*
 * <ol>
 *   <li>First, if only one {@code Foo} is null, that null value is treated as <i>greater</i>
 *   <li>Next, non-null {@code Foo} values are passed to {@code getBarFunction} (we will be
 *       comparing {@code Bar} values from now on)
 *   <li>Next, if only one {@code Bar} is null, that null value is treated as <i>lesser</i>
 *   <li>Finally, natural ordering is used (i.e. the result of {@code Bar.compareTo(Bar)} is

 *
 * <ol>
 *   <li>First, if only one {@code Foo} is null, that null value is treated as <i>greater</i>
 *   <li>Next, non-null {@code Foo} values are passed to {@code getBarFunction} (we will be
 *       comparing {@code Bar} values from now on)
 *   <li>Next, if only one {@code Bar} is null, that null value is treated as <i>lesser</i>
 *   <li>Finally, natural ordering is used (i.e. the result of {@code Bar.compareTo(Bar)} is

   * compares, using {@link #compare(double, double)}), the first pair of values that follow any
   * common prefix, or when one array is a prefix of the other, treats the shorter array as the
   * lesser. For example, {@code [] < [1.0] < [1.0, 2.0] < [2.0]}.
   *
   * <p>The returned comparator is inconsistent with {@link Object#equals(Object)} (since arrays

   * surprising for implementations of {@code HashCode} that represent the number in big-endian
   * since everything else in the hashing API uniformly treats multibyte values as little-endian.
   *
   * <p>To create a {@code HashCode} from its string representation, see {@link #fromString}.
   */
  @Override
  public final String toString() {
    byte[] bytes = getBytesInternal();

   * surprising for implementations of {@code HashCode} that represent the number in big-endian
   * since everything else in the hashing API uniformly treats multibyte values as little-endian.
   *
   * <p>To create a {@code HashCode} from its string representation, see {@link #fromString}.
   */
  @Override
  public final String toString() {
    byte[] bytes = getBytesInternal();

 * 2^32 + x}). The methods for which signedness is not an issue are in {@link Ints}, as well as
 * signed versions of methods for which signedness is an issue.
 *
 * <p>In addition, this class provides several static methods for converting an {@code int} to a
 * {@code String} and a {@code String} to an {@code int} that treat the {@code int} as an unsigned
 * number.
 *

 * 2^32 + x}). The methods for which signedness is not an issue are in {@link Ints}, as well as
 * signed versions of methods for which signedness is an issue.
 *
 * <p>In addition, this class provides several static methods for converting an {@code int} to a
 * {@code String} and a {@code String} to an {@code int} that treat the {@code int} as an unsigned
 * number.
 *

	//go:debug default=go1.21
	//go:debug panicnil=1
	//go:debug asynctimerchan=0

Starting in Go 1.21, the Go toolchain treats a `//go:debug` directive
with an unrecognized GODEBUG setting as an invalid program.
Programs with more than one `//go:debug` line for a given setting
are also treated as invalid.
(Older toolchains ignore `//go:debug` directives entirely.)

The defaults that will be compiled into a main package

    // array slot is read and written exactly once. However, it can have very poor memory locality:
    // benchmarking shows it can take 7 times longer than the other two in some cases. The other two
    // do n swaps, minus a delta (0 or 2 for Reversal, gcd(d, n) for Successive), so that's about

   * compares, using {@link #compare(long, long)}), the first pair of values that follow any common
   * prefix, or when one array is a prefix of the other, treats the shorter array as the lesser. For
   * example, {@code [] < [1L] < [1L, 2L] < [2L]}.
   *
   * <p>The returned comparator is inconsistent with {@link Object#equals(Object)} (since arrays