Merge branch 'branch-23.04' into fea-logger

conda/environments/all_cuda-118_arch-x86_64.yaml

@@ -42,13 +42,13 @@ dependencies:
 - ninja
 - notebook
 - numba>=0.56.2
-- numpy
+- numpy>=1.21
 - numpydoc
 - nvcc_linux-64=11.8
 - nvtx>=0.2.1
 - packaging
-- pandas>=1.0,<1.6.0dev0
-- pandoc<=2.0.0
+- pandas>=1.3,<1.6.0dev0
+- pandoc
 - pip
 - pre-commit
 - protobuf>=4.21.6,<4.22

conda/recipes/cudf/meta.yaml

@@ -62,10 +62,10 @@ requirements:
     - protobuf >=4.21.6,<4.22
     - python
     - typing_extensions
-    - pandas >=1.0,<1.6.0dev0
+    - pandas >=1.3,<1.6.0dev0
     - cupy >=9.5.0,<12.0.0a0
     - numba >=0.56.2
-    - numpy
+    - numpy >=1.21
     - {{ pin_compatible('pyarrow', max_pin='x.x.x') }}
     - libcudf {{ version }}
     - fastavro >=0.22.0

cpp/benchmarks/common/generate_input.cu

@@ -716,7 +716,7 @@ std::unique_ptr<cudf::column> create_random_column<cudf::list_view>(data_profile
       num_rows,
       std::move(offsets_column),
       std::move(current_child_column),
-      profile.get_null_probability().has_value() ? null_count : 0,  // cudf::UNKNOWN_NULL_COUNT,
+      profile.get_null_probability().has_value() ? null_count : 0,
       profile.get_null_probability().has_value() ? std::move(null_mask) : rmm::device_buffer{});
   }
   return list_column;  // return the top-level column

cpp/include/cudf/column/column.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2019-2022, NVIDIA CORPORATION.
+ * Copyright (c) 2019-2023, NVIDIA CORPORATION.
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -259,15 +259,15 @@ class column {
    * @brief Returns a reference to the specified child
    *
    * @param child_index Index of the desired child
-   * @return column& Reference to the desired child
+   * @return Reference to the desired child
    */
   column& child(size_type child_index) noexcept { return *_children[child_index]; };
 
   /**
    * @brief Returns a const reference to the specified child
    *
    * @param child_index Index of the desired child
-   * @return column const& Const reference to the desired child
+   * @return Const reference to the desired child
    */
   [[nodiscard]] column const& child(size_type child_index) const noexcept
   {
@@ -306,7 +306,7 @@ class column {
    * @brief Creates an immutable, non-owning view of the column's data and
    * children.
    *
-   * @return column_view The immutable, non-owning view
+   * @return The immutable, non-owning view
    */
   [[nodiscard]] column_view view() const;
 
@@ -316,7 +316,7 @@ class column {
    * This allows passing a `column` object directly into a function that
    * requires a `column_view`. The conversion is automatic.
    *
-   * @return column_view Immutable, non-owning `column_view`
+   * @return Immutable, non-owning `column_view`
    */
   operator column_view() const { return this->view(); };
 
@@ -330,7 +330,7 @@ class column {
    * if not, the null count will be recomputed on the next invocation of
    *`null_count()`.
    *
-   * @return mutable_column_view The mutable, non-owning view
+   * @return The mutable, non-owning view
    */
   mutable_column_view mutable_view();
 
@@ -346,7 +346,7 @@ class column {
    * Otherwise, the null count will be recomputed on the next invocation of
    * `null_count()`.
    *
-   * @return mutable_column_view Mutable, non-owning `mutable_column_view`
+   * @return Mutable, non-owning `mutable_column_view`
    */
   operator mutable_column_view() { return this->mutable_view(); };
 

cpp/include/cudf/column/column_device_view.cuh

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2019-2022, NVIDIA CORPORATION.
+ * Copyright (c) 2019-2023, NVIDIA CORPORATION.
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -129,7 +129,7 @@ class alignas(16) column_device_view_base {
    * or `std::is_same_v<T,void>` are true.
    *
    * @tparam The type to cast to
-   * @return T const* Typed pointer to underlying data
+   * @return Typed pointer to underlying data
    */
   template <typename T = void,
             CUDF_ENABLE_IF(std::is_same_v<T, void> or is_rep_layout_compatible<T>())>
@@ -151,7 +151,7 @@ class alignas(16) column_device_view_base {
    * false.
    *
    * @tparam T The type to cast to
-   * @return T const* Typed pointer to underlying data, including the offset
+   * @return Typed pointer to underlying data, including the offset
    */
   template <typename T, CUDF_ENABLE_IF(is_rep_layout_compatible<T>())>
   [[nodiscard]] CUDF_HOST_DEVICE T const* data() const noexcept
@@ -990,7 +990,7 @@ class alignas(16) mutable_column_device_view : public detail::column_device_view
    * `data<T>()`.
    *
    * @tparam The type to cast to
-   * @return T* Typed pointer to underlying data
+   * @return Typed pointer to underlying data
    */
   template <typename T = void,
             CUDF_ENABLE_IF(std::is_same_v<T, void> or is_rep_layout_compatible<T>())>
@@ -1009,7 +1009,7 @@ class alignas(16) mutable_column_device_view : public detail::column_device_view
    * @note If `offset() == 0`, then `head<T>() == data<T>()`
    *
    * @tparam T The type to cast to
-   * @return T* Typed pointer to underlying data, including the offset
+   * @return Typed pointer to underlying data, including the offset
    */
   template <typename T, CUDF_ENABLE_IF(is_rep_layout_compatible<T>())>
   CUDF_HOST_DEVICE T* data() const noexcept
@@ -1078,7 +1078,7 @@ class alignas(16) mutable_column_device_view : public detail::column_device_view
    * `mutable_column_device_view::has_element_accessor<T>()` is false.
    *
    * @tparam T The desired type
-   * @return T* Pointer to the first element after casting
+   * @return Pointer to the first element after casting
    */
   template <typename T, CUDF_ENABLE_IF(mutable_column_device_view::has_element_accessor<T>())>
   iterator<T> begin()
@@ -1094,7 +1094,7 @@ class alignas(16) mutable_column_device_view : public detail::column_device_view
    * `mutable_column_device_view::has_element_accessor<T>()` is false.
    *
    * @tparam T The desired type
-   * @return T const* Pointer to one past the last element after casting
+   * @return Pointer to one past the last element after casting
    */
   template <typename T, CUDF_ENABLE_IF(mutable_column_device_view::has_element_accessor<T>())>
   iterator<T> end()
@@ -1106,7 +1106,7 @@ class alignas(16) mutable_column_device_view : public detail::column_device_view
    * @brief Returns the specified child
    *
    * @param child_index The index of the desired child
-   * @return column_view The requested child `column_view`
+   * @return The requested child `column_view`
    */
   [[nodiscard]] __device__ mutable_column_device_view child(size_type child_index) const noexcept
   {
@@ -1173,7 +1173,7 @@ class alignas(16) mutable_column_device_view : public detail::column_device_view
    * device view of the specified column and it's children.
    *
    * @param source_view The `column_view` to use for this calculation.
-   * @return size_t The size in bytes of the amount of memory needed to hold a
+   * @return The size in bytes of the amount of memory needed to hold a
    * device view of the specified column and it's children
    */
   static std::size_t extent(mutable_column_view source_view);

cpp/include/cudf/column/column_view.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2019-2022, NVIDIA CORPORATION.
+ * Copyright (c) 2019-2023, NVIDIA CORPORATION.
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -66,7 +66,7 @@ class column_view_base {
    * or `std::is_same_v<T,void>` are true.
    *
    * @tparam The type to cast to
-   * @return T const* Typed pointer to underlying data
+   * @return Typed pointer to underlying data
    */
   template <typename T = void,
             CUDF_ENABLE_IF(std::is_same_v<T, void> or is_rep_layout_compatible<T>())>
@@ -85,7 +85,7 @@ class column_view_base {
    * false.
    *
    * @tparam T The type to cast to
-   * @return T const* Typed pointer to underlying data, including the offset
+   * @return Typed pointer to underlying data, including the offset
    */
   template <typename T, CUDF_ENABLE_IF(is_rep_layout_compatible<T>())>
   T const* data() const noexcept
@@ -101,7 +101,7 @@ class column_view_base {
    * false.
    *
    * @tparam T The desired type
-   * @return T const* Pointer to the first element after casting
+   * @return Pointer to the first element after casting
    */
   template <typename T, CUDF_ENABLE_IF(is_rep_layout_compatible<T>())>
   T const* begin() const noexcept
@@ -117,7 +117,7 @@ class column_view_base {
    * false.
    *
    * @tparam T The desired type
-   * @return T const* Pointer to one past the last element after casting
+   * @return Pointer to one past the last element after casting
    */
   template <typename T, CUDF_ENABLE_IF(is_rep_layout_compatible<T>())>
   T const* end() const noexcept
@@ -389,7 +389,7 @@ class column_view : public detail::column_view_base {
    * @brief Returns the specified child
    *
    * @param child_index The index of the desired child
-   * @return column_view The requested child `column_view`
+   * @return The requested child `column_view`
    */
   [[nodiscard]] column_view child(size_type child_index) const noexcept
   {
@@ -553,7 +553,7 @@ class mutable_column_view : public detail::column_view_base {
    * column, and instead, accessing the elements should be done via `data<T>()`.
    *
    * @tparam The type to cast to
-   * @return T* Typed pointer to underlying data
+   * @return Typed pointer to underlying data
    */
   template <typename T = void,
             CUDF_ENABLE_IF(std::is_same_v<T, void> or is_rep_layout_compatible<T>())>
@@ -572,7 +572,7 @@ class mutable_column_view : public detail::column_view_base {
    * @note If `offset() == 0`, then `head<T>() == data<T>()`
    *
    * @tparam T The type to cast to
-   * @return T* Typed pointer to underlying data, including the offset
+   * @return Typed pointer to underlying data, including the offset
    */
   template <typename T, CUDF_ENABLE_IF(is_rep_layout_compatible<T>())>
   T* data() const noexcept
@@ -588,7 +588,7 @@ class mutable_column_view : public detail::column_view_base {
    * false.
    *
    * @tparam T The desired type
-   * @return T* Pointer to the first element after casting
+   * @return Pointer to the first element after casting
    */
   template <typename T, CUDF_ENABLE_IF(is_rep_layout_compatible<T>())>
   T* begin() const noexcept
@@ -604,7 +604,7 @@ class mutable_column_view : public detail::column_view_base {
    * false.
    *
    * @tparam T The desired type
-   * @return T* Pointer to one past the last element after casting
+   * @return Pointer to one past the last element after casting
    */
   template <typename T, CUDF_ENABLE_IF(is_rep_layout_compatible<T>())>
   T* end() const noexcept
@@ -639,7 +639,7 @@ class mutable_column_view : public detail::column_view_base {
    * @brief Returns a reference to the specified child
    *
    * @param child_index The index of the desired child
-   * @return mutable_column_view The requested child `mutable_column_view`
+   * @return The requested child `mutable_column_view`
    */
   [[nodiscard]] mutable_column_view child(size_type child_index) const noexcept
   {
@@ -670,7 +670,7 @@ class mutable_column_view : public detail::column_view_base {
   /**
    * @brief Converts a mutable view into an immutable view
    *
-   * @return column_view An immutable view of the mutable view's elements
+   * @return An immutable view of the mutable view's elements
    */
   operator column_view() const;
 
@@ -684,7 +684,7 @@ class mutable_column_view : public detail::column_view_base {
  * @brief Counts the number of descendants of the specified parent.
  *
  * @param parent The parent whose descendants will be counted
- * @return size_type The number of descendants of the parent
+ * @return The number of descendants of the parent
  */
 size_type count_descendants(column_view parent);
 

cpp/include/cudf/concatenate.hpp

@@ -40,8 +40,7 @@ namespace cudf {
  *
  * @param views host_span of column views whose bitmasks will be concatenated
  * @param mr Device memory resource used for allocating the new device_buffer
- * @return rmm::device_buffer A `device_buffer` containing the bitmasks of all
- * the column views in the views vector
+ * @return A `device_buffer` containing the bitmasks of all the column views in the views vector
  */
 rmm::device_buffer concatenate_masks(
   host_span<column_view const> views,

cpp/include/cudf/copying.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2018-2022, NVIDIA CORPORATION.
+ * Copyright (c) 2018-2023, NVIDIA CORPORATION.
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -78,7 +78,7 @@ enum class out_of_bounds_policy : bool {
  * better performance. If `policy` is set to `DONT_CHECK` and there are out-of-bounds indices
  * in the gather map, the behavior is undefined. Defaults to `DONT_CHECK`.
  * @param[in] mr Device memory resource used to allocate the returned table's device memory
- * @return std::unique_ptr<table> Result of the gather
+ * @return Result of the gather
  */
 std::unique_ptr<table> gather(
   table_view const& source_table,
@@ -211,15 +211,15 @@ enum class mask_allocation_policy {
  * @brief Initializes and returns an empty column of the same type as the `input`.
  *
  * @param[in] input Immutable view of input column to emulate
- * @return std::unique_ptr<column> An empty column of same type as `input`
+ * @return An empty column of same type as `input`
  */
 std::unique_ptr<column> empty_like(column_view const& input);
 
 /**
  * @brief Initializes and returns an empty column of the same type as the `input`.
  *
  * @param[in] input Scalar to emulate
- * @return std::unique_ptr<column> An empty column of same type as `input`
+ * @return An empty column of same type as `input`
  */
 std::unique_ptr<column> empty_like(scalar const& input);
 
@@ -264,7 +264,7 @@ std::unique_ptr<column> allocate_like(
  * memory for the column's data or bitmask.
  *
  * @param[in] input_table Immutable view of input table to emulate
- * @return std::unique_ptr<table> A table of empty columns with the same types as the columns in
+ * @return A table of empty columns with the same types as the columns in
  * `input_table`
  */
 std::unique_ptr<table> empty_like(table_view const& input_table);
@@ -333,7 +333,7 @@ void copy_range_in_place(column_view const& source,
  * (exclusive)
  * @param target_begin The starting index of the target range (inclusive)
  * @param mr Device memory resource used to allocate the returned column's device memory
- * @return std::unique_ptr<column> The result target column
+ * @return The result target column
  */
 std::unique_ptr<column> copy_range(
   column_view const& source,
@@ -920,7 +920,7 @@ std::unique_ptr<table> boolean_mask_scatter(
  * @param input Column view to get the element from
  * @param index Index into `input` to get the element at
  * @param mr Device memory resource used to allocate the returned scalar's device memory
- * @return std::unique_ptr<scalar> Scalar containing the single value
+ * @return Scalar containing the single value
  */
 std::unique_ptr<scalar> get_element(
   column_view const& input,
@@ -960,7 +960,7 @@ enum class sample_with_replacement : bool {
  * @param seed Seed value to initiate random number generator
  * @param mr Device memory resource used to allocate the returned table's device memory
  *
- * @return std::unique_ptr<table> Table containing samples from `input`
+ * @return Table containing samples from `input`
  */
 std::unique_ptr<table> sample(
   table_view const& input,

cpp/include/cudf/io/data_sink.hpp

@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2020-2022, NVIDIA CORPORATION.
+ * Copyright (c) 2020-2023, NVIDIA CORPORATION.
  *
  * Licensed under the Apache License, Version 2.0 (the "License");
  * you may not use this file except in compliance with the License.
@@ -123,7 +123,7 @@ class data_sink {
    * instead of write() when possible.  However, it is still possible to receive
    * write() calls as well.
    *
-   * @return bool If this writer supports device_write() calls
+   * @return If this writer supports device_write() calls
    */
   [[nodiscard]] virtual bool supports_device_write() const { return false; }
 
@@ -194,7 +194,7 @@ class data_sink {
   /**
    * @pure @brief Returns the total number of bytes written into this sink
    *
-   * @return size_t Total number of bytes written into this sink
+   * @return Total number of bytes written into this sink
    */
   virtual size_t bytes_written() = 0;
 };