Comencemos con un ejemplo y luego veámoslo en detalle.

Creamos una función asíncrona `lifespan()` con `yield` así:

{* ../../docs_src/events/tutorial003_py310.py hl[16,19] *}

### Creando el `Settings` solo una vez con `lru_cache` { #creating-the-settings-only-once-with-lru-cache }

Leer un archivo desde el disco es normalmente una operación costosa (lenta), por lo que probablemente quieras hacerlo solo una vez y luego reutilizar el mismo objeto de configuraciones, en lugar de leerlo para cada request.

Pero cada vez que hacemos:

```Python
Settings()
```

Nós criamos uma função assíncrona `lifespan()` com `yield` assim:

{* ../../docs_src/events/tutorial003_py310.py hl[16,19] *}

Aqui estamos simulando a operação de *inicialização* custosa de carregar o modelo, colocando a (falsa) função do modelo no dicionário com modelos de Aprendizado de Máquina antes do `yield`. Esse código será executado **antes** de a aplicação **começar a receber requisições**, durante a *inicialização*.

      description: "How many tools (such as vet, gopls, gofmt, goimports, etc.) would be affected? "
    validations:
      required: false

  - type: input
    id: perf-costs
    attributes:
      label: Performance Costs
      description: "What is the compile time cost? What is the run time cost? "
    validations:
      required: false

  - type: textarea
    id: prototype
    attributes:

if [[ "$TFCI_DOCKER_PULL_ENABLE" == 1 ]]; then
  # Simple retry logic for docker-pull errors. Sleeps if a pull fails.
  # Pulling an already-pulled container image will finish instantly, so
  # repeating the command costs nothing.
  docker pull "$TFCI_DOCKER_IMAGE" || sleep 15
  docker pull "$TFCI_DOCKER_IMAGE" || sleep 30
  docker pull "$TFCI_DOCKER_IMAGE" || sleep 60
  docker pull "$TFCI_DOCKER_IMAGE"
fi 

import java.util.stream.Collectors;

public class ParallelDetector {

    private static Integer _defaultParallel = null;

    public static int findDefaultParallel(Project project) {
        // Since it costs IO to compute this, and is done at configuration time we want to cache this if possible
        // It's safe to store this in a static variable since it's just a primitive so leaking memory isn't an issue

the Program in a commercial product offering, such Contributor
("Commercial Contributor") hereby agrees to defend and indemnify every
other Contributor ("Indemnified Contributor") against any losses,
damages and costs (collectively "Losses") arising from claims, lawsuits
and other legal actions brought by a third party against the Indemnified
Contributor to the extent caused by the acts or omissions of such

### Criando o `Settings` apenas uma vez com `lru_cache` { #creating-the-settings-only-once-with-lru-cache }

Ler um arquivo do disco normalmente é uma operação custosa (lenta), então você provavelmente vai querer fazer isso apenas uma vez e depois reutilizar o mesmo objeto de configurações, em vez de lê-lo a cada requisição.

Mas toda vez que fizermos:

```Python
Settings()
```

such Contributor ("Commercial Contributor") hereby agrees to defend and indemnify
every other Contributor ("Indemnified Contributor") against any losses, damages
and costs (collectively "Losses") arising from claims, lawsuits and other
legal actions brought by a third party against the Indemnified Contributor
to the extent caused by the acts or omissions of such Commercial Contributor

 * could in principle cause problems for some users. Still, we expect that the benefits of the
 * nullness annotations in particular will outweigh the costs. (And it's worth noting that we have
 * released multiple ListenableFuture.class files that are not byte-for-byte compatible even from
 * the beginning, thanks to using different `-source -target` values for compiling our `-jre` and