*       cleared, we know that we're not executing inline any more
     * </ul>
     *
     * All the states where thread != currentThread are identical for our purposes, and so even
     * though it's racy, we don't care which of those values we get, so no need to synchronize.
     */
    @LazyInit @Nullable Thread thread;

    /** Only used by the thread associated with this object */
    @Nullable Runnable nextTask;

   * they are declared by the same {@link java.lang.reflect.GenericDeclaration} and have the same
   * name, even if their bounds differ.
   *
   * <p>While resolving a type variable from a {@code var -> type} map, we don't care whether the
   * type variable's bound has been partially resolved. As long as the type variable "identity"
   * matches.
   *
   * <p>On the other hand, if for example we are resolving {@code List<A extends B>} to {@code

     *     passed to {@code whenAllSucceed}, if that is the method you used to create this {@code
     *     FutureCombiner}). Even if you don't care about the value of the future, you should
     *     typically check whether it failed: See <a

	if took > maxDurationSecondsForLog {
		// Log if we took a lot of time to load.
		logger.Info("IAM expired STS purge took %.2fs", took)
	}

	// Store the newly populated map in the iam cache. This takes care of
	// removing stale entries from the existing map.
	cache.iamSTSAccountsMap = stsAccountsFromStore

	stsAccPoliciesFromStore.Range(func(k string, v MappedPolicy) bool {
		cache.iamSTSPolicyMap.Store(k, v)

   * non-deterministic behavior at an undetermined time in the future.
   *
   * <p>Note that the fail-fast behavior of an iterator cannot be guaranteed as it is, generally
   * speaking, impossible to make any hard guarantees in the presence of unsynchronized concurrent
   * modification. Fail-fast iterators throw {@code ConcurrentModificationException} on a

    this.allowsNullKeys = allowsNullKeys;
    this.allowsNullValues = allowsNullValues;
    this.supportsIteratorRemove = supportsIteratorRemove;
  }

  /**
   * Used by tests that require a map, but don't care whether it's populated or not.
   *
   * @return a new map instance.
   */
  protected Map<K, V> makeEitherMap() {
    try {
      return makePopulatedMap();
    } catch (UnsupportedOperationException e) {

          // have no idea.
          // TODO(lukes): consider building everything in terms of ListenableScheduledFuture then
          // the AbstractService could monitor the future directly. Rescheduling is still hard...
          // but it would help with some of these lock ordering issues.
          scheduleFailure = e;
          toReturn = new FutureAsCancellable(immediateCancelledFuture());
        } finally {

* Startup and shutdown events.
* Test client built on HTTPX.
* CORS, GZip, Static Files, Streaming responses.
* Session and Cookie support.
* 100% test coverage.
* 100% type annotated codebase.
* Few hard dependencies.

Starlette is currently the fastest Python framework tested. Only surpassed by Uvicorn, which is not a framework, but a server.

Starlette provides all the basic web microframework functionality.

                } else {
                    /*
                     * NOTE: Other configurators like the map-oriented one don't call into the listener, so do it the
                     * hard way.
                     */
                    validateParameters(mojoDescriptor, configuration, expressionEvaluator);
                }
            }

        } catch (ComponentConfigurationException e) {

      // implementation because:
      // 1. the string constructor can avoid an extra copy most of the time by correctly sizing the
      //    internal char array (hard to avoid using StringBuilder)
      // 2. we avoid extra copies into temporary buffers altogether
      // The downside is that this will cause us to store the file bytes in memory twice for a short
      // amount of time.