li,ue,?ts&ae&-&as,pa,su,vog-su,?ht&ron-pa,uos-pa,??ew-&su,ue,vog-su,???2-ts&ae&-su,ht&ron-pa,uos-pa,??ew-&su,ue,??3-ts&aeht&ron-pa,uos-pa,?ew-ue,??,nil-kaerts,o-morf,r&adhtiwtliub,ow&-&sndnyd,ta-sndnyd,?ten-orehkcats,??sedal,u,?l&a&-morf,colottad,rebil-a-si,?f-morf,i&-morf,am&-sndnyd,detsohpw,??l&ecelffaw,uf-ytnuob:.a&hpla,teb,?,?ppmswa,ru-&elpmis,taen,?ssukoreh,xegap,?m&n-morf,pml.ppa,rofe&pyt.orp,rerac-htlaeh,?sacrasevres,uirarret-yltsaf,?n&a&cilbuper-a-si,f&-sllub-a-si,racsan-a-si,?i&cisum-a-...

   * Foo<Enum<?>>} according to JLS. See testRecursiveWildcardSubtypeBug() for a real example.
   *
   * <p>It appears that properly handling recursive type bounds in the presence of implicit type
   * bounds is not easy. For now we punt, hoping that this defect should rarely cause issues in real
   * code.
   *
   * @param formalType is {@code Foo<formalType>} a supertype of {@code Foo<T>}?

 *             return createExpensiveGraph(key);
 *           }
 *         });
 * }</pre>
 *
 * <p>Or equivalently,
 *
 * <pre>{@code
 * // In real life this would come from a command-line flag or config file
 * String spec = "maximumSize=10000,expireAfterWrite=10m";
 *
 * LoadingCache<Key, Graph> graphs = CacheBuilder.from(spec)
 *     .removalListener(MY_LISTENER)