Si hubiéramos declarado `-> HeroPublic`, tu editor y linter se quejarían (con razón) de que estás devolviendo un `Hero` en lugar de un `HeroPublic`.

Al declararlo en `response_model` le estamos diciendo a **FastAPI** que haga lo suyo, sin interferir con las anotaciones de tipo y la ayuda de tu editor y otras herramientas.

///

### Leer Heroes con `HeroPublic` { #read-heroes-with-heropublic }

你的模型的 JSON Schema 会成为生成的 OpenAPI Schema 的一部分，并显示在交互式 API 文档中：

<img src="/img/tutorial/body/image01.png">

并且，还会用于需要它们的每个*路径操作*的 API 文档中：

<img src="/img/tutorial/body/image02.png">

## 编辑器支持 { #editor-support }

在编辑器中，函数内部你会在各处得到类型提示与补全（如果接收的不是 Pydantic 模型，而是 `dict`，就不会有这样的支持）：

<img src="/img/tutorial/body/image03.png">

还支持检查错误的类型操作：

<img src="/img/tutorial/body/image04.png">

这并非偶然，整个框架都是围绕这种设计构建的。

...desde:

```Python
        ... "item_name": item.name ...
```

...a:

```Python
        ... "item_price": item.price ...
```

...y observa cómo tu editor autocompleta los atributos y conoce sus tipos:

![editor support](https://fastapi.tiangolo.com/img/vscode-completion.png)

Depois, você poderia adicionar permissões sobre essa entidade, como "dirigir" (para o carro) ou "editar" (para o blog).

E então, poderia dar esse token JWT para um usuário (ou bot), e ele poderia usá-lo para realizar essas ações (dirigir o carro ou editar o blog) sem sequer precisar ter uma conta, apenas com o token JWT que sua API gerou para isso.

     * not be set within all security descriptors even though the ACE was in
     * face inherited. If an inherited ACE is added to a parent the Windows
     * ACL editor will rebuild all children ACEs and set this flag accordingly.
     * @return true if this ACE is inherited, false if it is direct
     */
    public boolean isInherited() {
        return (flags & FLAGS_INHERITED) != 0;

 *
 * <p>This class recursively discovers all POM files in a Maven project hierarchy
 * and loads them as domtrip Documents. Individual strategies can create domtrip Editors
 * from these Documents as needed for superior formatting preservation.
 */
public class PomDiscovery {

    private PomDiscovery() {
        // noop
    }

    /**

For example, you might want to try:

* [Speakeasy](https://speakeasy.com/editor?utm_source=fastapi+repo&utm_medium=github+sponsorship)
* [Stainless](https://www.stainless.com/?utm_source=fastapi&utm_medium=referral)
* [liblab](https://developers.liblab.com/tutorials/sdk-for-fastapi?utm_source=fastapi)

{* ../../docs_src/response_status_code/tutorial002_py310.py hl[1,6] *}

Diese sind nur eine Annehmlichkeit, sie enthalten dieselbe Zahl, aber so können Sie die Autovervollständigung Ihres Editors verwenden, um sie zu finden:

<img src="/img/tutorial/response-status-code/image02.png">

/// note | Technische Details

Sie könnten auch `from starlette import status` verwenden.

    - advanced/generate-clients.md
    - advanced/advanced-python-types.md
    - advanced/json-base64-bytes.md
    - advanced/strict-content-type.md
    - advanced/vibe.md
  - fastapi-cli.md
  - editor-support.md
  - Deployment:
    - deployment/index.md
    - deployment/versions.md
    - deployment/fastapicloud.md
    - deployment/https.md
    - deployment/manually.md
    - deployment/concepts.md

<img src="/img/tutorial/body/image01.png">

Et seront aussi utilisés dans chaque *chemin d'accès* de la documentation utilisant ces modèles :

<img src="/img/tutorial/body/image02.png">

## Support de l'éditeur { #editor-support }

Dans votre éditeur, vous aurez des annotations de type et de l'autocomplétion partout dans votre fonction (ce qui n'aurait pas été le cas si vous aviez reçu un `dict` plutôt qu'un modèle Pydantic) :