TF_Status* status);

// Report error (specified by error_code and error_message) to other tasks in
// the cluster.
TF_CAPI_EXPORT extern void TFE_ReportErrorToCluster(TFE_Context* ctx,
                                                    int error_code,
                                                    const char* error_message,

          "test_pluggable_device.so"));
  TF_Library* lib = TF_LoadPluggableDeviceLibrary(lib_path.c_str(), status);
  TF_Code code = TF_GetCode(status);
  string status_msg(TF_Message(status));
  TF_DeleteStatus(status);
  ASSERT_EQ(TF_OK, code) << status_msg;
  TF_DeletePluggableDeviceLibraryHandle(lib);
#endif  // !defined(TENSORFLOW_NO_SHARED_OBJECTS)
#endif  // !defined(PLATFORM_WINDOWS)
}

    TF_SetStatusFromGCSStatus(gcs_status, status);
    auto status_code = TF_GetCode(status);
    if (status_code != TF_OK && status_code != TF_NOT_FOUND) return;
    // If this file does not exist on server, we will need to sync it.
    bool sync_need = (status_code == TF_NOT_FOUND);
    file->plugin_file = new tf_writable_file::GCSFile(

  for (size_t i = 0; i < tasks.length; ++i) {
    const auto& result = (*results)[i];
    TF_Status s;
    TF_SetStatus(&s, static_cast<TF_Code>(result.error_code()),
                 std::string(result.error_message()).data());
    if (TF_GetCode(&s) != TF_Code::TF_OK) {
      tensorflow::CoordinationServiceError error;
      *error.mutable_source_task() = result.error_payload().source_task();

  EXPECT_EQ(TF_OK, TF_GetCode(status)) << TF_Message(status);

  TFE_TensorHandle* retvals[1];
  int num_retvals = 1;
  TFE_Execute(matmul, &retvals[0], &num_retvals, status);
  EXPECT_EQ(TF_OK, TF_GetCode(status)) << TF_Message(status);

  auto* retval_task0 =
      TFE_TensorHandleCopyToDevice(retvals[0], ctx, local_device_name, status);
  ASSERT_EQ(TF_OK, TF_GetCode(status)) << TF_Message(status);

      TFE_NewOp(context.get(), "VarHandleOp", status.get()), TFE_DeleteOp);
  ASSERT_TRUE(TF_GetCode(status.get()) == TF_OK) << TF_Message(status.get());
  TFE_OpSetAttrType(handle_op.get(), "dtype", TF_FLOAT);
  TFE_OpSetAttrShape(handle_op.get(), "shape", /*dims=*/nullptr, /*num_dims=*/0,
                     status.get());
  ASSERT_TRUE(TF_GetCode(status.get()) == TF_OK) << TF_Message(status.get());
  auto outputs =

EOF
  cat $BATS_TEST_TMPDIR/out
  [[ ! -s $BATS_TEST_TMPDIR/out ]]
}

@test "All tensorflow.org/code links point to real files" {
    for i in $(grep -onI 'https://www.tensorflow.org/code/[a-zA-Z0-9/._-]\+' -r tensorflow); do
        target=$(echo $i | sed 's!.*https://www.tensorflow.org/code/!!g')

        if [[ ! -f $target ]] && [[ ! -d $target ]]; then
            echo "$i" >> errors.txt
        fi

// without using tensorflow python code, hence left unspecified here.
//
// Gradient is the type returned by gradient functions. In Python TF it's either
// Tensor or IndexedSlices or None, which here we map to nullptr. Gradients need
// to allow their size to be computed and they need to be passable to a backward
// function and deleted (as the backprop code creates lots of gradients the user
// is not interested in).
//

                                             &input_tensors, &input_nodes);
  if (TF_GetCode(status) != TF_OK) return nullptr;

  // Process outputs.
  std::vector<tensorflow::OutputTensor> output_tensors;
  status->status = tensorflow::ProcessOutputs(fn_body, fn_name, noutputs,
                                              outputs, &output_tensors);
  if (TF_GetCode(status) != TF_OK) return nullptr;

  // Process output names.

      TFE_NewOp(context, "DestroyResourceOp", status), &TFE_DeleteOp);
  if (TF_GetCode(status) != TF_OK) return;
  TFE_OpAddInput(op.get(), handle_, status);
  if (TF_GetCode(status) != TF_OK) return;
  const char* device = TFE_TensorHandleDeviceName(handle_, status);
  if (TF_GetCode(status) != TF_OK) return;
  TFE_OpSetDevice(op.get(), device, status);
  if (TF_GetCode(status) != TF_OK) return;
  int num_retvals = 0;