* supplied by the user of the class. For example, {@code Multiset.Entry.getElement()} returns
 * {@code @ParametricNullness E}, which means:
 *
 * <ul>
 *   <li>{@code getElement} on a {@code Multiset.Entry<@NonNull String>} returns {@code @NonNull
 *       String}.
 *   <li>{@code getElement} on a {@code Multiset.Entry<@Nullable String>} returns {@code @Nullable
 *       String}.
 * </ul>
 *

 * supplied by the user of the class. For example, {@code Multiset.Entry.getElement()} returns
 * {@code @ParametricNullness E}, which means:
 *
 * <ul>
 *   <li>{@code getElement} on a {@code Multiset.Entry<@NonNull String>} returns {@code @NonNull
 *       String}.
 *   <li>{@code getElement} on a {@code Multiset.Entry<@Nullable String>} returns {@code @Nullable
 *       String}.
 * </ul>
 *

 * map from keys to <i>nonempty</i> collections of values:
 *
 * <ul>
 *   <li>a → 1, 2
 *   <li>b → 3
 * </ul>
 *
 * ... or as a single "flattened" collection of key-value pairs:
 *
 * <ul>
 *   <li>a → 1
 *   <li>a → 2
 *   <li>b → 3
 * </ul>
 *
 * <p><b>Important:</b> although the first interpretation resembles how most multimaps are

 *   <li>{@code expireAfterWrite=[duration]}: sets {@link CacheBuilder#expireAfterWrite}.
 *   <li>{@code refreshAfterWrite=[duration]}: sets {@link CacheBuilder#refreshAfterWrite}.
 *   <li>{@code weakKeys}: sets {@link CacheBuilder#weakKeys}.
 *   <li>{@code softValues}: sets {@link CacheBuilder#softValues}.
 *   <li>{@code weakValues}: sets {@link CacheBuilder#weakValues}.
 *   <li>{@code recordStats}: sets {@link CacheBuilder#recordStats}.
 * </ul>

 *
 * <ul>
 *   <li>An annotation named just "Suppress" might someday be treated by a non-Android tool as a
 *       suppression. This would follow the precedent of many of our annotation processors, which
 *       look for any annotation named, e.g., "GwtIncompatible," regardless of package.
 *   <li>An annotation named just "Suppress" might suggest to users that the test is suppressed

   * thread safe.
   *
   * <p>The initial state of the builder is such that:
   *
   * <ul>
   *   <li>There are no replacement mappings
   *   <li>{@code safeMin == Character.MIN_VALUE}
   *   <li>{@code safeMax == Character.MAX_VALUE}
   *   <li>{@code unsafeReplacement == null}
   * </ul>
   *
   * <p>For performance reasons escapers created by this builder are not Unicode aware and will not

 *
 * <ul>
 *   <li>An annotation named just "Suppress" might someday be treated by a non-Android tool as a
 *       suppression. This would follow the precedent of many of our annotation processors, which
 *       look for any annotation named, e.g., "GwtIncompatible," regardless of package.
 *   <li>An annotation named just "Suppress" might suggest to users that the test is suppressed

     * <ul>
     *   <li>If it starts with either "pom." or "project.", the expression is evaluated against the model.</li>
     *   <li>If the value is null, get the value from the context.</li>
     *   <li>If the value is null, but the context contains the expression, don't replace the expression string
     *       with the value, and continue to find other expressions.</li>

/**
 * Specific problems during resolution that we want to account for:
 * <ul>
 *   <li>missing metadata</li>
 *   <li>version range violations</li>
 *   <li>version circular dependencies</li>
 *   <li>missing artifacts</li>
 *   <li>network/transfer errors</li>
 *   <li>file system errors: permissions</li>
 * </ul>
 *
 * TODO carlos: all these possible has*Exceptions and get*Exceptions methods make the clients too

 *
 * <p>Valid:
 *
 * <ul>
 *   <li>Live: valid key/value are set
 *   <li>Loading: loading is pending
 * </ul>
 *
 * <p>Invalid:
 *
 * <ul>
 *   <li>Expired: time expired (key/value may still be set)
 *   <li>Collected: key/value was partially collected, but not yet cleaned up
 *   <li>Unset: marked as unset, awaiting cleanup or reuse
 * </ul>
 */
@GwtIncompatible