<ol class="breadcrumb float-sm-right">
                            <li class="breadcrumb-item"><la:link href="/admin/dict">
                                <la:message key="labels.dict_list_link"/>
                            </la:link></li>
                            <li class="breadcrumb-item"><la:message key="labels.dict_stopwords_list_link"/></li>
                        </ol>
                    </div>
                </div>

                        <ol class="breadcrumb float-sm-right">
                            <li class="breadcrumb-item"><la:link href="/admin/dict">
                                <la:message key="labels.dict_list_link"/>
                            </la:link></li>
                            <li class="breadcrumb-item"><la:message key="labels.dict_stemmeroverride_list_link"/></li>
                        </ol>
                    </div>
                </div>

 *
 * <p>Advantages compared to {@code double[]}:
 *
 * <ul>
 *   <li>All the many well-known advantages of immutability (read <i>Effective Java</i>, third
 *       edition, Item 17).
 *   <li>Has the value-based (not identity-based) {@link #equals}, {@link #hashCode}, and {@link
 *       #toString} behavior you expect.
 *   <li>Offers useful operations beyond just {@code get} and {@code length}, so you don't have to

 *   <li>It doesn't propagate exceptions, so apps don't have a way to react to them.
 *   <li>It doesn't interoperate well with RxJava, coroutines, and other more commonly used
 *       alternatives.
 *   <li>It imposes requirements on the lifecycle of its subscribers. For example, if an event
 *       occurs between when one subscriber is removed and the next subscriber is added, the event
 *       is dropped.

 * actions taken by the java garbage collection system:
 *
 * <ul>
 *   <li>invoking the {@code finalize} methods of unreachable objects
 *   <li>clearing weak references to unreachable referents
 *   <li>enqueuing weak references to unreachable referents in their reference queue
 * </ul>
 *

   *       fail.
   *   <li>If the constructor or factory method takes a parameter that {@link
   *       AbstractPackageSanityTests} doesn't know how to construct, the test will fail.
   *   <li>If there is no visible constructor or visible static factory method declared by {@code
   *       C}, {@code C} is skipped for serialization test, even if it implements {@link
   *       Serializable}.

 * </pre>
 *
 * <p>Here are some examples of recognized formats:
 *
 * <ul>
 *   <li>example.com
 *   <li>example.com:80
 *   <li>192.0.2.1
 *   <li>192.0.2.1:80
 *   <li>[2001:db8::1] - {@link #getHost()} omits brackets
 *   <li>[2001:db8::1]:80 - {@link #getHost()} omits brackets
 *   <li>2001:db8::1 - Use {@link #requireBracketsForIPv6()} to prohibit this
 * </ul>
 *

 *   <li>Convert multiple instances "forward" using {@code converter.convertAll(as)}.
 *   <li>Convert in the "backward" direction using {@code converter.reverse().convert(b)} or {@code
 *       converter.reverse().convertAll(bs)}.
 *   <li>Use {@code converter} or {@code converter.reverse()} anywhere a {@link
 *       java.util.function.Function} is accepted (for example {@link java.util.stream.Stream#map
 *       Stream.map}).

 * 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()
 *         <li>3 levels: average of 77ns per lock()/unlock()

   * all references {@code x}, {@code y}, and {@code z} (any of which may be null):
   *
   * <ul>
   *   <li>{@code equivalent(x, x)} is true (<i>reflexive</i> property)
   *   <li>{@code equivalent(x, y)} and {@code equivalent(y, x)} each return the same result
   *       (<i>symmetric</i> property)
   *   <li>If {@code equivalent(x, y)} and {@code equivalent(y, z)} are both true, then {@code