if (ops_->path_exists == nullptr)
    return errors::Unimplemented(tensorflow::strings::StrCat(
        "Filesystem for ", fname, " does not support FileExists()"));

  UniquePtrTo_TF_Status plugin_status(TF_NewStatus(), TF_DeleteStatus);
  const std::string translated_name = TranslateName(fname);
  ops_->path_exists(filesystem_.get(), translated_name.c_str(),

    TF_VLog(3, "CreateDir: directory already exists, not uploading %s", path);
    return TF_SetStatus(status, TF_ALREADY_EXISTS, path);
  }

  auto metadata = gcs_file->gcs_client.InsertObject(
      bucket, object, "",
      // Adding this parameter means HTTP_CODE_PRECONDITION_FAILED
      // will be returned if the object already exists, so avoid reuploading.
      gcs::IfGenerationMatch(0), gcs::Fields(""));

  }

  close(fd);
}

static void CreateDir(const TF_Filesystem* filesystem, const char* path,
                      TF_Status* status) {
  if (strlen(path) == 0)
    TF_SetStatus(status, TF_ALREADY_EXISTS, "already exists");
  else if (mkdir(path, /*mode=*/0755) != 0)
    TF_SetStatusFromIOError(status, errno, path);
  else
    TF_SetStatus(status, TF_OK, "");
}

  virtual Status AsyncWait() = 0;

  // Add a function (serialized FunctionDef protocol buffer) so that it can
  // be executed as an op. Return error if the function with the same name
  // already exists.
  virtual Status AddFunctionDef(const FunctionDef& fdef) = 0;

  // Notifies about the function removal.
  virtual Status AddRemoveFunctionNotifier(const string& func,

Status GradientRegistry::Register(
    const string& op_name, GradientFunctionFactory gradient_function_factory) {
  auto iter = registry_.find(op_name);
  if (iter != registry_.end()) {
    const string error_msg = "Gradient already exists for op: " + op_name + ".";
    return errors::AlreadyExists(error_msg);
  }
  registry_.insert({op_name, gradient_function_factory});
  return absl::OkStatus();
}
Status GradientRegistry::Lookup(

  status->status = c.construction_status();
  if (!status->status.ok()) return;

  if (op_reg_data->shape_inference_fn == nullptr) {
    status->status =
        InvalidArgument("No shape inference function exists for op '",
                        node_def.op(), "', did you forget to define it?");
    return;
  }

  status->status = c.Run(op_reg_data->shape_inference_fn);
  if (!status->status.ok()) return;

your changes by running the command:

```bash
tensorflow/tools/ci_build/ci_build.sh CPU tensorflow/tools/ci_build/ci_sanity.sh
```

This will catch most license, Python coding style and BUILD file issues that
may exist in your changes.

#### Running unit tests

There are two ways to run TensorFlow unit tests.

1.  Using tools and libraries installed directly on your system.

    Refer to the

    int64_t init_timeout_in_ms, int retries, TF_Status* status,
    bool clear_existing_contexts);

// Checks whether a remote worker is alive or not. This will return true even if
// the context doesn't exist on the remote worker.
TF_CAPI_EXPORT extern bool TFE_ContextCheckAlive(TFE_Context* ctx,
                                                 const char* worker_name,
                                                 TF_Status* status);

namespace {

template <typename BackwardFunction, typename TapeTensor>
struct BackpropInitialState {
  OpTape<BackwardFunction, TapeTensor> op_tape;

  // Map from tensor ID to how many references still exist for this tensor in
  // the tape.
  std::unordered_map<int64_t, int64_t> tensor_usage_counts;

  // Maps from op ID to how many output tensors of this op still need to have
  // their gradients computed.

                    output_path: str,
                    verbose: bool = False):
  """Creates a JUnit-based XML file, with each invocation as a testcase."""
  os.makedirs(os.path.dirname(output_path), exist_ok=True)
  failure_count = 0
  error_count = 0

  date_time = datetime.datetime
  attrib = {'name': 'Bazel Invocations', 'time': '0.0',
            'timestamp': date_time.isoformat(date_time.utcnow())}