return (ImmutableMapEntry<K, V>[]) new ImmutableMapEntry<?, ?>[size];
  }

  ImmutableMapEntry(K key, V value) {
    super(key, value);
    checkEntryNotNull(key, value);
  }

  // Redeclare methods to make them `final`, just to be extra-safe.

  @Override
  @ParametricNullness
  public final K getKey() {
    return super.getKey();
  }

  @Override
  @ParametricNullness

{* ../../docs_src/response_model/tutorial003_py310.py hl[24] *}

...we declared the `response_model` to be our model `UserOut`, that doesn't include the password:

{* ../../docs_src/response_model/tutorial003_py310.py hl[22] *}

So, **FastAPI** will take care of filtering out all the data that is not declared in the output model (using Pydantic).

### `response_model` or Return Type { #response-model-or-return-type }

  public Spliterator<E> spliterator() {
    return Spliterators.spliterator(array, SPLITERATOR_CHARACTERISTICS);
  }

  // TODO(lowasser): benchmark optimizations for equals() and see if they're worthwhile

  // redeclare to help optimizers with b/310253115
  @SuppressWarnings("RedundantOverride")
  @Override
  @J2ktIncompatible
  @GwtIncompatible
    Object writeReplace() {
    return super.writeReplace();
  }

        return axes.get(axis).get(axisIndex);
      }

      @Override
      boolean isPartialView() {
        return true;
      }

      // redeclare to help optimizers with b/310253115
      @SuppressWarnings("RedundantOverride")
      @J2ktIncompatible // serialization
      @Override
      @GwtIncompatible // serialization
      Object writeReplace() {

Comme tous les **champs changent réellement** (le type inclut désormais `None` et ils ont maintenant une valeur par défaut de `None`), nous devons les **redéclarer**.

Nous n'avons pas vraiment besoin d'hériter de `HeroBase` puisque nous redéclarons tous les champs. Je le laisse hériter juste pour la cohérence, mais ce n'est pas nécessaire. C'est plutôt une question de goût personnel. 🤷

          // Since this is an artificially generated type variable, we don't bother checking
          // subtyping between declared type bound and actual type bound. So it's possible that we
          // may generate something like <capture#1-of ? extends Foo&SubFoo>.
          // Checking subtype between declared and actual type bounds
          // adds recursive isSubtypeOf() call and feels complicated.

{* ../../docs_src/response_change_status_code/tutorial001_py310.py hl[1,9,12] *}

And then you can return any object you need, as you normally would (a `dict`, a database model, etc).

And if you declared a `response_model`, it will still be used to filter and convert the object you returned.

    return CollectSpliterators.indexed(
        size(),
        ImmutableList.SPLITERATOR_CHARACTERISTICS | Spliterator.SORTED | Spliterator.DISTINCT,
        delegateList()::get,
        comparator());
  }

  // redeclare to help optimizers with b/310253115
  @SuppressWarnings("RedundantOverride")
  @Override
  @J2ktIncompatible
  @GwtIncompatible
    Object writeReplace() {
    return super.writeReplace();
  }

    }

#### Adding additional integration tests

Additional integration tests for a certain elasticsearch modules that are specific to certain cluster configuration can be declared in a separate so called `qa` subproject of your module.

The benefit of a dedicated project for these tests are:
- `qa` projects are dedicated two specific usecases and easier to maintain

// not as the type of a variable.
type comparable interface{ comparable }

// iota is a predeclared identifier representing the untyped integer ordinal
// number of the current const specification in a (usually parenthesized)
// const declaration. It is zero-indexed.
const iota = 0 // Untyped int.

// nil is a predeclared identifier representing the zero value for a
// pointer, channel, func, interface, map, or slice type.