int num_values) = 0;
  virtual Status SetAttrTypeList(const char* attr_name, const DataType* values,
                                 int num_values) = 0;
  virtual Status SetAttrBoolList(const char* attr_name,
                                 const unsigned char* values,
                                 int num_values) = 0;
  virtual Status SetAttrShapeList(const char* attr_name, const int64_t** dims,

  /// found, call `compute_func`. If `compute_func` set `status` to `TF_OK`,
  /// store a copy of the output parameter in the cache, and another copy in
  /// `value`.
  void LookupOrCompute(const std::string& key, T* value,
                       const ComputeFunc& compute_func, TF_Status* status) {
    if (max_age_ == 0) {
      return compute_func(key, value, status);
    }

                         const void* const* values, const size_t* lengths,
                         int num_values, ForwardOperation*);
Status SetAttrFloatList(AbstractOperation*, const char* attr_name,
                        const float* values, int num_values, ForwardOperation*);
Status SetAttrIntList(AbstractOperation*, const char* attr_name,
                      const int64_t* values, int num_values, ForwardOperation*);

  TF_Tensor* output_tensor(int i) { return output_values_[i]; }

  TF_Session* mutable_session() { return session_; }

 private:
  void DeleteInputValues();
  void ResetOutputValues();

  TF_Session* session_;
  std::vector<TF_Output> inputs_;
  std::vector<TF_Tensor*> input_values_;
  std::vector<TF_Output> outputs_;
  std::vector<TF_Tensor*> output_values_;
  std::vector<TF_Operation*> targets_;
};

// read current values from streamz counters defined in other modules.
//
// The code under test will have created streamz counters like this:
// auto* streamz = tensorflow::monitoring::Counter<1>::New("name",
// "description", "label");
// and then incremented that counter for various values of label:
// streamz->GetCell("label-value")->IncrementBy(1);
//

  TF_DeleteTensor(t);
  TF_DeleteStatus(status);
  return th;
}

// Return a scalar tensor handle with given values.
template <class T, TF_DataType datatype>
TFE_TensorHandle* TestScalarTensorHandle(TFE_Context* ctx, const T value) {
  T data[] = {value};
  return TestTensorHandleWithDims<T, datatype>(ctx, data, nullptr, 0);
}

// Return a tensor handle containing a 100x100 matrix of floats

TF_CAPI_EXPORT extern uint64_t TF_NowMicros(void);

// Returns the number of seconds since the Unix epoch.
TF_CAPI_EXPORT extern uint64_t TF_NowSeconds(void);

// Populates a TF_ThreadOptions struct with system-default values.
TF_CAPI_EXPORT extern void TF_DefaultThreadOptions(TF_ThreadOptions* options);

// Returns a new thread that is running work_func and is identified
// (for debugging/performance-analysis) by thread_name.
//

  // Returns a Boolean hint indicating whether callers should prefer
  // `SummarizeValue` to resolving this handle and formatting the tensor.
  //
  // For example some tensor handles may represent distributed values, in which
  // case placement information is lost when resolving the handle.
  //
  // If false, a caller might implement pretty-printing by resolving and

  *tensor =
      unwrap(TF_CreateAbstractTensorFromEagerTensor(input_eager, status.get()));
  return absl::OkStatus();
}

// Return a scalar tensor handle with given value.
template <class T, TF_DataType datatype>
Status TestScalarTensorHandle(AbstractContext* ctx, const T value,
                              AbstractTensorHandle** tensor) {
  std::unique_ptr<TF_Status, decltype(&TF_DeleteStatus)> status(
      TF_NewStatus(), TF_DeleteStatus);

  std::vector<TF_Operation*> return_nodes;
  std::vector<const char*> missing_unused_key_names;
  std::vector<int> missing_unused_key_indexes;

  // Backing memory for missing_unused_key_names values.
  std::vector<tensorflow::string> missing_unused_key_names_data;
};

struct TF_DeviceList {
  std::vector<tensorflow::DeviceAttributes> response;
};

struct TF_Function {