你可以使用**服务器发送事件**（SSE）向客户端流式发送数据。

这类似于[流式传输 JSON Lines](stream-json-lines.md)，但使用 `text/event-stream` 格式，浏览器原生通过 [`EventSource` API](https://developer.mozilla.org/en-US/docs/Web/API/EventSource) 支持。

/// info | 信息

新增于 FastAPI 0.135.0。

///

## 什么是服务器发送事件？ { #what-are-server-sent-events }

SSE 是一种通过 HTTP 从服务器向客户端流式传输数据的标准。

每个事件是一个带有 `data`、`event`、`id` 和 `retry` 等“字段”的小文本块，以空行分隔。

看起来像这样：

```

		Hidden: true,
	},
	cli.StringFlag{
		Name:  "certs-dir, S",
		Value: defaultCertsDir.Get(),
		Usage: "path to certs directory",
	},
	cli.BoolFlag{
		Name:  "quiet",
		Usage: "disable startup and info messages",
	},
	cli.BoolFlag{
		Name:  "anonymous",
		Usage: "hide sensitive information from logging",
	},
	cli.BoolFlag{
		Name:  "json",
		Usage: "output logs in JSON format",
	},

* `users:read` o `users:write` son ejemplos comunes.
* `instagram_basic` es usado por Facebook / Instagram.
* `https://www.googleapis.com/auth/drive` es usado por Google.

/// info | Información

En OAuth2 un "scope" es solo un string que declara un permiso específico requerido.

No importa si tiene otros caracteres como `:` o si es una URL.

Esos detalles son específicos de la implementación.

            "name": "The Baz"
        }
    ]
}
```

/// info | 情報

`images`キーが画像オブジェクトのリストを持つようになったことに注目してください。

///

## 深くネストされたモデル { #deeply-nested-models }

深くネストされた任意のモデルを定義することができます:

{* ../../docs_src/body_nested_models/tutorial007_py310.py hl[7,12,18,21,25] *}

/// info | 情報

`Offer`は`Item`のリストであり、それらがさらにオプションの`Image`のリストを持っていることに注目してください。

///

Bunu yapmak için önce şunları import edin:

* `fastapi` içinden `Query`
* `typing` içinden `Annotated`

{* ../../docs_src/query_params_str_validations/tutorial002_an_py310.py hl[1,3] *}

/// info | Bilgi

FastAPI, 0.95.0 sürümünde `Annotated` desteğini ekledi (ve önermeye başladı).

Daha eski bir sürüm kullanıyorsanız `Annotated` kullanmaya çalışırken hata alırsınız.

		// we are only interested in b.ObjectLockConfigXML or objectLockConfig value
	}

	for _, legacyFile := range legacyConfigs {
		configFile := path.Join(bucketMetaPrefix, b.Name, legacyFile)

		configData, info, err := readConfigWithMetadata(ctx, objectAPI, configFile, ObjectOptions{})
		if err != nil {
			if _, ok := err.(ObjectExistsAsDirectory); ok {
				// in FS mode it possible that we have actual

  <id>lifecycle</id>
  <name>Lifecycle</name>
  <description>
    Configuration of custom lifecycle mappings for the plugin, as generally stored in
    {@code META-INF/maven/lifecycle.xml} in a plugin's jar artifact.
  </description>
  <classes>
    <class rootElement="true" xml.tagName="lifecycles" xsd.compositor="sequence">
      <name>LifecycleConfiguration</name>

 {{{./apidocs/org/apache/maven/cli/logging/Slf4jConfigurationFactory.html}Slf4jConfigurationFactory}} /
 {{{./apidocs/org/apache/maven/cli/logging/Slf4jConfiguration.html}Slf4jConfiguration}}.

~~ TODO document META-INF/maven/slf4j-configuration.properties

* Getting Logger Instance

 Starting with Maven 3.1.0, SLF4J Logger can be used directly. This technique can be used safely in Maven core

 * @deprecated since 4.0.0, use {@code maven-api-impl} jar instead
 */
@Deprecated(since = "4.0.0")
class PluginsMetadataGenerator implements MetadataGenerator {
    private static final String PLUGIN_DESCRIPTOR_LOCATION = "META-INF/maven/plugin.xml";

    private final Logger logger = LoggerFactory.getLogger(getClass());

    private final Map<Object, PluginsMetadata> processedPlugins;

    private final Date timestamp;

        try {
            final Bucket existingBucket = storage.get(bucket);
            if (existingBucket == null) {
                storage.create(BucketInfo.newBuilder(bucket).build());
                logger.info("Created storage bucket: {}", bucket);
            }
        } catch (final Exception e) {
            logger.warn("Failed to ensure bucket exists: {}", bucket, e);
        }
    }

    @Override