Function TestStringDictionaryAppendIndices

cpp/src/arrow/array/array_dict_test.cc:429–460

template <typename BuilderType, typename IndexType, typename AppendCType>
void TestStringDictionaryAppendIndices() {
  auto index_type = TypeTraits<IndexType>::type_singleton();

  auto ex_dict = ArrayFromJSON(utf8(), R"(["c", "a", "b", "d"])");
  auto invalid_dict = ArrayFromJSON(binary(), R"(["e", "f"])");

  BuilderType builder;
  ASSERT_OK(builder.InsertMemoValues(*ex_dict));

  // Inserting again should have no effect
  ASSERT_OK(builder.InsertMemoValues(*ex_dict));

  // Type mismatch
  ASSERT_RAISES(Invalid, builder.InsertMemoValues(*invalid_dict));

  std::vector<AppendCType> raw_indices = {0, 1, 2, -1, 3};
  std::vector<uint8_t> is_valid = {1, 1, 1, 0, 1};
  for (int i = 0; i < 2; ++i) {
    ASSERT_OK(builder.AppendIndices(
        raw_indices.data(), static_cast<int64_t>(raw_indices.size()), is_valid.data()));
  }

  ASSERT_EQ(10, builder.length());

  std::shared_ptr<Array> result;
  ASSERT_OK(builder.Finish(&result));

  auto ex_indices = ArrayFromJSON(index_type, R"([0, 1, 2, null, 3, 0, 1, 2, null, 3])");
  auto dtype = dictionary(index_type, utf8());
  DictionaryArray expected(dtype, ex_indices, ex_dict);
  ASSERT_TRUE(expected.Equals(result));
}

TEST(TestStringDictionaryBuilder, AppendIndices) {
  // Currently AdaptiveIntBuilder only accepts int64_t in bulk appends