}
    }

    // subsequent calls should call the loader again, not get the old exception
    assertThrows(UncheckedExecutionException.class, () -> cache.getUnchecked("bar"));
    assertThat(callCount.get()).isEqualTo(2);
  }

  /**
   * Test-helper method that performs {@code nThreads} concurrent calls to {@code cache.get(key)} or

### Bug or Regression

- Avoid GCE API calls when initializing GCE CloudProvider for Kubelets.

        public boolean isFinished() {
            return finished;
        }

        /**
         * Stops the crawling operation gracefully.
         * If a data store is currently active, this method calls
         * its stop method to halt the crawling process.
         */
        public void stopCrawling() {
            if (dataStore != null) {
                dataStore.stop();
            }
        }

   * table and deduplicating elements as they come, so it only allocates O(max(distinct,
   * expectedCapacity)) rather than O(calls to add).
   *
   * <p>This implementation attempts to detect hash flooding, and if it's identified, falls back to
   * JdkBackedSetBuilderImpl.
   */
  private static final class RegularSetBuilderImpl<E> extends SetBuilderImpl<E> {

     * The configuration string format is "clientName:pattern,clientName:pattern,..."
     * Results are cached to improve performance on subsequent calls.
     *
     * @param value the client rule configuration string
     * @return a list of pairs containing client names and their corresponding compiled patterns
     */

     */
    ContiguousSet<Integer> enormous = ContiguousSet.create(Range.<Integer>all(), integers());
    assertTrue(enormous instanceof RegularContiguousSet);
    // We can't use reserializeAndAssert because it calls hashCode, which is enormously slow.
    ContiguousSet<Integer> enormousReserialized = reserialize(enormous);
    assertEquals(enormous, enormousReserialized);
  }

 * tailored to that use case.
 *
 * <p>This uses the same efficient implementation as {@link Ordering#leastOf(Iterable, int)},
 * offering expected O(n + k log k) performance (worst case O(n log k)) for n calls to {@link
 * #offer} and a call to {@link #topK}, with O(k) memory. In comparison, quickselect has the same
 * asymptotics but requires O(n) memory, and a {@code PriorityQueue} implementation takes O(n log

        int n;
        final NameQueryRequest request = new NameQueryRequest(name);
        final NameQueryResponse response = new NameQueryResponse();

        if (addr != null) { /* UniAddress calls always use this
                             * because it specifies addr
                             */
            request.addr = addr; /* if addr ends with 255 flag it bcast */

//
// This call may not block for execution of ops enqueued concurrently with this
// call.
TF_CAPI_EXPORT extern void TFE_ExecutorWaitForAllPendingNodes(
    TFE_Executor*, TF_Status* status);

// When an error happens, any pending operations are discarded, and newly issued
// ops return an error. This call clears the error state and re-enables
// execution of newly issued ops.
//

        /**
         * The last non-TESTRESPONSE snippet.
         */
        Snippet previousTest

        /**
         * Called each time a snippet is encountered. Tracks the snippets and
         * calls buildTest to actually build the test.
         */
        void handleSnippet(Snippet snippet) {
            if (RestTestsFromSnippetsTask.isConsoleCandidate(snippet)) {