*
   * How this works is best explained with an example:
   *
   * For a RateLimiter that produces 1 token per second, every second that goes by with the
   * RateLimiter being unused, we increase storedPermits by 1. Say we leave the RateLimiter unused
   * for 10 seconds (i.e., we expected a request at time X, but we are at time X + 10 seconds before
   * a request actually arrives; this is also related to the point made in the last paragraph), thus

            ImmutableSet.of(
                DurationSpec.of(0, SECONDS), DurationSpec.of(1, SECONDS), DurationSpec.of(1, DAYS)))
        .withExpireAfterAccesses(
            ImmutableSet.of(
                DurationSpec.of(0, SECONDS), DurationSpec.of(1, SECONDS), DurationSpec.of(1, DAYS)))
        .withRefreshes(ImmutableSet.of(DurationSpec.of(1, SECONDS), DurationSpec.of(1, DAYS)));
  }

            ImmutableSet.of(
                DurationSpec.of(0, SECONDS), DurationSpec.of(1, SECONDS), DurationSpec.of(1, DAYS)))
        .withExpireAfterAccesses(
            ImmutableSet.of(
                DurationSpec.of(0, SECONDS), DurationSpec.of(1, SECONDS), DurationSpec.of(1, DAYS)))
        .withRefreshes(ImmutableSet.of(DurationSpec.of(1, SECONDS), DurationSpec.of(1, DAYS)));
  }

import static com.google.common.util.concurrent.Uninterruptibles.getUninterruptibly;
import static java.util.concurrent.TimeUnit.MILLISECONDS;
import static java.util.concurrent.TimeUnit.SECONDS;
import static junit.framework.Assert.assertFalse;
import static junit.framework.Assert.fail;

import com.google.common.annotations.GwtCompatible;
import java.util.concurrent.ExecutionException;

      if (requestCaching.maxAgeSeconds != -1) {
        freshMillis = minOf(freshMillis, SECONDS.toMillis(requestCaching.maxAgeSeconds.toLong()))
      }

      var minFreshMillis: Long = 0
      if (requestCaching.minFreshSeconds != -1) {
        minFreshMillis = SECONDS.toMillis(requestCaching.minFreshSeconds.toLong())
      }

      var maxStaleMillis: Long = 0

// +protobuf.options.(gogoproto.goproto_stringer)=false
message MicroTime {
  // Represents seconds of UTC time since Unix epoch
  // 1970-01-01T00:00:00Z. Must be from 0001-01-01T00:00:00Z to
  // 9999-12-31T23:59:59Z inclusive.
  optional int64 seconds = 1;

  // Non-negative fractions of a second at nanosecond resolution. Negative
  // second values with fractions must still have non-negative nanos values

* Então ele retorna a **resposta**.

/// note | "Detalhes técnicos"

Se você tiver dependências com `yield`, o código de saída será executado *depois* do middleware.

Se houver alguma tarefa em segundo plano (documentada posteriormente), ela será executada *depois* de todo o middleware.

///

## Criar um middleware

Para criar um middleware, use o decorador `@app.middleware("http")` logo acima de uma função.

    //      last modified: 105 seconds ago
    //             served:   5 seconds ago
    //   default lifetime: (105 - 5) / 10 = 10 seconds
    //            expires:  10 seconds from served date = 5 seconds from now
    server.enqueue(
      MockResponse.Builder()
        .addHeader("Last-Modified: " + formatDate(-105, TimeUnit.SECONDS))
        .addHeader("Date: " + formatDate(-5, TimeUnit.SECONDS))

    }

    /**
     * Sets the minimum number of seconds that a response will continue to be fresh for. If the
     * response will be stale when [minFresh] have elapsed, the cached response will not be used and
     * a network request will be made.
     *
     * @param minFresh a non-negative integer. This is stored and transmitted with
     *     [TimeUnit.SECONDS] precision; finer precision will be lost.
     */

import static com.google.common.util.concurrent.Uninterruptibles.awaitUninterruptibly;
import static java.util.concurrent.Executors.newSingleThreadExecutor;
import static java.util.concurrent.TimeUnit.SECONDS;
import static org.junit.Assert.assertThrows;

import com.google.common.util.concurrent.ClosingFuture.ClosingCallable;
import com.google.common.util.concurrent.ClosingFuture.DeferredCloser;