deletion_vectors.hpp

Go to the documentation of this file.

/*
 * SPDX-FileCopyrightText: Copyright (c) 2025, NVIDIA CORPORATION.
 * SPDX-License-Identifier: Apache-2.0
 */
 
#pragma once
 
#include <cudf/io/parquet.hpp>
#include <cudf/io/types.hpp>
#include <cudf/types.hpp>
#include <cudf/utilities/export.hpp>
 
#include <queue>
 
namespace CUDF_EXPORT cudf {
namespace io::parquet::experimental {
 
class chunked_parquet_reader {
 public:
  struct roaring_bitmap_impl;
 
  chunked_parquet_reader(
    std::size_t chunk_read_limit,
    parquet_reader_options const& options,
    cudf::host_span<cuda::std::byte const> serialized_roaring64,
    cudf::host_span<size_t const> row_group_offsets,
    cudf::host_span<size_type const> row_group_num_rows,
    rmm::cuda_stream_view stream      = cudf::get_default_stream(),
    rmm::device_async_resource_ref mr = cudf::get_current_device_resource_ref());
 
  chunked_parquet_reader(
    std::size_t chunk_read_limit,
    std::size_t pass_read_limit,
    parquet_reader_options const& options,
    cudf::host_span<cuda::std::byte const> serialized_roaring64,
    cudf::host_span<size_t const> row_group_offsets,
    cudf::host_span<size_type const> row_group_num_rows,
    rmm::cuda_stream_view stream      = cudf::get_default_stream(),
    rmm::device_async_resource_ref mr = cudf::get_current_device_resource_ref());
 
  ~chunked_parquet_reader();
 
  [[nodiscard]] bool has_next() const;
 
  [[nodiscard]] table_with_metadata read_chunk();
 
 private:
  std::unique_ptr<cudf::io::chunked_parquet_reader> _reader;
  std::queue<size_t> _row_group_row_offsets;
  std::queue<size_type> _row_group_row_counts;
  std::unique_ptr<roaring_bitmap_impl> _deletion_vector;
  size_t _start_row;
  bool _is_unspecified_row_group_data;
  rmm::cuda_stream_view _stream;
  rmm::device_async_resource_ref _mr;
  rmm::device_async_resource_ref _table_mr;
};
 
table_with_metadata read_parquet(
  parquet_reader_options const& options,
  cudf::host_span<cuda::std::byte const> serialized_roaring64,
  cudf::host_span<size_t const> row_group_offsets,
  cudf::host_span<size_type const> row_group_num_rows,
  rmm::cuda_stream_view stream      = cudf::get_default_stream(),
  rmm::device_async_resource_ref mr = rmm::mr::get_current_device_resource_ref());
  // end of group
 
}  // namespace io::parquet::experimental
}  // namespace CUDF_EXPORT cudf