* En responses, el `set` se convertirá en una `list`.
        * El esquema generado especificará que los valores del `set` son únicos (usando `uniqueItems` de JSON Schema).
* `bytes`:
    * `bytes` estándar de Python.
    * En requests y responses se tratará como `str`.
    * El esquema generado especificará que es un `str` con "binary" como "format".
* `Decimal`:

        ResponseData data = new ResponseData();

        data.setMimeType("text/html");
        assertEquals("text/html", data.getMimeType());

        data.setMimeType("application/json");
        assertEquals("application/json", data.getMimeType());

        data.setMimeType(null);
        assertNull(data.getMimeType());
    }

    public void test_methodGetterSetter() {
        // Test method getter/setter

        "default configs should be preserved"
    )


def test_get_users():
    response = client.get("/users/foo")
    assert response.status_code == 200, response.text

            pass  # pragma: nocover

    client = TestClient(app)
    response = client.get("/multi-query", params={"foo": "5"})
    assert response.status_code == 200
    assert response.json() == 5

    response = client.get("/multi-query", params={"foo": "123"})
    assert response.status_code == 422

    response = client.get("/multi-query", params={"foo": "1"})

* the API docs
* the automatic docs

* Data Science
* Deep Learning
* Machine Learning
* Dependency Injection
* HTTP Basic authentication
* HTTP Digest
* ISO format
* the JSON Schema standard
* the JSON schema
* the schema definition
* Password Flow
* Mobile

* deprecated
* designed
* invalid
* on the fly
* standard
* default
* case-sensitive
* case-insensitive

                errors=exc.errors(include_url=False), loc_prefix=loc
            )

    def serialize(
        self,
        value: Any,
        *,
        mode: Literal["json", "python"] = "json",
        include: Union[IncEx, None] = None,
        exclude: Union[IncEx, None] = None,
        by_alias: bool = True,
        exclude_unset: bool = False,
        exclude_defaults: bool = False,

### 标签顺序

每个标签元数据字典的顺序也定义了在文档用户界面显示的顺序。

例如按照字母顺序，即使 `users` 排在 `items` 之后，它也会显示在前面，因为我们将它的元数据添加为列表内的第一个字典。

## OpenAPI URL

默认情况下，OpenAPI 模式服务于 `/openapi.json`。

但是你可以通过参数 `openapi_url` 对其进行配置。

例如，将其设置为服务于 `/api/v1/openapi.json`：

{* ../../docs_src/metadata/tutorial002.py hl[3] *}

如果你想完全禁用 OpenAPI 模式，可以将其设置为 `openapi_url=None`，这样也会禁用使用它的文档用户界面。

## 文档 URLs

你可以配置两个文档用户界面，包括：

여기 쿼리 매개변수 `needy`는 `str`형인 필수 쿼리 매개변수입니다.

브라우저에서 아래와 같은 URL을 연다면:

```
http://127.0.0.1:8000/items/foo-item
```

...필수 매개변수 `needy`를 넣지 않았기 때문에 아래와 같은 오류를 보게 됩니다:

```JSON
{
    "detail": [
        {
            "loc": [
                "query",
                "needy"
            ],
            "msg": "field required",
            "type": "value_error.missing"

              <resources>
                <resource>
                  <directory>../../guava</directory>
                  <includes>
                    <include>module.json</include>
                  </includes>
                  <filtering>true</filtering>
                </resource>
              </resources>
            </configuration>
          </execution>

* `username`: `Rick`
* `password`: `Portal Gun`
* `extra`: `Mr. Poopybutthole`

Ele receberá um retorno de erro informando-o que o campo `extra` não é permitido:

```json
{
    "detail": [
        {
            "type": "extra_forbidden",
            "loc": ["body", "extra"],
            "msg": "Extra inputs are not permitted",
            "input": "Mr. Poopybutthole"
        }