///

## Cookies com Modelos Pydantic { #cookies-with-a-pydantic-model }

Declare os parâmetros de **cookie** de que você precisa em um **modelo Pydantic**, e depois declare o parâmetro como `Cookie`:

{* ../../docs_src/cookie_param_models/tutorial001_an_py310.py hl[9:12,16] *}

   *
   * @return this method does not return; it always throws
   * @throws IOException when the given throwable is an IOException
   * @throws X1 when the given throwable is of the declared type X1
   * @throws X2 when the given throwable is of the declared type X2
   */
  public <X1 extends Exception, X2 extends Exception> RuntimeException rethrow(
      Throwable e, Class<X1> declaredType1, Class<X2> declaredType2) throws IOException, X1, X2 {

Por ejemplo, ampliando el modelo anterior, podrías decidir que deseas tener otra clave `importance` en el mismo cuerpo, además de `item` y `user`.

Si lo declaras tal cual, debido a que es un valor singular, **FastAPI** asumirá que es un parámetro de query.

Pero puedes instruir a **FastAPI** para que lo trate como otra clave del cuerpo usando `Body`:

You can declare a parameter of type `Response` in your *path operation function*.

And then you can set cookies in that *temporal* response object.

{* ../../docs_src/response_cookies/tutorial002_py310.py hl[1, 8:9] *}

And then you can return any object you need, as you normally would (a `dict`, a database model, etc).

We can do better...

## What makes a dependency { #what-makes-a-dependency }

Up to now you have seen dependencies declared as functions.

But that's not the only way to declare dependencies (although it would probably be the more common).

The key factor is that a dependency should be a "callable".

            the dependency is declared again for the rest of the request (for example
            if the dependency is needed by several dependencies), the value will be
            re-used for the rest of the request.

            Set `use_cache` to `False` to disable this behavior and ensure the
            dependency is called again (if declared more than once) in the same request.

You can declare a parameter of type `Response` in your *path operation function* (as you can do for cookies).

And then you can set headers in that *temporal* response object.

{* ../../docs_src/response_headers/tutorial002_py310.py hl[1, 7:8] *}

And then you can return any object you need, as you normally would (a `dict`, a database model, etc).

Et si vous avez déclaré un `response_model`, il sera toujours utilisé pour filtrer et convertir l'objet que vous avez renvoyé.

**FastAPI** utilisera cette réponse temporaire pour extraire les en-têtes (ainsi que les cookies et le code de statut), et les placera dans la réponse finale qui contient la valeur que vous avez renvoyée, filtrée par tout `response_model`.

```JSON
{
    "name": "Foo",
    "price": 45.2
}
```

## Declare como um parâmetro { #declare-it-as-a-parameter }

Para adicioná-lo à sua *operação de rota*, declare-o da mesma maneira que você declarou parâmetros de rota e de consulta:

{* ../../docs_src/body/tutorial001_py310.py hl[16] *}

...e declare o seu tipo como o modelo que você criou, `Item`.

## Resultados { #results }

Fíjate que `Field` se importa directamente desde `pydantic`, no desde `fastapi` como el resto (`Query`, `Path`, `Body`, etc).

///

## Declarar atributos del modelo { #declare-model-attributes }

Después puedes utilizar `Field` con los atributos del modelo:

{* ../../docs_src/body_fields/tutorial001_an_py310.py hl[11:14] *}

`Field` funciona de la misma manera que `Query`, `Path` y `Body`, tiene todos los mismos parámetros, etc.