Support gcc 12 as the C++ compiler (#13316)

Corrects compilation issues found by gcc 12 and the new warnings it provides.

Three major classes of issues was found:

- incorrect lambda captures of the wrong container for the iterators being used
- incorrect implicit conversions
- warnings generated by using std::transform and std::initializer_lists

Authors:
  - Robert Maynard (https://github.com/robertmaynard)

Approvers:
  - Bradley Dice (https://github.com/bdice)
  - David Wendt (https://github.com/davidwendt)
  - Nghia Truong (https://github.com/ttnghia)
  - Mark Harris (https://github.com/harrism)
  - Vyas Ramasubramani (https://github.com/vyasr)

URL: #13316

cpp/cmake/thirdparty/patches/nvbench_override.json

@@ -12,6 +12,11 @@
           "file" : "nvbench/use_existing_fmt.diff",
           "issue" : "Fix add support for using an existing fmt [https://github.com/NVIDIA/nvbench/pull/125]",
           "fixed_in" : ""
+        },
+        {
+          "file" : "nvbench/public_fmt_dep_in_conda.diff",
+          "issue" : "Propagate fmt requirement in conda envs [https://github.com/NVIDIA/nvbench/pull/127]",
+          "fixed_in" : ""
         }
       ]
     }

cpp/include/cudf_test/iterator_utilities.hpp

@@ -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.
@@ -121,15 +121,17 @@ template <typename Iter>
  * The returned iterator yields `false` (to mark `null`) at the indices corresponding to the
  * pointers having `nullptr` values and `true` for the remaining indices.
  *
+ * @note The input vector is referenced by the transform iterator, so the
+ * lifespan must be just as long as the iterator.
+ *
  * @tparam T the data type
  * @param ptrs The data pointers for which the validity iterator is computed
  * @return auto Validity iterator
  */
 template <class T>
 [[maybe_unused]] static auto nulls_from_nullptrs(std::vector<T const*> const& ptrs)
 {
-  // The vector `indices` is copied into the lambda as it can be destroyed at the caller site.
-  return thrust::make_transform_iterator(ptrs.begin(), [ptrs](auto ptr) { return ptr != nullptr; });
+  return thrust::make_transform_iterator(ptrs.begin(), [](auto ptr) { return ptr != nullptr; });
 }
 
 }  // namespace iterators

cpp/include/cudf_test/type_lists.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.
@@ -84,11 +84,13 @@ std::enable_if_t<cudf::is_fixed_width<TypeParam>() && !cudf::is_timestamp_t<Type
                  thrust::host_vector<TypeParam>>
 make_type_param_vector(std::initializer_list<T> const& init_list)
 {
-  thrust::host_vector<TypeParam> vec(init_list.size());
-  std::transform(std::cbegin(init_list), std::cend(init_list), std::begin(vec), [](auto const& e) {
-    if constexpr (std::is_unsigned_v<TypeParam>) { return static_cast<TypeParam>(std::abs(e)); }
-    return static_cast<TypeParam>(e);
-  });
+  std::vector<T> input{init_list};
+  std::vector<TypeParam> vec(init_list.size());
+  std::transform(
+    std::cbegin(input), std::cend(input), std::begin(vec), [](auto const& e) -> TypeParam {
+      if constexpr (std::is_unsigned_v<TypeParam>) { return static_cast<TypeParam>(std::abs(e)); }
+      return static_cast<TypeParam>(e);
+    });
   return vec;
 }
 

cpp/src/io/json/json_column.cu

@@ -813,10 +813,10 @@ std::pair<std::unique_ptr<column>, std::vector<column_name_info>> device_json_co
 
       // For string columns return ["offsets", "char"] schema
       if (target_type.id() == type_id::STRING) {
-        return {std::move(col), {{"offsets"}, {"chars"}}};
+        return {std::move(col), std::vector<column_name_info>{{"offsets"}, {"chars"}}};
       }
       // Non-string leaf-columns (e.g., numeric) do not have child columns in the schema
-      return {std::move(col), {}};
+      return {std::move(col), std::vector<column_name_info>{}};
     }
     case json_col_t::StructColumn: {
       std::vector<std::unique_ptr<column>> child_columns;
@@ -860,7 +860,7 @@ std::pair<std::unique_ptr<column>, std::vector<column_name_info>> device_json_co
                                                        data_type{type_id::INT8},
                                                        0,
                                                        rmm::device_buffer{0, stream, mr}),
-                                                     {}}
+                                                     std::vector<column_name_info>{}}
           : device_json_column_to_cudf_column(
               json_col.child_columns.begin()->second,
               d_input,

cpp/src/io/json/nested_json_gpu.cu

@@ -1678,11 +1678,11 @@ std::pair<std::unique_ptr<column>, std::vector<column_name_info>> json_column_to
 
       // For string columns return ["offsets", "char"] schema
       if (target_type.id() == type_id::STRING) {
-        return {std::move(col), {{"offsets"}, {"chars"}}};
+        return {std::move(col), std::vector<column_name_info>{{"offsets"}, {"chars"}}};
       }
       // Non-string leaf-columns (e.g., numeric) do not have child columns in the schema
       else {
-        return {std::move(col), {}};
+        return {std::move(col), std::vector<column_name_info>{}};
       }
       break;
     }
@@ -1724,7 +1724,8 @@ std::pair<std::unique_ptr<column>, std::vector<column_name_info>> json_column_to
       auto [child_column, names] =
         json_col.child_columns.empty()
           ? std::pair<std::unique_ptr<column>,
-                      std::vector<column_name_info>>{std::make_unique<column>(), {}}
+                      std::vector<column_name_info>>{std::make_unique<column>(),
+                                                     std::vector<column_name_info>{}}
           : json_column_to_cudf_column(json_col.child_columns.begin()->second,
                                        d_input,
                                        options,

cpp/tests/io/json_test.cpp

@@ -1568,7 +1568,7 @@ TEST_P(JsonReaderParamTest, JsonDtypeParsing)
 
   auto make_validity = [](std::vector<int> const& validity) {
     return cudf::detail::make_counting_transform_iterator(
-      0, [=](auto i) -> bool { return static_cast<bool>(validity[i]); });
+      0, [&](auto i) -> bool { return static_cast<bool>(validity[i]); });
   };
 
   constexpr int int_ignore{};

cpp/tests/scalar/scalar_test.cpp

@@ -33,7 +33,7 @@ TYPED_TEST_SUITE(TypedScalarTestWithoutFixedPoint, cudf::test::FixedWidthTypesWi
 TYPED_TEST(TypedScalarTest, DefaultValidity)
 {
   using Type = cudf::device_storage_type_t<TypeParam>;
-  Type value = cudf::test::make_type_param_scalar<TypeParam>(7);
+  Type value = static_cast<Type>(cudf::test::make_type_param_scalar<TypeParam>(7));
   cudf::scalar_type_t<TypeParam> s(value);
 
   EXPECT_TRUE(s.is_valid());
@@ -71,7 +71,7 @@ TYPED_TEST(TypedScalarTestWithoutFixedPoint, SetNull)
 TYPED_TEST(TypedScalarTest, CopyConstructor)
 {
   using Type = cudf::device_storage_type_t<TypeParam>;
-  Type value = cudf::test::make_type_param_scalar<TypeParam>(8);
+  Type value = static_cast<Type>(cudf::test::make_type_param_scalar<TypeParam>(8));
   cudf::scalar_type_t<TypeParam> s(value);
   auto s2 = s;