* {@link #writeTo} and {@link #readFrom} methods. Both serialized forms will continue to be
 * supported by future versions of this library. However, serial forms generated by newer versions
 * of the code may not be readable by older versions of the code (e.g., a serialized Bloom filter
 * generated today may <i>not</i> be readable by a binary that was compiled 6 months ago).
 *

///

### Read Heroes with `HeroPublic` { #read-heroes-with-heropublic }

We can do the same as before to **read** `Hero`s, again, we use `response_model=list[HeroPublic]` to ensure that the data is validated and serialized correctly.

{* ../../docs_src/sql_databases/tutorial002_an_py310.py ln[65:72] hl[65] *}

### Read One Hero with `HeroPublic` { #read-one-hero-with-heropublic }

We can **read** a single hero:

Bir [Response Model](../tutorial/response-model.md) deklare ederseniz, FastAPI veriyi Pydantic kullanarak JSON'a serialize etmek için bunu kullanır.

Bir response model deklare etmezseniz, FastAPI [JSON Uyumlu Encoder](../tutorial/encoder.md)'da anlatılan `jsonable_encoder`'ı kullanır ve bunu bir `JSONResponse` içine koyar.

    For example, `data="hello"` produces `data: "hello"` on the wire (with
    quotes).
    """

    data: Annotated[
        Any,
        Doc(
            """
            The event payload.

            Can be any JSON-serializable value: a Pydantic model, dict, list,
            string, number, etc. It is **always** serialized to JSON: strings

### JSON Performansı { #json-performance }

        start = System.currentTimeMillis();
        SerializeUtil.serialize(hmap);
        System.out.println("HashMap serialize:" + (System.currentTimeMillis() - start));

        start = System.currentTimeMillis();
        SerializeUtil.serialize(amap);
        System.out.println("ArrayMap serialize:" + (System.currentTimeMillis() - start));
    }

    private static class MyKey {

// Sets `value` to the binary-serialized TensorShapeProto of the value of
// `attr_name` attribute of `oper`.
TF_CAPI_EXPORT extern void TF_OperationGetAttrTensorShapeProto(
    TF_Operation* oper, const char* attr_name, TF_Buffer* value,
    TF_Status* status);

// Fills in `values` with binary-serialized TensorShapeProto values of the

        assertTrue(s.contains("hello"));
    }

    @Test
    @DisplayName("Should serialize and deserialize preserving message and cause")
    void testJavaSerialization() throws Exception {
        // Given
        SpnegoException original = new SpnegoException("serialize me", new IllegalArgumentException("iaex"));

        // When
        byte[] bytes;

Com um modelo de resposta, o FastAPI usará o Pydantic para serializar os dados para JSON, sem passos intermediários, como convertê-los com `jsonable_encoder`, o que aconteceria em qualquer outro caso.

E, por baixo dos panos, o Pydantic usa os mesmos mecanismos em Rust que o `orjson` para serializar para JSON, então você já terá a melhor performance com um modelo de resposta.

			if err != nil {
				err = msgp.WrapError(err)
				return
			}
		}
	}
	o = bts
	return
}

// Msgsize returns an upper bound estimate of the number of bytes occupied by the serialized message
func (z *allTierStats) Msgsize() (s int) {
	s = 1 + 3 + msgp.MapHeaderSize
	if z.Tiers != nil {
		for za0001, za0002 := range z.Tiers {
			_ = za0002