};

struct TF_Library {
  void* lib_handle;
  TF_Buffer op_list;
};

struct TF_Graph {
  TF_Graph();

  mutable tensorflow::mutex mu;
  tensorflow::Graph graph TF_GUARDED_BY(mu);

  // Runs shape inference.
  tensorflow::ShapeRefiner refiner TF_GUARDED_BY(mu);

  // Maps from name of an operation to the Node* in 'graph'.
  std::unordered_map<tensorflow::string, tensorflow::Node*> name_map
      TF_GUARDED_BY(mu);

  // Computes the rank of the tensor handle.
  //
  // Shapes are specified via callbacks because retrieving the shape of a tensor
  // is a blocking operation for async eager; custom devices should avoid
  // retrieving shapes of tensors they wrap until the custom device tensor's
  // shape is explicitly requested where possible.
  int (*num_dims)(void* data, TF_Status* status);

  // Computes the axis length at `dim_index`.

 *
 * The task runner is also responsible for releasing held threads when the library is unloaded.
 * This is for the benefit of container environments that implement code unloading.
 *
 * Most applications should share a process-wide [TaskRunner] and use queues for per-client work.
 */
class TaskRunner(
  val backend: Backend,
  internal val logger: Logger = TaskRunner.logger,
) {
  val lock: ReentrantLock = ReentrantLock()

/**
 * A set of tasks that are executed in sequential order.
 *
 * Work within queues is not concurrent. This is equivalent to each queue having a dedicated thread
 * for its work; in practice a set of queues may share a set of threads to save resources.
 */
class TaskQueue internal constructor(
  internal val taskRunner: TaskRunner,
  internal val name: String,
) {
  val lock: ReentrantLock = ReentrantLock()

                dr.server = resp.referrals[di].path.substring(1).toLowerCase();
            } else {
                dfsPathSplit(resp.referrals[di].node, arr);
                dr.server = arr[1];
                dr.share = arr[2];
                dr.path = arr[3];
            }
            dr.pathConsumed = resp.pathConsumed;

            di++;
            if (di == rn)
                break;

efficiently makes your stuff load faster and saves bandwidth.

OkHttp is an HTTP client that’s efficient by default:

 * HTTP/2 support allows all requests to the same host to share a socket.
 * Connection pooling reduces request latency (if HTTP/2 isn’t available).
 * Transparent GZIP shrinks download sizes.
 * Response caching avoids the network completely for repeat requests.

   * scattered in memory and thus don't interfere much with each
   * other. But Atomic objects residing in arrays will tend to be
   * placed adjacent to each other, and so will most often share
   * cache lines (with a huge negative performance impact) without
   * this precaution.
   *
   * In part because Cells are relatively large, we avoid creating

import jcifs.internal.util.SMBUtil;
import jcifs.util.Hexdump;


/**
 * 
 * 
 */
public class SmbComNTCreateAndX extends AndXServerMessageBlock implements Request<SmbComNTCreateAndXResponse> {

    // share access specified in SmbFile

    // create disposition

    /*
     * Creates a new file or supersedes the existing one
     */

    static final int FILE_SUPERSEDE = 0x0;

    /*

## Caveat

If you need to perform heavy background computation and you don't necessarily need it to be run by the same process (for example, you don't need to share memory, variables, etc), you might benefit from using other bigger tools like <a href="https://docs.celeryq.dev" class="external-link" target="_blank">Celery</a>.

L'ajout d'un `\f` (un caractère d'échappement "form feed") va permettre à **FastAPI** de tronquer la sortie utilisée pour OpenAPI à ce stade.

Il n'apparaîtra pas dans la documentation, mais d'autres outils (tel que Sphinx) pourront utiliser le reste.

```Python hl_lines="19-29"
{!../../docs_src/path_operation_advanced_configuration/tutorial004.py!}
```