Remove orc::gpu namespace (rapidsai#17891)

Moving forward with removal of the (redundant) `gpu` namespace in cuIO. Also moved the entire ORC implementation to `cudf::io::orc::detail`, leaving only the implementation of the public API in `cudf::io::orc`. Also removed a few unused headers, or moved them to be included in the right files. Authors: - Vukasin Milovanovic (https://github.com/vuule) Approvers: - Yunsong Wang (https://github.com/PointKernel) - Muhammad Haseeb (https://github.com/mhaseeb123) URL: rapidsai#17891
kingcrimsontianyu · Feb 1, 2025 · a6acba0 · a6acba0
1 parent aeb90de
commit a6acba0
Show file tree

Hide file tree

Showing 19 changed files with 282 additions and 329 deletions.
diff --git a/cpp/include/cudf/io/orc_metadata.hpp b/cpp/include/cudf/io/orc_metadata.hpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2019-2024, NVIDIA CORPORATION.
+ * Copyright (c) 2019-2025, NVIDIA CORPORATION.
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -171,12 +171,12 @@ using statistics_type = std::variant<no_statistics,
                                      timestamp_statistics>;
 
 //! Orc I/O interfaces
-namespace orc {
+namespace orc::detail {
 // forward declare the type that ProtobufReader uses. The `cudf::io::column_statistics` objects,
 // returned from `read_parsed_orc_statistics`, are constructed from
-// `cudf::io::orc::column_statistics` objects that `ProtobufReader` initializes.
+// `cudf::io::orc::detail::column_statistics` objects that `ProtobufReader` initializes.
 struct column_statistics;
-}  // namespace orc
+}  // namespace orc::detail
 
 /**
  * @brief Contains per-column ORC statistics.
@@ -194,7 +194,7 @@ struct column_statistics {
    *
    * @param detail_statistics The statistics to initialize the object with
    */
-  column_statistics(orc::column_statistics&& detail_statistics);
+  column_statistics(orc::detail::column_statistics&& detail_statistics);
 };
 
 /**

diff --git a/cpp/src/io/functions.cpp b/cpp/src/io/functions.cpp
@@ -292,7 +292,7 @@ raw_orc_statistics read_raw_orc_statistics(source_info const& src_info,
     CUDF_FAIL("Unsupported source type");
   }
 
-  orc::metadata const metadata(source.get(), stream);
+  orc::detail::metadata const metadata(source.get(), stream);
 
   // Initialize statistics to return
   raw_orc_statistics result;
@@ -318,7 +318,7 @@ raw_orc_statistics read_raw_orc_statistics(source_info const& src_info,
   return result;
 }
 
-column_statistics::column_statistics(orc::column_statistics&& cs)
+column_statistics::column_statistics(orc::detail::column_statistics&& cs)
 {
   number_of_values = cs.number_of_values;
   has_null         = cs.has_null;
@@ -350,9 +350,9 @@ parsed_orc_statistics read_parsed_orc_statistics(source_info const& src_info,
   result.column_names = raw_stats.column_names;
 
   auto parse_column_statistics = [](auto const& raw_col_stats) {
-    orc::column_statistics stats_internal;
-    orc::ProtobufReader(reinterpret_cast<uint8_t const*>(raw_col_stats.c_str()),
-                        raw_col_stats.size())
+    orc::detail::column_statistics stats_internal;
+    orc::detail::ProtobufReader(reinterpret_cast<uint8_t const*>(raw_col_stats.c_str()),
+                                raw_col_stats.size())
       .read(stats_internal);
     return column_statistics(std::move(stats_internal));
   };
@@ -373,7 +373,7 @@ parsed_orc_statistics read_parsed_orc_statistics(source_info const& src_info,
   return result;
 }
 namespace {
-orc_column_schema make_orc_column_schema(host_span<orc::SchemaType const> orc_schema,
+orc_column_schema make_orc_column_schema(host_span<orc::detail::SchemaType const> orc_schema,
                                          uint32_t column_id,
                                          std::string column_name)
 {
@@ -400,7 +400,7 @@ orc_metadata read_orc_metadata(source_info const& src_info, rmm::cuda_stream_vie
   auto sources = make_datasources(src_info);
 
   CUDF_EXPECTS(sources.size() == 1, "Only a single source is currently supported.");
-  auto const footer = orc::metadata(sources.front().get(), stream).ff;
+  auto const footer = orc::detail::metadata(sources.front().get(), stream).ff;
 
   return {{make_orc_column_schema(footer.types, 0, "")},
           footer.numberOfRows,

diff --git a/cpp/src/io/orc/dict_enc.cu b/cpp/src/io/orc/dict_enc.cu
@@ -24,7 +24,7 @@
 
 #include <rmm/cuda_stream_view.hpp>
 
-namespace cudf::io::orc::gpu {
+namespace cudf::io::orc::detail {
 
 /**
  * @brief Counts the number of characters in each rowgroup of each string column.
@@ -266,4 +266,4 @@ void get_dictionary_indices(device_2dspan<stripe_dictionary> dictionaries,
     <<<dictionaries.count(), block_size, 0, stream.value()>>>(dictionaries, columns);
 }
 
-}  // namespace cudf::io::orc::gpu
+}  // namespace cudf::io::orc::detail
diff --git a/cpp/src/io/orc/orc.cpp b/cpp/src/io/orc/orc.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2019-2024, NVIDIA CORPORATION.
+ * Copyright (c) 2019-2025, NVIDIA CORPORATION.
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -16,6 +16,7 @@
 
 #include "orc.hpp"
 
+#include "io/comp/io_uncomp.hpp"
 #include "orc_field_reader.hpp"
 #include "orc_field_writer.hpp"
 
@@ -25,7 +26,7 @@
 
 #include <string>
 
-namespace cudf::io::orc {
+namespace cudf::io::orc::detail {
 
 namespace {
 [[nodiscard]] constexpr uint32_t varint_size(uint64_t val)
@@ -496,7 +497,7 @@ metadata::metadata(datasource* const src, rmm::cuda_stream_view stream) : source
   buffer =
     source->host_read(len - ps_length - 1 - ps.footerLength - ps.metadataLength, ps.metadataLength);
   auto const md_data = decompressor->decompress_blocks({buffer->data(), buffer->size()}, stream);
-  orc::ProtobufReader(md_data.data(), md_data.size()).read(md);
+  ProtobufReader(md_data.data(), md_data.size()).read(md);
 
   init_parent_descriptors();
   init_column_names();
@@ -546,4 +547,4 @@ void metadata::init_parent_descriptors()
   }
 }
 
-}  // namespace cudf::io::orc
+}  // namespace cudf::io::orc::detail
diff --git a/cpp/src/io/orc/orc.hpp b/cpp/src/io/orc/orc.hpp
@@ -16,8 +16,6 @@
 
 #pragma once
 
-#include "io/comp/io_uncomp.hpp"
-
 #include <cudf/column/column_device_view.cuh>
 #include <cudf/io/datasource.hpp>
 #include <cudf/io/orc_metadata.hpp>
@@ -34,9 +32,7 @@
 #include <string>
 #include <vector>
 
-namespace cudf {
-namespace io {
-namespace orc {
+namespace cudf::io::orc::detail {
 
 static constexpr uint32_t block_header_size = 3;
 // Seconds from January 1st, 1970 to January 1st, 2015
@@ -710,6 +706,4 @@ struct rowgroup_rows {
   [[nodiscard]] CUDF_HOST_DEVICE constexpr auto size() const noexcept { return end - begin; }
 };
 
-}  // namespace orc
-}  // namespace io
-}  // namespace cudf
+}  // namespace cudf::io::orc::detail
diff --git a/cpp/src/io/orc/orc_field_reader.hpp b/cpp/src/io/orc/orc_field_reader.hpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2020-2024, NVIDIA CORPORATION.
+ * Copyright (c) 2020-2025, NVIDIA CORPORATION.
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -17,17 +17,15 @@
 
 #include "orc.hpp"
 
-#include <string>
+#include <cstdint>
 
 /**
  * @file orc_field_reader.hpp
  * @brief Functors to encapsulate common functionality required to implement
  * ProtobufReader::read(...) functions
  */
 
-namespace cudf {
-namespace io {
-namespace orc {
+namespace cudf::io::orc::detail {
 
 /**
  * @brief Functor to run an operator for a specified field.
@@ -90,6 +88,4 @@ inline void ProtobufReader::function_builder(T& s, size_t maxlen, std::tuple<Ope
   CUDF_EXPECTS(m_cur <= end, "Current pointer to metadata stream is out of bounds");
 }
 
-}  // namespace orc
-}  // namespace io
-}  // namespace cudf
+}  // namespace cudf::io::orc::detail
diff --git a/cpp/src/io/orc/orc_field_writer.hpp b/cpp/src/io/orc/orc_field_writer.hpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2020-2024, NVIDIA CORPORATION.
+ * Copyright (c) 2020-2025, NVIDIA CORPORATION.
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -26,9 +26,7 @@
  * `ProtobufWriter::write(...)` functions
  */
 
-namespace cudf {
-namespace io {
-namespace orc {
+namespace cudf::io::orc::detail {
 
 struct ProtobufWriter::ProtobufFieldWriter {
   int struct_size{0};
@@ -129,6 +127,4 @@ struct ProtobufWriter::ProtobufFieldWriter {
   size_t value() { return struct_size; }
 };
 
-}  // namespace orc
-}  // namespace io
-}  // namespace cudf
+}  // namespace cudf::io::orc::detail
diff --git a/cpp/src/io/orc/orc_gpu.hpp b/cpp/src/io/orc/orc_gpu.hpp
@@ -18,7 +18,6 @@
 
 #include "io/comp/comp.hpp"
 #include "io/statistics/statistics.cuh"
-#include "io/utilities/column_buffer.hpp"
 #include "orc.hpp"
 
 #include <cudf/detail/cuco_helpers.hpp>
@@ -33,10 +32,7 @@
 
 #include <cuco/static_map.cuh>
 
-namespace cudf {
-namespace io {
-namespace orc {
-namespace gpu {
+namespace cudf::io::orc::detail {
 
 using cudf::detail::device_2dspan;
 using cudf::detail::host_2dspan;
@@ -65,9 +61,7 @@ auto constexpr VALUE_SENTINEL = size_type{-1};
 struct CompressedStreamInfo {
   CompressedStreamInfo() = default;
   explicit constexpr CompressedStreamInfo(uint8_t const* compressed_data_, size_t compressed_size_)
-    : compressed_data(compressed_data_),
-      uncompressed_data(nullptr),
-      compressed_data_size(compressed_size_)
+    : compressed_data(compressed_data_), compressed_data_size(compressed_size_)
   {
   }
   uint8_t const* compressed_data{};  // [in] base ptr to compressed stream data
@@ -500,7 +494,4 @@ void reduce_pushdown_masks(device_span<orc_column_device_view const> orc_columns
                            device_2dspan<cudf::size_type> set_counts,
                            rmm::cuda_stream_view stream);
 
-}  // namespace gpu
-}  // namespace orc
-}  // namespace io
-}  // namespace cudf
+}  // namespace cudf::io::orc::detail