* large that doing the lookup is noticeably slower than redoing the work would be.
         *
         * Ideally we'd have a real eviction policy, but until we see a problem in practice, I hope
         * that this will suffice. I have not even benchmarked with different size limits.
         */
        if (validClasses.size() > 1000) {
          validClasses.clear();
        }

---

Typische Beispiele für CPU-lastige Vorgänge sind Dinge, die komplexe mathematische Berechnungen erfordern.

Zum Beispiel:

        }
        Type[] resolvedBounds = new TypeResolver(forDependants).resolveTypes(bounds);
        /*
         * We'd like to simply create our own TypeVariable with the newly resolved bounds. There's
         * just one problem: We want to interoperate properly with the platform's built-in
         * implementation of TypeVariable, but the behavior of the built-in implementation differs
         * across platforms:
         *

        if (type.isInstance(value)) {
            return true;
        }
        // likely Boolean -> boolean, Short -> int etc. conversions, it's not the problem case we try to avoid
        return ((type.isPrimitive() || type.getName().startsWith("java.lang."))
                && value.getClass().getName().startsWith("java.lang."));
    }

                // Don't want to have two workers pulling from the queue.
                return;
              } else {
                // Increment the run counter to avoid the ABA problem of a submitter marking the
                // thread as QUEUED after it already ran and exhausted the queue before returning
                // from execute().
                workerRunCount++;

                // Don't want to have two workers pulling from the queue.
                return;
              } else {
                // Increment the run counter to avoid the ABA problem of a submitter marking the
                // thread as QUEUED after it already ran and exhausted the queue before returning
                // from execute().
                workerRunCount++;

         * large that doing the lookup is noticeably slower than redoing the work would be.
         *
         * Ideally we'd have a real eviction policy, but until we see a problem in practice, I hope
         * that this will suffice. I have not even benchmarked with different size limits.
         */
        if (validClasses.size() > 1000) {
          validClasses.clear();
        }

Comme l'éditeur connaît le type des variables, vous n'avez pas seulement l'auto-complétion, mais aussi de la détection d'erreurs :

<img src="/img/python-types/image04.png">

Maintenant que vous avez connaissance du problème, convertissez `age` en <abbr title="string">chaîne de caractères</abbr> grâce à `str(age)` :

{*../../docs_src/python_types/tutorial004.py hl[2] *}

## Déclarer des types

   * for an RPC may wait for a similar call that requests a long timeout, or a call by an
   * unprivileged user may return a resource accessible only to a privileged user making a similar
   * call. To prevent this problem, create a key object that includes all values that affect the
   * result of the query. Or use {@code LoadingCache.get(K)}, which lacks the ability to refer to
   * state other than that in the key.
   *

    assertEquals(Integer.MAX_VALUE, ms.tailMultiset("a", CLOSED).size());
  }

  @J2ktIncompatible
  @GwtIncompatible // reflection
  @AndroidIncompatible // Reflection bug, or actual binary compatibility problem?
  public void testElementSetBridgeMethods() {
    for (Method m : TreeMultiset.class.getMethods()) {
      if (m.getName().equals("elementSet") && m.getReturnType().equals(SortedSet.class)) {
        return;
      }