## Pydantic `bytes` { #pydantic-bytes }

Você pode declarar um modelo Pydantic com campos `bytes` e então usar `val_json_bytes` na configuração do modelo para indicar que deve usar base64 para *validar* os dados JSON de entrada; como parte dessa validação, ele decodificará a string base64 em bytes.

{* ../../docs_src/json_base64_bytes/tutorial001_py310.py ln[1:9,29:35] hl[9] *}

    echo \"Failed attempts: $ATTEMPTS\" ;
    if [ $ATTEMPTS -gt $LIMIT ]; then
      exit 1 ;
    fi ;
    sleep 2 ; # 2 second intervals between attempts
    $MC_COMMAND ;
    STATUS=$? ;
  done ;
  set -e ; # reset `e` as active
  return 0
}

# checkSvcacctExists ()
# Check if the svcacct exists, by using the exit code of `mc admin user svcacct info`
checkSvcacctExists() {

catch() {
	if [ $# -ne 0 ]; then
		echo "error on line $1"
		for site in sitea siteb; do
			echo "$site server logs ========="
			cat "/tmp/${site}_1.log"
			echo "==========================="
			cat "/tmp/${site}_2.log"
		done
	fi

	echo "Cleaning up instances of MinIO"
	pkill minio
	pkill -9 minio
	rm -rf /tmp/multisitea
	rm -rf /tmp/multisiteb
	rm -rf /tmp/data
	if [ $# -ne 0 ]; then
		exit $#
	fi
}

catch

        final PrintWriter writer = new PrintWriter(stringWriter);

        chatApiManager.sendSseEvent(writer, "done", Map.of("sessionId", "session-123"));

        final String output = stringWriter.toString();
        assertTrue(output.contains("event: done"));
        assertTrue(output.contains("\"sessionId\":\"session-123\""));
    }

    @Test

    val existing = tags[type]
    if (existing != null) return existing

    if (computed == null) {
      computed = compute()
    }

    // If we successfully add the computed element, we're done.
    val newTags = tags.plus(type, computed)
    if (compareAndSet(tags, newTags)) return computed

    // We lost the race. Possibly to other code that was putting a *different* key. Try again!
  }

Você também define de alguma maneira em quais **momentos** a sua aplicação mandará essas requisições ou eventos.

E os **seus usuários** definem de alguma forma (em algum painel por exemplo) a **URL** que a sua aplicação deve enviar essas requisições.

Toda a **lógica** sobre como cadastrar as URLs para os webhooks e o código para enviar de fato as requisições cabe a você definir. Você escreve da maneira que você desejar no **seu próprio código**.

   */
  void moveLastEntry(int dstIndex) {
    int srcIndex = size() - 1;
    if (dstIndex < srcIndex) {
      // move last entry to deleted spot
      keys[dstIndex] = keys[srcIndex];
      values[dstIndex] = values[srcIndex];
      keys[srcIndex] = null;
      values[srcIndex] = 0;

      // move the last entry to the removed spot, just like we moved the element
      long lastEntry = entries[srcIndex];

    int NT_STATUS_PASSWORD_EXPIRED = 0xC0000071;
    /** The referenced account is currently disabled */
    int NT_STATUS_ACCOUNT_DISABLED = 0xC0000072;
    /** No mapping between account names and security IDs was done */
    int NT_STATUS_NONE_MAPPED = 0xC0000073;
    /** The security ID structure is invalid */
    int NT_STATUS_INVALID_SID = 0xC0000078;
    /** The disk is full */
    int NT_STATUS_DISK_FULL = 0xC000007f;

Поэтому, если мы создадим Pydantic-объект `user_in` таким способом:

```Python
user_in = UserIn(username="john", password="secret", email="john.doe@example.com")
```

и затем вызовем:

```Python
user_dict = user_in.model_dump()
```

то теперь у нас есть `dict` с данными в переменной `user_dict` (это `dict` вместо объекта Pydantic-модели).

    List<Integer> list = new ArrayList<>(set);
    inputOrder.arrange(list);
    input = ImmutableList.copyOf(list);
  }

  @Benchmark
  int collections(int reps) {
    int dummy = 0;
    // Yes, this could be done more elegantly
    if (mutable) {
      for (int i = 0; i < reps; i++) {
        List<Integer> copy = new ArrayList<>(input);
        sort(copy);
        dummy += copy.get(0);
      }
    } else {