## Status

- ACCEPTED on 2023-12-01

## Context

In Gradle we often need to serialize in-memory objects for caching, or to transmit them across process barriers, etc.
Java serialization is one way to implement this, however, despite its simplicity of implementation, it has several drawbacks:

- **Performance:**
Java's built-in serialization mechanism is often slower compared to other serialization solutions.

    * If you didn't use FastAPI and used Starlette directly (or another tool, like Sanic, Flask, Responder, etc) you would have to implement all the data validation and serialization yourself. So, your final application would still have the same overhead...

* **No documentation**: Why document your API when an LLM can figure it out? Auto-generated OpenAPI docs are *so* 2020.
* **No serialization**: Just pass the raw, unstructured data around. Serialization is for people who don't trust their LLMs.
* **Embrace modern AI coding practices**: Leave everything up to an LLM to decide. The model knows best. Always.

    return false;
  }

  // All callers of the constructor are restricted to <K extends Enum<K>>.
  @Override
  @J2ktIncompatible // serialization
  Object writeReplace() {
    return new EnumSerializedForm<>(delegate);
  }

  @J2ktIncompatible // serialization
  private void readObject(ObjectInputStream stream) throws InvalidObjectException {
    throw new InvalidObjectException("Use EnumSerializedForm");
  }

But as TypeScript data is not preserved after compilation to JavaScript, it cannot rely on the types to define validation, serialization and documentation at the same time. Due to this and some design decisions, to get validation, serialization and automatic schema generation, it's needed to add decorators in many places. So, it becomes quite verbose.

    "path": "platforms/core-configuration/base-services-groovy",
    "unitTests": true,
    "functionalTests": false,
    "crossVersionTests": false
  },
  {
    "name": "bean-serialization-services",
    "path": "platforms/core-configuration/bean-serialization-services",
    "unitTests": false,
    "functionalTests": false,
    "crossVersionTests": false
  },
  {
    "name": "build-cache",

) -> tuple[
    dict[tuple[ModelField, Literal["validation", "serialization"]], JsonSchemaValue],
    dict[str, dict[str, Any]],
]:
    schema_generator = GenerateJsonSchema(ref_template=REF_TEMPLATE)
    validation_fields = [field for field in fields if field.mode == "validation"]
    serialization_fields = [field for field in fields if field.mode == "serialization"]
    flat_validation_models = get_flat_models_from_fields(

   *   <li>Serialization test is not performed on method return values unless the method is a
   *       visible static factory method whose return type is {@code C} or {@code C}'s subtype.
   * </ul>
   *
   * <p>In all cases, if {@code C} needs custom logic for testing serialization, you can add an
   * explicit {@code testSerializable()} test in the corresponding {@code CTest} class, and {@code

   * size before writing to pre-allocated space (e.g., for memory-mapped files or fixed-size
   * records).
   *
   * @since NEXT
   */
  public long serializedSize() {
    // We return a long (and not an int) because the max serialized size is approximately
    // 17 billion bytes (2^31 * 8 bytes).

    // The serialization header is 6 bytes:

   *   <li>Serialization test is not performed on method return values unless the method is a
   *       visible static factory method whose return type is {@code C} or {@code C}'s subtype.
   * </ul>
   *
   * <p>In all cases, if {@code C} needs custom logic for testing serialization, you can add an
   * explicit {@code testSerializable()} test in the corresponding {@code CTest} class, and {@code