después* de que se envíe el response.

Si hay un query en el request, se escribirá en el log en una tarea en segundo plano.

* Luego devuelve la **response**.

/// note | Detalles Técnicos

Si tienes dependencias con `yield`, el código de salida se ejecutará *después* del middleware.

Si hubiera tareas en segundo plano (cubiertas en la sección [Tareas en segundo plano](background-tasks.md){.internal-link target=_blank}, lo verás más adelante), se ejecutarán *después* de todo el middleware.

///

## Crear un middleware { #create-a-middleware }

* [LastaFlute](https://github.com/lastaflute/lastaflute "LastaFlute"): Framework Web
* [Lasta Job](https://github.com/lastaflute/lasta-job "Lasta Job"): Programador de Tareas
* [Fess Crawler](https://github.com/codelibs/fess-crawler "Fess Crawler"): Rastreador Web

				errs = append(errs, e)
			}
		}
	}
	return tgts, errs
}

// Split targets into two groups:
//
//	group1 contains all targets of type t
//	group2 contains the remaining targets
func splitTargets(targets []Target, t types.TargetType) (group1 []Target, group2 []Target) {
	for _, target := range targets {
		if target.Type() == t {
			group1 = append(group1, target)
		} else {

* Então ele retorna a **resposta**.

/// note | Detalhes Técnicos

Se você tiver dependências com `yield`, o código de saída será executado *depois* do middleware.

Se houver alguma tarefa em segundo plano (abordada na seção [Tarefas em segundo plano](background-tasks.md){.internal-link target=_blank}, que você verá mais adiante), ela será executada *depois* de todo o middleware.

///

## Criar um middleware { #create-a-middleware }

## Adicione a tarefa em segundo plano { #add-the-background-task }

Dentro da sua *função de operação de rota*, passe sua função de tarefa para o objeto de *tarefas em segundo plano* com o método `.add_task()`:

{* ../../docs_src/background_tasks/tutorial001_py39.py hl[14] *}

O `.add_task()` recebe como argumentos:

* Uma função de tarefa a ser executada em segundo plano (`write_notification`).

            } else {
                t.url = args[ai];
            }
        }

        final Thread[] threads = new Thread[t.numThreads];
        int ti;

        for (ti = 0; ti < t.numThreads; ti++) {
            threads[ti] = new Thread(t);
            logger.info(threads[ti].getName());
            threads[ti].start();
        }

        while (t.numComplete < t.numThreads) {
            long delay;

 *   <li>a {@linkplain ThreadFactoryBuilder#setNameFormat naming format}
 *   <li>a {@linkplain Thread#setPriority thread priority}
 *   <li>an {@linkplain Thread#setUncaughtExceptionHandler uncaught exception handler}
 *   <li>a {@linkplain ThreadFactory#newThread backing thread factory}
 * </ul>
 *
 * <p>If no backing thread factory is provided, a default backing thread factory is used as if by

#### Do-stuff-later pool

Sometimes there's an action required like calling the application layer or responding to a ping, and the thread discovering the action is not the thread that should do the work. We enqueue a runnable on this executor and it gets handled by one of the executor's threads.

### Locks

 * <p>
 * The context includes information such as the session ID, active thread count, access count, crawler status,
 * URL filter, rule manager, interval controller, robots.txt URL set, sitemaps, number of threads,
 * maximum thread check count, maximum depth, and maximum access count.
 * </p>
 *
 * <p>
 * It also provides thread-local storage for sitemaps, allowing each thread to have its own set of sitemaps.
 * </p>
 */