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

```Python hl_lines="1  7-8"
{!../../docs_src/response_headers/tutorial002.py!}
```

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

And if you declared a `response_model`, it will still be used to filter and convert the object you returned.

使用 `startup` 事件声明 `app` 启动前运行的函数：

```Python hl_lines="8"
{!../../docs_src/events/tutorial001.py!}
```

本例中，`startup` 事件处理器函数为项目数据库（只是**字典**）提供了一些初始值。

**FastAPI** 支持多个事件处理器函数。

只有所有 `startup` 事件处理器运行完毕，**FastAPI** 应用才开始接收请求。

## `shutdown` 事件

使用 `shutdown` 事件声明 `app` 关闭时运行的函数：

```Python hl_lines="6"
{!../../docs_src/events/tutorial002.py!}
```

此处，`shutdown` 事件处理器函数在 `log.txt` 中写入一行文本 `Application shutdown`。

```Python hl_lines="8"
{!../../docs_src/path_params_numeric_validations/tutorial004.py!}
```

## Validações numéricas: maior que e menor que ou igual

O mesmo se aplica para:

* `gt`: maior que (`g`reater `t`han)
* `le`: menor que ou igual (`l`ess than or `e`qual)

```Python hl_lines="9"
{!../../docs_src/path_params_numeric_validations/tutorial005.py!}
```

## Paramètres optionnels

De la même façon, vous pouvez définir des paramètres de requête comme optionnels, en leur donnant comme valeur par défaut `None` :

{* ../../docs_src/query_params/tutorial002.py hl[9] *}

Ici, le paramètre `q` sera optionnel, et aura `None` comme valeur par défaut.

/// check | "Remarque"

import os
import shutil

from dirty_equals import IsDict
from fastapi.testclient import TestClient

from docs_src.additional_responses.tutorial002 import app

client = TestClient(app)


def test_path_operation():
    response = client.get("/items/foo")
    assert response.status_code == 200, response.text
    assert response.json() == {"id": "foo", "value": "there goes my hero"}


def test_path_operation_img():

import pytest
from fastapi.testclient import TestClient

from ...utils import needs_pydanticv2


@pytest.fixture(name="client")
def get_client() -> TestClient:
    from docs_src.separate_openapi_schemas.tutorial002 import app

    client = TestClient(app)
    return client


def test_create_item(client: TestClient) -> None:
    response = client.post("/items/", json={"name": "Foo"})
    assert response.status_code == 200, response.text

```Python hl_lines="2  6"
{!../../docs_src/advanced_middleware/tutorial001.py!}
```

## `TrustedHostMiddleware`

Erzwingt, dass alle eingehenden Requests einen korrekt gesetzten `Host`-Header haben, um sich vor HTTP-Host-Header-Angriffen zu schützen.

```Python hl_lines="2  6-8"
{!../../docs_src/advanced_middleware/tutorial002.py!}
```

Die folgenden Argumente werden unterstützt:

✋️ 🚥 👆 ✔️ 📚 🎻 🤥 🤭, 👉 👌 🎱 ⚙️ 👫 🏋️ 🕸 🛠️ ⚙️ FastAPI. 👶

///

## 🎻 `response_model`

👆 💪 ⚙️ `dataclasses` `response_model` 🔢:

```Python hl_lines="1  7-13  19"
{!../../docs_src/dataclasses/tutorial002.py!}
```

🎻 🔜 🔁 🗜 Pydantic 🎻.

👉 🌌, 🚮 🔗 🔜 🎦 🆙 🛠️ 🩺 👩‍💻 🔢:

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

## 🎻 🔁 📊 📊

👆 💪 🌀 `dataclasses` ⏮️ 🎏 🆎 ✍ ⚒ 🐦 📊 📊.

Você pode usar isso para adicionar um `example` para cada campo:

```Python hl_lines="4  10-13"
{!../../docs_src/schema_extra_example/tutorial002.py!}
```

/// warning | "Atenção"

Lembre-se de que esses argumentos extras passados ​​não adicionarão nenhuma validação, apenas informações extras, para fins de documentação.

///

因此，开发时仍需要使用 Pydantic 模型。

但如果数据类很多，这一技巧能给 FastAPI 开发 Web API 增添不少助力。🤓

///

## `response_model` 使用数据类

在 `response_model` 参数中使用 `dataclasses`：

```Python hl_lines="1  7-13  19"
{!../../docs_src/dataclasses/tutorial002.py!}
```

本例把数据类自动转换为 Pydantic 数据类。

API 文档中也会显示相关概图：

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

## 在嵌套数据结构中使用数据类

您还可以把 `dataclasses` 与其它类型注解组合在一起，创建嵌套数据结构。