`user_in` ist ein Pydantic-Modell der Klasse `UserIn`.

Pydantic-Modelle haben eine `.model_dump()`-Methode, die ein <abbr title="Dictionary – Zuordnungstabelle: In anderen Sprachen auch Hash, Map, Objekt, Assoziatives Array genannt">`dict`</abbr> mit den Daten des Modells zurückgibt.

Wenn wir also ein Pydantic-Objekt `user_in` erstellen, etwa so:

```Python

       * UnsignedBytesBenchmark. TODO(cpovirk): benchmark this
       */
      /*
       * A CopyOnWriteArraySet<WeakReference> is faster than a newSetFromMap of a MapMaker map with
       * weakKeys() and concurrencyLevel(1), even up to at least 12 cached exception types.
       */
      private static final Set<WeakReference<Class<? extends Exception>>> validClasses =

            java.lang.reflect.Method[] methods = SmbSession.class.getDeclaredMethods();
            java.util.Set<String> methodNames =
                    java.util.Arrays.stream(methods).map(java.lang.reflect.Method::getName).collect(java.util.stream.Collectors.toSet());

            assertTrue(methodNames.contains("close"), "Should contain close method");

            return result.getArtifactResults().stream()
                    .filter(ArtifactResult::isResolved)
                    .map(ArtifactResult::getArtifact)
                    .collect(Collectors.toList());
        } catch (PluginResolutionException | InterpolatorException e) {
            throw new ExtensionResolutionException(extension, e);
        }

Hier simulieren wir den langsamen *Startup*, das Laden des Modells, indem wir die (Fake-)Modellfunktion vor dem `yield` in das <abbr title="Dictionary – Zuordnungstabelle: In anderen Sprachen auch Hash, Map, Objekt, Assoziatives Array genannt">Dictionary</abbr> mit Modellen für maschinelles Lernen einfügen. Dieser Code wird ausgeführt, **bevor** die Anwendung **beginnt, Requests entgegenzunehmen**, während des *Startups*.

	uuids := make([]uuid.UUID, 5)
	for i := range uuids {
		uuids[i] = uuid.UUID([16]byte{15: uint8(i + 1)})
	}
	fivs := make([]FileInfo, 5)
	objInfos := make([]ObjectInfo, 5)
	objRetentionMeta := make(map[string]string)
	objRetentionMeta[strings.ToLower(xhttp.AmzObjectLockMode)] = string(objectlock.RetCompliance)
	// Set retain until date 12 hours into the future

	EnvKafkaProducerCompressionCodec = "MINIO_NOTIFY_KAFKA_PRODUCER_COMPRESSION_CODEC"
	EnvKafkaProducerCompressionLevel = "MINIO_NOTIFY_KAFKA_PRODUCER_COMPRESSION_LEVEL"
)

var codecs = map[string]sarama.CompressionCodec{
	"none":   sarama.CompressionNone,
	"gzip":   sarama.CompressionGZIP,
	"snappy": sarama.CompressionSnappy,
	"lz4":    sarama.CompressionLZ4,
	"zstd":   sarama.CompressionZSTD,
}

  public static OptionalInt findLast(IntStream stream) {
    // findLast(Stream) does some allocation, so we might as well box some more
    java.util.Optional<Integer> boxedLast = findLast(stream.boxed());
    return boxedLast.map(OptionalInt::of).orElse(OptionalInt.empty());
  }

  /**
   * Returns the last element of the specified stream, or {@link OptionalLong#empty} if the stream
   * is empty.
   *

  fun check(
    hostname: String,
    peerCertificates: List<Certificate>,
  ) = check(hostname) {
    (certificateChainCleaner?.clean(peerCertificates, hostname) ?: peerCertificates)
      .map { it as X509Certificate }
  }

  internal fun check(
    hostname: String,
    cleanedPeerCertificatesFn: () -> List<X509Certificate>,
  ) {
    val pins = findMatchingPins(hostname)

		return strings.Compare(results[i].Name, results[j].Name) <= -1
	})

	if len(results) != 2 || results[0].Name != user2.Name || results[1].Name != user3.Name {
		t.Errorf("Scan into struct map, got %#v", results)
	}

	type ID uint64
	var id ID
	DB.Raw("select id from users where id = ?", user2.ID).Scan(&id)
	if uint(id) != user2.ID {
		t.Errorf("Failed to scan to customized data type")
	}