Merge remote-tracking branch 'upstream/main' into ci/cudf_polars/polars_tests_rapidsmpf

Author: mroeschke

cpp/CMakeLists.txt

@@ -536,6 +536,7 @@ add_library(
   src/io/parquet/experimental/hybrid_scan_preprocess.cu
   src/io/parquet/experimental/page_index_filter.cu
   src/io/parquet/experimental/page_index_filter_utils.cu
+  src/io/parquet/expression_transform_helpers.cpp
   src/io/parquet/io_utils/parquet_io_utils.cpp
   src/io/parquet/page_data.cu
   src/io/parquet/chunk_dict.cu

cpp/benchmarks/io/cuio_common.cpp

@@ -274,9 +274,8 @@ void drop_page_cache_if_enabled(std::vector<std::string> const& file_paths)
   }
 
   if (not is_file_scope) {
-    if (not kvikio::drop_system_page_cache()) {
-      CUDF_FAIL("Failed to execute the drop cache command");
-    }
+    auto const is_cache_dropped = kvikio::drop_system_page_cache();
+    CUDF_EXPECTS(is_cache_dropped, "Failed to execute the drop cache command");
     return;
   }
 

cpp/doxygen/developer_guide/DEVELOPER_GUIDE.md

@@ -1142,6 +1142,19 @@ occurred and is used for the exception's `what()` message. If the conditional ev
 `false`, then an error has occurred and an instance of the exception class in the third argument
 (or the default, `cudf::logic_error`) is thrown.
 
+The condition (first argument) of `CUDF_EXPECTS` should be a pure predicate that only inspects
+state without modifying it. If the result of an operation with a side effect needs to be checked,
+capture the result in a variable first:
+
+```c++
+// WRONG — side effect (initialization) inside the condition:
+CUDF_EXPECTS(reader.init(source), "Failed to initialize reader");
+
+// RIGHT — capture the result, then check it:
+auto const init_ok = reader.init(source);
+CUDF_EXPECTS(init_ok, "Failed to initialize reader");
+```
+
 There are times where a particular code path, if reached, should indicate an error no matter what.
 For example, often the `default` case of a `switch` statement represents an invalid alternative.
 Use the `CUDF_FAIL` macro for such errors. This is effectively the same as calling
@@ -1153,6 +1166,11 @@ Example:
 CUDF_FAIL("This code path should not be reached.");
 ```
 
+Prefer `CUDF_EXPECTS` over `if (condition) { CUDF_FAIL(reason); }` when the condition has no side
+effects. The `if`/`CUDF_FAIL` pattern is appropriate when cleanup or other actions must be
+performed before throwing, or when the condition itself involves side effects that cannot be
+separated from the check.
+
 ### CUDA Error Checking
 
 Use the `CUDF_CUDA_TRY` macro to check for the successful completion of CUDA runtime API functions.

cpp/include/cudf/dictionary/detail/concatenate.hpp

@@ -1,5 +1,5 @@
 /*
- * SPDX-FileCopyrightText: Copyright (c) 2020-2024, NVIDIA CORPORATION.
+ * SPDX-FileCopyrightText: Copyright (c) 2020-2026, NVIDIA CORPORATION.
  * SPDX-License-Identifier: Apache-2.0
  */
 #pragma once
@@ -12,7 +12,7 @@
 
 #include <rmm/cuda_stream_view.hpp>
 
-namespace CUDF_EXPORT cudf {
+namespace cudf {
 namespace dictionary::detail {
 /**
  * @brief Returns a single column by vertically concatenating the given vector of
@@ -31,4 +31,4 @@ std::unique_ptr<column> concatenate(host_span<column_view const> columns,
                                     rmm::device_async_resource_ref mr);
 
 }  // namespace dictionary::detail
-}  // namespace CUDF_EXPORT cudf
+}  // namespace cudf

cpp/include/cudf/dictionary/detail/encode.hpp

@@ -1,5 +1,5 @@
 /*
- * SPDX-FileCopyrightText: Copyright (c) 2020-2024, NVIDIA CORPORATION.
+ * SPDX-FileCopyrightText: Copyright (c) 2020-2026, NVIDIA CORPORATION.
  * SPDX-License-Identifier: Apache-2.0
  */
 #pragma once
@@ -12,7 +12,7 @@
 
 #include <rmm/cuda_stream_view.hpp>
 
-namespace CUDF_EXPORT cudf {
+namespace cudf {
 namespace dictionary::detail {
 /**
  * @brief Construct a dictionary column by dictionary encoding an existing column.
@@ -73,4 +73,4 @@ std::unique_ptr<column> decode(dictionary_column_view const& dictionary_column,
 data_type get_indices_type_for_size(size_type keys_size);
 
 }  // namespace dictionary::detail
-}  // namespace CUDF_EXPORT cudf
+}  // namespace cudf

cpp/include/cudf/dictionary/detail/merge.hpp

@@ -1,5 +1,5 @@
 /*
- * SPDX-FileCopyrightText: Copyright (c) 2020-2024, NVIDIA CORPORATION.
+ * SPDX-FileCopyrightText: Copyright (c) 2020-2026, NVIDIA CORPORATION.
  * SPDX-License-Identifier: Apache-2.0
  */
 #pragma once
@@ -11,7 +11,7 @@
 
 #include <rmm/cuda_stream_view.hpp>
 
-namespace CUDF_EXPORT cudf {
+namespace cudf {
 namespace dictionary::detail {
 
 /**
@@ -36,4 +36,4 @@ std::unique_ptr<column> merge(dictionary_column_view const& lcol,
                               rmm::device_async_resource_ref mr);
 
 }  // namespace dictionary::detail
-}  // namespace CUDF_EXPORT cudf
+}  // namespace cudf

cpp/include/cudf/dictionary/detail/replace.hpp

@@ -1,5 +1,5 @@
 /*
- * SPDX-FileCopyrightText: Copyright (c) 2020-2024, NVIDIA CORPORATION.
+ * SPDX-FileCopyrightText: Copyright (c) 2020-2026, NVIDIA CORPORATION.
  * SPDX-License-Identifier: Apache-2.0
  */
 #pragma once
@@ -12,7 +12,7 @@
 
 #include <rmm/cuda_stream_view.hpp>
 
-namespace CUDF_EXPORT cudf {
+namespace cudf {
 namespace dictionary::detail {
 
 /**
@@ -51,4 +51,4 @@ std::unique_ptr<column> replace_nulls(dictionary_column_view const& input,
                                       rmm::device_async_resource_ref mr);
 
 }  // namespace dictionary::detail
-}  // namespace CUDF_EXPORT cudf
+}  // namespace cudf

cpp/include/cudf/dictionary/detail/search.hpp

@@ -1,18 +1,17 @@
 /*
- * SPDX-FileCopyrightText: Copyright (c) 2020-2024, NVIDIA CORPORATION.
+ * SPDX-FileCopyrightText: Copyright (c) 2020-2026, NVIDIA CORPORATION.
  * SPDX-License-Identifier: Apache-2.0
  */
 #pragma once
 
 #include <cudf/dictionary/dictionary_column_view.hpp>
 #include <cudf/scalar/scalar.hpp>
 #include <cudf/utilities/default_stream.hpp>
-#include <cudf/utilities/export.hpp>
 #include <cudf/utilities/memory_resource.hpp>
 
 #include <rmm/cuda_stream_view.hpp>
 
-namespace CUDF_EXPORT cudf {
+namespace cudf {
 namespace dictionary {
 namespace detail {
 
@@ -27,30 +26,6 @@ std::unique_ptr<scalar> get_index(dictionary_column_view const& dictionary,
                                   rmm::cuda_stream_view stream,
                                   rmm::device_async_resource_ref mr);
 
-/**
- * @brief Get the index for a key if it were added to the given dictionary.
- *
- * The actual index is returned if the `key` is already part of the dictionary's key set.
- *
- * @code{.pseudo}
- * d1 = {["a","c","d"],[2,0,1,0]}
- * idx = get_insert_index(d1,"b")
- * idx is 1
- * @endcode{.pseudo}
- *
- * @throw cudf::logic_error if `key.type() != dictionary.keys().type()`
- *
- * @param dictionary The dictionary to search for the key.
- * @param key The value to search for in the dictionary keyset.
- * @param stream CUDA stream used for device memory operations and kernel launches.
- * @param mr Device memory resource used to allocate the returned column's device memory.
- * @return Numeric scalar index value of the key within the dictionary
- */
-std::unique_ptr<scalar> get_insert_index(dictionary_column_view const& dictionary,
-                                         scalar const& key,
-                                         rmm::cuda_stream_view stream,
-                                         rmm::device_async_resource_ref mr);
-
 }  // namespace detail
 }  // namespace dictionary
-}  // namespace CUDF_EXPORT cudf
+}  // namespace cudf

cpp/include/cudf/dictionary/detail/update_keys.hpp

@@ -1,5 +1,5 @@
 /*
- * SPDX-FileCopyrightText: Copyright (c) 2020-2024, NVIDIA CORPORATION.
+ * SPDX-FileCopyrightText: Copyright (c) 2020-2026, NVIDIA CORPORATION.
  * SPDX-License-Identifier: Apache-2.0
  */
 #pragma once
@@ -13,7 +13,7 @@
 
 #include <rmm/cuda_stream_view.hpp>
 
-namespace CUDF_EXPORT cudf {
+namespace cudf {
 namespace dictionary::detail {
 /**
  * @copydoc cudf::dictionary::add_keys(dictionary_column_view const&,column_view
@@ -92,4 +92,4 @@ std::pair<std::vector<std::unique_ptr<column>>, std::vector<table_view>> match_d
   std::vector<table_view> tables, rmm::cuda_stream_view stream, rmm::device_async_resource_ref mr);
 
 }  // namespace dictionary::detail
-}  // namespace CUDF_EXPORT cudf
+}  // namespace cudf

cpp/include/cudf/io/csv.hpp

@@ -1,5 +1,5 @@
 /*
- * SPDX-FileCopyrightText: Copyright (c) 2020-2025, NVIDIA CORPORATION.
+ * SPDX-FileCopyrightText: Copyright (c) 2020-2026, NVIDIA CORPORATION.
  * SPDX-License-Identifier: Apache-2.0
  */
 
@@ -795,9 +795,7 @@ class csv_reader_options {
    */
   void set_na_values(std::vector<std::string> vals)
   {
-    if ((!vals.empty()) and (!_na_filter)) {
-      CUDF_FAIL("Can't set na_values when na_filtering is disabled");
-    }
+    CUDF_EXPECTS(vals.empty() or _na_filter, "Can't set na_values when na_filtering is disabled");
 
     _na_values = std::move(vals);
   }