Method TypedColumnWriterImpl

cpp/src/parquet/column_writer.cc:1286–1327

  using T = typename ParquetType::c_type;

  TypedColumnWriterImpl(ColumnChunkMetaDataBuilder* metadata,
                        std::unique_ptr<PageWriter> pager, const bool use_dictionary,
                        Encoding::type encoding, const WriterProperties* properties,
                        BloomFilter* bloom_filter)
      : ColumnWriterImpl(metadata, std::move(pager), use_dictionary, encoding,
                         properties) {
    current_encoder_ = MakeEncoder(ParquetType::type_num, encoding, use_dictionary,
                                   descr_, properties->memory_pool());
    // We have to dynamic_cast as some compilers don't want to static_cast
    // through virtual inheritance.
    current_value_encoder_ =
        dynamic_cast<TypedEncoder<ParquetType>*>(current_encoder_.get());
    // Will be null if not using dictionary, but that's ok
    current_dict_encoder_ =
        dynamic_cast<DictEncoder<ParquetType>*>(current_encoder_.get());

    if (bloom_filter != nullptr) {
      bloom_filter_writer_ = std::make_unique<BloomFilterWriter>(descr_, bloom_filter);
    }

    // GH-46205: Geometry/Geography are the first non-nested logical types to have a
    // SortOrder::UNKNOWN. Currently, the presence of statistics is tied to
    // having a known sort order and so null counts will be missing.
    if (properties->statistics_enabled(descr_->path())) {
      if (SortOrder::UNKNOWN != descr_->sort_order()) {
        page_statistics_ = MakeStatistics<ParquetType>(descr_, allocator_);
        chunk_statistics_ = MakeStatistics<ParquetType>(descr_, allocator_);
      }
      if (descr_->logical_type() != nullptr && descr_->logical_type()->is_geometry()) {
        chunk_geospatial_statistics_ = std::make_shared<geospatial::GeoStatistics>();
      }
    }

    if (properties->size_statistics_level() == SizeStatisticsLevel::ColumnChunk ||
        properties->size_statistics_level() == SizeStatisticsLevel::PageAndColumnChunk) {
      page_size_statistics_ = SizeStatistics::Make(descr_);
      chunk_size_statistics_ = SizeStatistics::Make(descr_);
    }
    pages_change_on_record_boundaries_ =
        properties->data_page_version() == ParquetDataPageVersion::V2 ||
        properties->page_index_enabled(descr_->path());
  }

  int64_t Close() override { return ColumnWriterImpl::Close(); }