peer.acceptFrame() // HEADERS STREAM 3
    peer.acceptFrame() // DATA STREAM 3 "abcde"
    peer.acceptFrame() // HEADERS STREAM 3
    peer.play()

    // Play it back.
    val connection = connect(peer)
    val stream = connection.newStream(headerEntries("a", "android"), true)
    stream.enqueueTrailers(headersOf("foo", "bar"))
    val sink = stream.sink.buffer()
    sink.writeUtf8("abcdefghi")

Dies wäre bei großen oder unendlichen Streams noch wichtiger.

///

/// tip | Tipp

Anstatt eine `StreamingResponse` direkt zurückzugeben, sollten Sie wahrscheinlich dem Stil in [Daten streamen](./stream-data.md) folgen. Das ist wesentlich bequemer und behandelt den Abbruch im Hintergrund für Sie.

Wenn Sie JSON Lines streamen, folgen Sie dem Tutorial [JSON Lines streamen](../tutorial/stream-json-lines.md).

///

Esto sería aún más importante con streams grandes o infinitos.

///

/// tip | Consejo

En lugar de devolver un `StreamingResponse` directamente, probablemente deberías seguir el estilo en [Stream Data](./stream-data.md), es mucho más conveniente y maneja la cancelación por detrás de escena por ti.

Si estás transmitiendo JSON Lines, sigue el tutorial [Stream JSON Lines](../tutorial/stream-json-lines.md).

///

Isso seria ainda mais importante com streams grandes ou infinitos.

///

/// tip | Dica

Em vez de retornar uma `StreamingResponse` diretamente, você deveria provavelmente seguir o estilo em [Transmitir Dados](./stream-data.md), é muito mais conveniente e lida com cancelamento nos bastidores para você.

	if debugPrint {
		fmt.Println("RegisterStreamingHandler: subroute:", h.Subroute)
	}
	if h.Subroute == "" {
		if m.handlers.hasAny(id) && !id.isTestHandler() {
			return ErrHandlerAlreadyExists
		}
		m.handlers.streams[id] = &h
		return nil
	}
	subID := makeSubHandlerID(id, h.Subroute)
	if m.handlers.hasSubhandler(subID) && !id.isTestHandler() {
		return ErrHandlerAlreadyExists
	}
	m.handlers.subStreams[subID] = &h

- Text logging in Kubernetes components now uses [textlogger](https://pkg.go.dev/k8s.io/klog/v2@v2.120.0/textlogger). The same split streams of info and error log entries with buffering of info entries is now also supported for text output (off by default, alpha feature). Previously, this was only supported for JSON. Performance is better also without split streams. ([#114672](https://github.com/kubernetes/kubernetes/pull/114672), [@pohly](https://github.com/pohly))

				merged = append(merged, latest)

				// Calculate latest 'n' non-free versions.
				if !latest.header.FreeVersion() {
					nVersions++
				}
			}
		}

		// Remove from all streams up until latest modtime or if selected.
		for i, vers := range versions {
			for _, ver := range vers {
				// Truncate later modtimes, not selected.
				if ver.header.ModTime > latest.header.ModTime {

            return endOfData();
          }
        };
      }

      @Override
      public Stream<E> stream() {
        return Stream.concat(set1.stream(), set2.stream().filter((E e) -> !set1.contains(e)));
      }

      @Override
      public Stream<E> parallelStream() {
        return stream().parallel();
      }

      @Override
      public boolean contains(@Nullable Object object) {

        private void writeObject(ObjectOutputStream stream) throws IOException {
      stream.defaultWriteObject();
      stream.writeObject(factory);
      stream.writeObject(backingMap());
    }

    @GwtIncompatible
    @J2ktIncompatible
        @SuppressWarnings("unchecked") // reading data stored by writeObject
    private void readObject(ObjectInputStream stream) throws IOException, ClassNotFoundException {

* Get public IP for Azure vmss nodes. ([#68498](https://github.com/kubernetes/kubernetes/pull/68498), [@feiskyer](https://github.com/feiskyer))
* adjusted http/2 buffer sizes for apiservers to prevent starvation issues between concurrent streams ([#67902](https://github.com/kubernetes/kubernetes/pull/67902), [@liggitt](https://github.com/liggitt))