Method GetInternal

cpp/src/parquet/decoder.cc:2049–2119

  }

  int GetInternal(ByteArray* buffer, int max_values) {
    // Decode up to `max_values` strings into an internal buffer
    // and reference them into `buffer`.
    max_values = std::min(max_values, num_valid_values_);
    if (max_values == 0) {
      return max_values;
    }

    int suffix_read = suffix_decoder_.Decode(buffer, max_values);
    if (ARROW_PREDICT_FALSE(suffix_read != max_values)) {
      ParquetException::EofException("Read " + std::to_string(suffix_read) +
                                     ", expecting " + std::to_string(max_values) +
                                     " from suffix decoder");
    }

    int64_t data_size = 0;
    const int32_t* prefix_len_ptr =
        buffered_prefix_length_->data_as<int32_t>() + prefix_len_offset_;
    for (int i = 0; i < max_values; ++i) {
      if (prefix_len_ptr[i] == 0) {
        // We don't need to copy the suffix if the prefix length is 0.
        continue;
      }
      if (ARROW_PREDICT_FALSE(prefix_len_ptr[i] < 0)) {
        throw ParquetException("negative prefix length in DELTA_BYTE_ARRAY");
      }
      if (buffer[i].len == 0 && i != 0) {
        // We don't need to copy the prefix if the suffix length is 0
        // and this is not the first run (that is, the prefix doesn't point
        // to the mutable `last_value_`).
        continue;
      }
      if (ARROW_PREDICT_FALSE(AddWithOverflow(data_size, prefix_len_ptr[i], &data_size) ||
                              AddWithOverflow(data_size, buffer[i].len, &data_size))) {
        throw ParquetException("excess expansion in DELTA_BYTE_ARRAY");
      }
    }
    PARQUET_THROW_NOT_OK(buffered_data_->Resize(data_size));

    std::string_view prefix{last_value_};
    uint8_t* data_ptr = buffered_data_->mutable_data();
    if (max_values > 0) {
      BuildBufferInternal</*is_first_run=*/true>(prefix_len_ptr, 0, buffer, &prefix,
                                                 &data_ptr);
    }
    for (int i = 1; i < max_values; ++i) {
      BuildBufferInternal</*is_first_run=*/false>(prefix_len_ptr, i, buffer, &prefix,
                                                  &data_ptr);
    }
    DCHECK_EQ(data_ptr - buffered_data_->mutable_data(), data_size);
    prefix_len_offset_ += max_values;
    this->num_values_ -= max_values;
    num_valid_values_ -= max_values;
    last_value_ = std::string{prefix};

    if (num_valid_values_ == 0) {
      last_value_in_previous_page_ = last_value_;
    }