public class EnumsBenchmark {

  @Param({"Small", "Medium", "Large"})
  String enumSize;

  @Param({"0.2", "0.8"})
  float hitRate;

  // We could avoid the raw type here by initializing this with a ternary (? SmallEnum.class : ...).
  // However, we end up needing a raw type in getIfPresent, as discussed there.
  @SuppressWarnings("rawtypes")
  private Class<? extends Enum> enumType;

  private String[] sampleData;

    val value = headers.value(i)

    when {
      name.equals("Cache-Control", ignoreCase = true) -> {
        if (headerValue != null) {
          // Multiple cache-control headers means we can't use the raw value.
          canUseHeaderValue = false
        } else {
          headerValue = value
        }
      }

      name.equals("Pragma", ignoreCase = true) -> {

{* ../../docs_src/server_sent_events/tutorial002_py310.py hl[4,26] *}

Le champ `data` est toujours encodé en JSON. Vous pouvez passer toute valeur sérialisable en JSON, y compris des modèles Pydantic.

## Données brutes { #raw-data }

Si vous devez envoyer des données **sans** encodage JSON, utilisez `raw_data` au lieu de `data`.

	for i := range listener.listeners {
		addrs = append(addrs, listener.listeners[i].Addr())
	}

	return addrs
}

// TCPOptions specify customizable TCP optimizations on raw socket
type TCPOptions struct {
	UserTimeout int // this value is expected to be in milliseconds

	// When the net.Conn is a remote drive this value is honored, we close the connection to remote peer proactively.

    ArtifactResolutionResult resolve(ArtifactResolutionRequest request);

    // Install

    // Deploy

    // Map types of artifacts

    //
    // Raw file transfers
    //
    void publish(
            ArtifactRepository repository, File source, String remotePath, ArtifactTransferListener transferListener)
            throws ArtifactTransferFailedException;

从 `fastapi.sse` 导入 `ServerSentEvent`：

{* ../../docs_src/server_sent_events/tutorial002_py310.py hl[4,26] *}

`data` 字段始终会被编码为 JSON。你可以传入任何可被序列化为 JSON 的值，包括 Pydantic 模型。

## 原始数据 { #raw-data }

如果你需要发送**不**进行 JSON 编码的数据，请使用 `raw_data` 而不是 `data`。

这对于发送预格式化文本、日志行或特殊的 <dfn title="用于指示特殊条件或状态的值">"哨兵"</dfn> 值（例如 `[DONE]`）很有用。

{* ../../docs_src/server_sent_events/tutorial003_py310.py hl[17] *}

    /** 32-bit status codes flag */
    int FLAGS2_STATUS32 = 0x4000;
    /** Unicode strings flag */
    int FLAGS2_UNICODE = 0x8000;

    /** No capabilities */
    int CAP_NONE = 0x0000;
    /** Raw mode capability */
    int CAP_RAW_MODE = 0x0001;
    /** MPX mode capability */
    int CAP_MPX_MODE = 0x0002;
    /** Unicode capability */
    int CAP_UNICODE = 0x0004;
    /** Large files capability */

        // phase 1
        DefaultModelBuildingResult result = new DefaultModelBuildingResult();

        DefaultModelProblemCollector problems = new DefaultModelProblemCollector(result);

        // read and validate raw model
        Model inputModel = request.getRawModel();
        if (inputModel == null) {
            inputModel = readModel(request.getModelSource(), request.getPomFile(), request, problems);
        }

Write throughput is typically at least 500MB/s per CPU core,
and scales with the number of available CPU cores.
Decompression speed is typically at least 1GB/s.

This means that in cases where raw IO is below these numbers
compression will not only reduce disk usage but also help increase system throughput.
Typically, enabling compression on spinning disk systems
will increase speed when the content can be compressed.

func (e *SelectStatement) EvalFrom(format string, input Record) ([]*Record, error) {
	if !e.selectAST.From.HasKeypath() {
		return []*Record{&input}, nil
	}
	_, rawVal := input.Raw()

	if format != "json" {
		return nil, errDataSource(errors.New("path not supported"))
	}
	switch rec := rawVal.(type) {
	case jstream.KVS: