long largeOps = largeCounter.total();

      double ratio = (double) largeOps / smallOps;
      assertWithMessage(
              "ratio of equals/hashCode/compareTo operations to build with %s entries versus %s"
                  + " entries",
              largeSize, smallSize)
          .that(ratio)
          .isAtMost(
              2
                  * constructionAsymptotics.applyAsDouble(largeSize)

**Elasticsearch version** (`bin/elasticsearch --version`):

**Plugins installed**: []

**JVM version** (`java -version`):

**OS version** (`uname -a` if on a Unix-like system):

**Description of the problem including expected versus actual behavior**:

**Steps to reproduce**:

Please include a *minimal* but *complete* recreation of the problem,
including (e.g.) index creation, mappings, settings, query etc.  The easier

   * node(s) to any node reachable from the start node(s), and has no paths from any start node to
   * any other start node, such as a tree or forest.
   *
   * <p>{@code forTree()} is especially useful (versus {@code forGraph()}) in cases where the data
   * structure being traversed is, in addition to being a tree/forest, also defined <a
   * href="https://github.com/google/guava/wiki/GraphsExplained#non-recursiveness">recursively</a>.

 * hook into this and this is the model we need to follow - how many of our REST services do not map
 * directory to a method in the application interface?
 *
 * - research extension points versus plugins
 *
 * from igor: two plugins A and B, both depend on the same library but use different versions, say
 * lib 1.0 and lib 2.0 when debugger hits a breakpoint inside a class from the library, IDE needs to

    assertTrue(elementsEqual(a, b));

    // An element differs.
    a = asList(4, 8, 15, 12, 23, 42);
    b = asList(4, 8, 15, 16, 23, 42);
    assertFalse(elementsEqual(a, b));

    // null versus non-null.
    a = Arrays.<@Nullable Integer>asList(4, 8, 15, null, 23, 42);
    b = asList(4, 8, 15, 16, 23, 42);
    assertFalse(elementsEqual(a, b));
    assertFalse(elementsEqual(b, a));

    // An element differs.
    a = asList(4, 8, 15, 12, 23, 42);
    b = asList(4, 8, 15, 16, 23, 42);
    assertFalse(elementsEqual(a.iterator(), b.iterator()));

    // null versus non-null.
    a = Arrays.<@Nullable Integer>asList(4, 8, 15, null, 23, 42);
    b = asList(4, 8, 15, 16, 23, 42);
    assertFalse(elementsEqual(a.iterator(), b.iterator()));

Au final, peu importe le cas parmi ceux ci-dessus, FastAPI fonctionnera de manière asynchrone et sera extrêmement rapide.

Mais si vous suivez bien les instructions ci-dessus, il pourra effectuer quelques optimisations et ainsi améliorer les performances.

## Détails techniques { #technical-details }

           rules:
           - apiGroups: [""]
             resources: ["events"]
             verbs:
             - create
             - patch
             - update
           - apiGroups: [""]
             resources: ["secrets"]
             verbs:
             - get
             - create
         - apiVersion: rbac.authorization.k8s.io/v1
           kind: ClusterRoleBinding

## Sobre Starlette { #about-starlette }

Não é recomendado fixar a versão de `starlette`.

Versões diferentes de **FastAPI** utilizarão uma versão específica e mais recente de Starlette.

Então, você pode deixar **FastAPI** usar a versão correta do Starlette.

## Sobre Pydantic { #about-pydantic }

Por causa disso, versões do FastAPI anteriores à 0.99.0 ainda usavam versões do OpenAPI inferiores à 3.1.0.

///

### `examples` no Pydantic e no FastAPI { #pydantic-and-fastapi-examples }