* lookups.
 *
 * <p>During construction, names are normalized in two ways:
 *
 * <ol>
 *   <li>ASCII uppercase characters are converted to lowercase.
 *   <li>Unicode dot separators other than the ASCII period ({@code '.'}) are converted to the ASCII
 *       period.
 * </ol>
 *
 * <p>The normalized values will be returned from {@link #toString()} and {@link #parts()}, and will

 * <p>There are two entry points for creating a {@code Traverser}: {@link
 * #forTree(SuccessorsFunction)} and {@link #forGraph(SuccessorsFunction)}. You should choose one
 * based on your answers to the following questions:
 *
 * <ol>
 *   <li>Is there only one path to any node that's reachable from any start node? (If so, the graph
 *       to be traversed is a tree or forest even if it is a subgraph of a graph which is neither.)

 *     <li><code>alpha</code> or <code>a</code></li>
 *     <li><code>beta</code> or <code>b</code></li>
 *     <li><code>milestone</code> or <code>m</code></li>
 *     <li><code>rc</code> or <code>cr</code></li>
 *     <li><code>snapshot</code></li>
 *     <li><code>ga</code> or <code>final</code></li>
 *     <li><code>sp</code></li>
 *     </ol>

   *
   * <ol>
   *   <li>'this' and {@code formalType} are equal.
   *   <li>'this' and {@code formalType} have equal canonical form.
   *   <li>{@code formalType} is {@code <? extends Foo>} and 'this' is a subtype of {@code Foo}.
   *   <li>{@code formalType} is {@code <? super Foo>} and 'this' is a supertype of {@code Foo}.
   * </ol>
   *

     *
     * <p>This method performs the main logic of ServiceManager in the following steps.
     *
     * <ol>
     *   <li>Update the {@link #servicesByState()}
     *   <li>Update the {@link #startupTimers}
     *   <li>Based on the new state queue listeners to run
     *   <li>Run the listeners (outside of the lock)
     * </ol>
     */
    void transitionService(Service service, State from, State to) {
      checkNotNull(service);

 * example round-trip using {@link com.google.common.primitives.Doubles#stringConverter}:
 *
 * <ol>
 *   <li>{@code stringConverter().convert("1.00")} returns the {@code Double} value {@code 1.0}
 *   <li>{@code stringConverter().reverse().convert(1.0)} returns the string {@code "1.0"} --
 *       <i>not</i> the same string ({@code "1.00"}) we started with
 * </ol>
 *
 * <p>Note that it should still be the case that the round-tripped and original objects are

 * example round-trip using {@link com.google.common.primitives.Doubles#stringConverter}:
 *
 * <ol>
 *   <li>{@code stringConverter().convert("1.00")} returns the {@code Double} value {@code 1.0}
 *   <li>{@code stringConverter().reverse().convert(1.0)} returns the string {@code "1.0"} --
 *       <i>not</i> the same string ({@code "1.00"}) we started with
 * </ol>
 *
 * <p>Note that it should still be the case that the round-tripped and original objects are

 * <p>The executor follows these main steps:</p>
 * <ol>
 *   <li>Initial plan creation based on project dependencies and task segments</li>
 *   <li>Concurrent execution of build steps while maintaining dependency order</li>
 *   <li>Dynamic replanning when necessary (e.g., for forked executions)</li>
 *   <li>Project setup, execution, and teardown phases management</li>
 * </ol>
 *
 * <h2>Thread Management:</h2>

  static byte[] createBuffer() {
    return new byte[BUFFER_SIZE];
  }

  /**
   * There are three methods to implement {@link FileChannel#transferTo(long, long,
   * WritableByteChannel)}:
   *
   * <ol>
   *   <li>Use sendfile(2) or equivalent. Requires that both the input channel and the output
   *       channel have their own file descriptors. Generally this only happens when both channels

 * As a result, it usually helps to read chained ordering expressions <i>backwards</i>. For example,
 * when {@code compare} is called on the above ordering:
 *
 * <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