const void* value, size_t length) {
  tensorflow::StringPiece s(static_cast<const char*>(value), length);
  desc->node_builder.Attr(attr_name, s);
}

void TF_SetAttrStringList(TF_OperationDescription* desc, const char* attr_name,
                          const void* const* values, const size_t* lengths,
                          int num_values) {

    // more than fits in a 32-bit integer.
    size_t requested_read_length;
    if (n > INT32_MAX)
      requested_read_length = INT32_MAX;
    else
      requested_read_length = n;

    // `pread` returns a `ssize_t` on POSIX, but due to interface being
    // cross-platform, return type of `Read` is `int64_t`.
    int64_t r = int64_t{pread(posix_file->fd, dst, requested_read_length,
                              static_cast<off_t>(offset))};

                         int num_values, ForwardOperation* forward_op_) {
  std::vector<StringPiece> v(num_values);
  for (int i = 0; i < num_values; ++i) {
    v[i] = StringPiece(static_cast<const char*>(values[i]), lengths[i]);
  }
  forward_op_->attrs.Set(attr_name, v);
  return op_->SetAttrStringList(attr_name, values, lengths, num_values);
}

  } else {
    shape_string = "<error computing shape>";
  }
  std::string value_string;
  if (!SummarizeValue(value_string).ok()) {
    value_string = "<error computing value>";
  }
  if (value_string.length() > 100) {
    // The default NumPy-style output can be distractingly long in error
    // messages.
    value_string = absl::StrCat(value_string.substr(0, 100), " [...]");
  }
  Status s;

      } else {
        // Generalize differing axis lengths to "variable"/"unknown".
        for (int axis_index = 0; axis_index < combined_shape.dims();
             ++axis_index) {
          int64_t axis_length = combined_shape.dim_size(axis_index);
          if (axis_length != component_shape.dim_size(axis_index)) {
            axis_length = -1;
          }
          TF_RETURN_IF_ERROR(

  ASSERT_EQ(errors::OK, s.code()) << s.message();
  string message = "This is the way!";
  s = SetAttrString(check_numerics_op.get(), "message", message.data(),
                    message.length(), &forward_op);
  ASSERT_EQ(errors::OK, s.code()) << s.message();
  int num_retvals = 1;
  std::vector<AbstractTensorHandle*> outputs(1);
  GradientRegistry registry;
  s = RegisterGradients(&registry);

                           const size_t* lengths, int num_values) override {
    if (strcmp(attr_name, tensorflow::kColocationAttrName) == 0) {
      op_->colocation_constraints.clear();
      for (int i = 0; i < num_values; ++i) {
        op_->colocation_constraints.emplace(static_cast<const char*>(values[i]),
                                            lengths[i]);
      }
    } else {

        values_vector.reserve(s_size);
        absl::InlinedVector<size_t, 4> lengths_vector;
        lengths_vector.reserve(s_size);
        for (int i = 0; i < s_size; ++i) {
          const string& v = default_value.list().s(i);
          values_vector.push_back(v.data());
          lengths_vector.push_back(v.size());
        }
        TFE_OpSetAttrStringList(op, attr_name, values_vector.data(),

  tensorflow::string value = cell->cell.value();
  void* data = tensorflow::port::Malloc(value.length());
  value.copy(static_cast<char*>(data), value.length(), 0);
  buf->data = data;
  buf->length = value.length();
  buf->data_deallocator = [](void* data, size_t length) {
    tensorflow::port::Free(data);
  };
}

TFE_MonitoringStringGauge0* TFE_MonitoringNewStringGauge0(

      (TF_GetCode(status) != TF_OUT_OF_RANGE)) {
    return -1;
  }
  int64_t read;
  auto content_length = stream.headers().find("content-length");
  if (content_length == stream.headers().end()) {
    // When we read a file with offset that is bigger than the actual file size.
    // GCS will return an empty header (e.g no `content-length` header). In this
    // case, we will set read to `0` and continue.
    read = 0;