}

  @GwtIncompatible // Shorts.fromByteArray, Shorts.toByteArray
  public void testByteArrayRoundTrips() {
    Random r = new Random(5);
    byte[] b = new byte[Shorts.BYTES];

    // total overkill, but, it takes 0.1 sec so why not...
    for (int i = 0; i < 10000; i++) {
      short num = (short) r.nextInt();
      assertThat(Shorts.fromByteArray(Shorts.toByteArray(num))).isEqualTo(num);

      r.nextBytes(b);

   * given {@code Future}. If the given {@code Future} fails, the returned {@code Future} fails with
   * the same exception (and the function is not invoked).
   *
   * <p>More precisely, the returned {@code Future} takes its result from a {@code Future} produced
   * by applying the given {@code AsyncFunction} to the result of the original {@code Future}.
   * Example usage:
   *
   * {@snippet :

   *
   * <p>For more on IPv4 mapped addresses see section 2.5.5.2 of <a target="_parent"
   * href="http://tools.ietf.org/html/rfc4291#section-2.5.5.2">RFC 4291</a>.
   *
   * <p>Note: This method takes a {@code String} argument because {@link InetAddress} automatically
   * collapses mapped addresses to IPv4. (It is actually possible to avoid this using one of the

on:  `But why did they live at the bottom of a well?'

  `Take some more tea,' the March Hare said to Alice, very
earnestly.

  `I've had nothing yet,' Alice replied in an offended tone, `so
I can't take more.'

  `You mean you can't take LESS,' said the Hatter:  `it's very
easy to take MORE than nothing.'

  `Nobody asked YOUR opinion,' said Alice.

 *  Fix: Don't crash if a cookie's value is a lone double quote character.
 *  Fix: Don't crash when canceling an event source created by `EventSources.processResponse()`.
 *  New: `Cache` now has a public constructor that takes an [okio.FileSystem]. This should make it
    possible to implement decorators for cache encryption or compression.
 *  New: `Cookie.newBuilder()` to build upon an existing cookie.

    Ordering<Integer> integers = Ordering.explicit(1);

    // Like by like equals like
    Ordering<Number> unusedA = numbers.compound(numbers);

    // The compound takes the more specific type of the two, regardless of order

    Ordering<Number> unusedB = numbers.compound(objects);
    Ordering<Number> unusedC = objects.compound(numbers);

    Ordering<Integer> integers = Ordering.explicit(1);

    // Like by like equals like
    Ordering<Number> unusedA = numbers.compound(numbers);

    // The compound takes the more specific type of the two, regardless of order

    Ordering<Number> unusedB = numbers.compound(objects);
    Ordering<Number> unusedC = objects.compound(numbers);

 * Benchmarks (as of December 2011) show that:
 *
 * <ul>
 *   <li>for an unnested {@code lock()} and {@code unlock()}, a cycle detecting lock takes 38ns as
 *       opposed to the 24ns taken by a plain lock.
 *   <li>for nested locking, the cost increases with the depth of the nesting:
 *       <ul>
 *         <li>2 levels: average of 64ns per lock()/unlock()

## Version 3.0.0

_2016-01-13_

This release commits to a stable 3.0 API. Read the 3.0.0-RC1 changes for advice
on upgrading from 2.x to 3.x.

 *  **The `Callback` interface now takes a `Call`**. This makes it easier to
    check if the call was canceled from within the callback. When migrating
    async calls to this new API, `Call` is now the first parameter for both
    `onResponse()` and `onFailure()`.

    thread.setPriority(Thread.MAX_PRIORITY);
    Thread.class.getMethod("resume").invoke(thread);
    thread.join();
    // It's possible to race and suspend the thread just before the park call actually takes effect,
    // causing the thread to be suspended for 3.5 seconds, and then park itself for 2 seconds after
    // being resumed. To avoid a flake in this scenario, calculate how long that thread actually