*
   * <ul>
   *   <li>Any {@link ExecutionException} has its <i>cause</i> wrapped in an {@code X} if the cause
   *       is a checked exception, an {@link UncheckedExecutionException} if the cause is a {@code
   *       RuntimeException}, or an {@link ExecutionError} if the cause is an {@code Error}.
   *   <li>Any {@link InterruptedException} is wrapped in an {@code X} (after restoring the

                    } else {
                        logger.debug("Skipping step {} because a dependency has failed", step);
                    }
                } else {
                    // Skip because predecessors failed or were skipped
                    logger.debug(
                            "Skipping step {} because one or more predecessors did not execute successfully", step);
                }

            """
            An alternative name for the parameter field.

            This will be used to extract the data and for the generated OpenAPI.
            It is particularly useful when you can't use the name you want because it
            is a Python reserved keyword or similar.
            """
        ),
    ] = None,
    alias_priority: Annotated[
        Union[int, None],
        Doc(
            """

 *
 *  * https://example.com/
 *
 * This is because those two hosts share the same IP address. This is an old, bad design decision
 * that makes `java.net.URL` unusable for many things. It shouldn't be used as a [Map] key or in a
 * [Set]. Doing so is both inefficient because equality may require a DNS lookup, and incorrect
 * because unequal URLs may be equal because of how they are hosted.
 *
 * ### Equal URLs should be equal
 *

 *  Fix: Calling `disconnect()` on a connecting `HttpUrlConnection` could cause it to retry in an
    infinite loop! This regression was introduced in OkHttp 2.7.0.
 *  Fix: Drop cookies that contain ASCII NULL and other bad characters. Previously such cookies
    would cause OkHttp to crash when they were included in a request.

              size,
              (e1, e2) -> {
                // requireNonNull is safe because the first `size` elements have been filled in.
                requireNonNull(e1);
                requireNonNull(e2);
                return comparator.compare(e1.getKey(), e2.getKey());
              });
          // requireNonNull is safe because the first `size` elements have been filled in.

common:nvcc_clang --config=cuda_nvcc

# Debug config
common:dbg -c dbg
# Only include debug info for files under tensorflow/, excluding kernels, to
# reduce the size of the debug info in the binary. This is because if the debug
# sections in the ELF binary are too large, errors can occur. See
# https://github.com/tensorflow/tensorflow/issues/48919.
# Users can still include debug info for a specific kernel, e.g. with:

   *
   * - method reference: Inside Google, CacheBuilder is used from the implementation of a custom
   *   ClassLoader that is sometimes used as a system classloader. That's a problem because
   *   method-reference linking tries to look up the system classloader, and it fails because there
   *   isn't one yet.
   *
   * - lambda: Outside Google, we got a report of a similar problem in
   *   https://github.com/google/guava/issues/6565
   */

      }

      @Override
      public I next() {
        if (!hasNext()) {
          throw new NoSuchElementException();
        }
        /*
         * requireNonNull is safe because our callers always pass non-null arguments. Each element
         * of the array becomes null only when we iterate past it and then clear it.
         */
        I result = requireNonNull(elements[index]);

    assertThat(response.body.string()).isEqualTo("abc")
    assertThat(server.takeRequest().exchangeIndex).isEqualTo(0)

    // Note that although we have two routes available, we only use one. The retry is permitted
    // because there are routes available, but it chooses the existing connection since it isn't
    // yet considered unhealthy.
    assertThat(server.takeRequest().exchangeIndex).isEqualTo(1)
  }

  @Test