Replace unnecessary uses of UNKNOWN_NULL_COUNT (#13102)

This PR replaces uses of `cudf::UNKNOWN_NULL_COUNT` where the null count is either already known or trivially computed. Contributes to #11968 Authors: - Vyas Ramasubramani (https://github.com/vyasr) Approvers: - David Wendt (https://github.com/davidwendt) - Nghia Truong (https://github.com/ttnghia) URL: #13102
rapidsai · Apr 10, 2023 · cab6522 · cab6522
1 parent 30411b5
commit cab6522
Show file tree

Hide file tree

Showing 6 changed files with 12 additions and 16 deletions.
diff --git a/cpp/src/binaryop/compiled/binary_ops.cu b/cpp/src/binaryop/compiled/binary_ops.cu
@@ -80,8 +80,8 @@ scalar_as_column_view::return_type scalar_as_column_view::operator()<cudf::strin
   auto col_v = column_view(s.type(),
                            1,
                            nullptr,
-                           (bitmask_type const*)s.validity_data(),
-                           cudf::UNKNOWN_NULL_COUNT,
+                           reinterpret_cast<bitmask_type const*>(s.validity_data()),
+                           static_cast<size_type>(!s.is_valid(stream)),
                            0,
                            {offsets_column->view(), chars_column_v});
   return std::pair{col_v, std::move(offsets_column)};

diff --git a/cpp/src/copying/pack.cpp b/cpp/src/copying/pack.cpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2021-2022, NVIDIA CORPORATION.
+ * Copyright (c) 2021-2023, NVIDIA CORPORATION.
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -159,12 +159,8 @@ packed_columns::metadata pack_metadata(ColumnIter begin,
 
   // first metadata entry is a stub indicating how many total (top level) columns
   // there are
-  metadata.emplace_back(data_type{type_id::EMPTY},
-                        static_cast<size_type>(std::distance(begin, end)),
-                        UNKNOWN_NULL_COUNT,
-                        -1,
-                        -1,
-                        0);
+  metadata.emplace_back(
+    data_type{type_id::EMPTY}, static_cast<size_type>(std::distance(begin, end)), 0, -1, -1, 0);
 
   std::for_each(begin, end, [&metadata, &contiguous_buffer, &buffer_size](column_view const& col) {
     build_column_metadata(metadata, col, contiguous_buffer, buffer_size);

diff --git a/cpp/src/interop/from_arrow.cu b/cpp/src/interop/from_arrow.cu
@@ -284,7 +284,7 @@ std::unique_ptr<column> dispatch_to_cudf_column::operator()<cudf::string_view>(
   auto out_col        = make_strings_column(num_rows,
                                      std::move(offsets_column),
                                      std::move(chars_column),
-                                     UNKNOWN_NULL_COUNT,
+                                     array.null_count(),
                                      std::move(*get_mask_buffer(array, stream, mr)));
 
   return num_rows == array.length()
@@ -324,7 +324,7 @@ std::unique_ptr<column> dispatch_to_cudf_column::operator()<cudf::dictionary32>(
   return make_dictionary_column(std::move(keys_column),
                                 std::move(indices_column),
                                 std::move(*(column_contents.null_mask)),
-                                UNKNOWN_NULL_COUNT);
+                                array.null_count());
 }
 
 template <>
@@ -357,7 +357,7 @@ std::unique_ptr<column> dispatch_to_cudf_column::operator()<cudf::struct_view>(
   }
 
   return make_structs_column(
-    array.length(), move(child_columns), UNKNOWN_NULL_COUNT, std::move(out_mask), stream, mr);
+    array.length(), move(child_columns), array.null_count(), std::move(out_mask), stream, mr);
 }
 
 template <>
@@ -381,7 +381,7 @@ std::unique_ptr<column> dispatch_to_cudf_column::operator()<cudf::list_view>(
   auto out_col        = make_lists_column(num_rows,
                                    std::move(offsets_column),
                                    std::move(child_column),
-                                   UNKNOWN_NULL_COUNT,
+                                   array.null_count(),
                                    std::move(*get_mask_buffer(array, stream, mr)),
                                    stream,
                                    mr);

diff --git a/cpp/src/lists/copying/concatenate.cu b/cpp/src/lists/copying/concatenate.cu
@@ -14,7 +14,6 @@
  * limitations under the License.
  */
 
-#include <algorithm>
 #include <cudf/column/column.hpp>
 #include <cudf/column/column_device_view.cuh>
 #include <cudf/column/column_factories.hpp>
@@ -29,6 +28,7 @@
 
 #include <thrust/transform.h>
 
+#include <algorithm>
 #include <memory>
 
 namespace cudf {

diff --git a/cpp/src/lists/dremel.cu b/cpp/src/lists/dremel.cu
@@ -147,7 +147,7 @@ dremel_data get_encoding(column_view h_col,
                          col.size(),
                          col.head(),
                          col.null_mask(),
-                         UNKNOWN_NULL_COUNT,
+                         col.null_count(),
                          col.offset(),
                          std::move(children));
     };

diff --git a/cpp/src/transform/transform.cpp b/cpp/src/transform/transform.cpp
@@ -78,7 +78,7 @@ std::unique_ptr<column> transform(column_view const& input,
   CUDF_EXPECTS(is_fixed_width(input.type()), "Unexpected non-fixed-width type.");
 
   std::unique_ptr<column> output = make_fixed_width_column(
-    output_type, input.size(), copy_bitmask(input), cudf::UNKNOWN_NULL_COUNT, stream, mr);
+    output_type, input.size(), copy_bitmask(input), input.null_count(), stream, mr);
 
   if (input.is_empty()) { return output; }