/**
   * Opens a new buffered {@link Writer} for writing to this sink. The returned stream is not
   * required to be a {@link BufferedWriter} in order to allow implementations to simply delegate to
   * {@link #openStream()} when the stream returned by that method does not benefit from additional
   * buffering. This method returns a new, independent writer each time it is called.
   *

  /**
   * This is invoked for the last bytes of the input, which are not enough to fill a whole chunk.
   * The passed {@code ByteBuffer} is guaranteed to be non-empty.
   *
   * <p>This implementation simply pads with zeros and delegates to {@link #process(ByteBuffer)}.
   */
  protected void processRemaining(ByteBuffer bb) {
    Java8Compatibility.position(bb, bb.limit()); // move at the end

    /*
     * Except for \n, \t, and \r, all ASCII control characters are replaced with the Unicode
     * replacement character.
     *
     * Implementation note: An alternative to the following would be to make a map that simply
     * replaces the allowed ASCII whitespace characters with themselves and to set the minimum safe
     * character to 0x20. However this would slow down the escaping of simple strings that contain
     * \t, \n, or \r.

      }
    }
    return utf8Length;
  }

  /**
   * Returns {@code true} if {@code bytes} is a <i>well-formed</i> UTF-8 byte sequence according to
   * Unicode 6.0. Note that this is a stronger criterion than simply whether the bytes can be
   * decoded. For example, some versions of the JDK decoder will accept "non-shortest form" byte
   * sequences, but encoding never reproduces these. Such byte sequences are <i>not</i> considered

    //    second thread can then dispatch to the subscriber it took before the first thread does.
    //
    // All this makes me really wonder if there's any value in queueing here at all. A dispatcher
    // that simply loops through the subscribers and dispatches the event to each would actually
    // probably provide a stronger order guarantee, though that order would obviously be different
    // in some cases.

h3. Creation of an isolated classloader

    return (graph instanceof ImmutableValueGraph)
        ? (ImmutableValueGraph<N, V>) graph
        : new ImmutableValueGraph<N, V>(graph);
  }

  /**
   * Simply returns its argument.
   *
   * @deprecated no need to use this
   */
  @Deprecated
  public static <N, V> ImmutableValueGraph<N, V> copyOf(ImmutableValueGraph<N, V> graph) {
    return checkNotNull(graph);
  }

    return (graph instanceof ImmutableValueGraph)
        ? (ImmutableValueGraph<N, V>) graph
        : new ImmutableValueGraph<N, V>(graph);
  }

  /**
   * Simply returns its argument.
   *
   * @deprecated no need to use this
   */
  @Deprecated
  public static <N, V> ImmutableValueGraph<N, V> copyOf(ImmutableValueGraph<N, V> graph) {
    return checkNotNull(graph);
  }

  /**
   * Returns an ordering based on an <i>existing</i> comparator instance. Note that it is
   * unnecessary to create a <i>new</i> anonymous inner class implementing {@code Comparator} just
   * to pass it in here. Instead, simply subclass {@code Ordering} and implement its {@code compare}
   * method directly.
   *
   * <p>The returned object is serializable if {@code comparator} is serializable.
   *

        return clazz;
    }

    /**
     * <p>Find a Method using the methodKey provided.</p>
     * <p>Look in the methodMap for an entry. If found,
     * it'll either be a CACHE_MISS, in which case we
     * simply give up, or it'll be a Method, in which
     * case, we return it.</p>
     * <p>If nothing is found, then we must actually go
     * and introspect the method from the MethodMap.</p>
     * @param name Method name.