ARROW-17694: [C++] Remove std::optional backport (#14105)

Just use the C++17 standard library version.

Authored-by: Antoine Pitrou <antoine@python.org>
Signed-off-by: Antoine Pitrou <antoine@python.org>

.github/workflows/cpp.yml

@@ -257,7 +257,7 @@ jobs:
       - name: Install ccache
         shell: bash
         run: |
-          ci/scripts/install_ccache.sh 4.6.2 /usr
+          ci/scripts/install_ccache.sh 4.6.3 /usr
       - name: Setup ccache
         shell: bash
         run: |
@@ -271,8 +271,11 @@ jobs:
         uses: actions/cache@v2
         with:
           path: ${{ steps.ccache-info.outputs.cache-dir }}
-          key: cpp-ccache-windows-${{ hashFiles('cpp/**') }}
-          restore-keys: cpp-ccache-windows-
+          key: cpp-ccache-windows-${{ env.CACHE_VERSION }}-${{ hashFiles('cpp/**') }}
+          restore-keys: cpp-ccache-windows-${{ env.CACHE_VERSION }}-
+        env:
+          # We can invalidate the current cache by updating this.
+          CACHE_VERSION: "2022-09-13"
       - name: Build
         shell: cmd
         run: |

LICENSE.txt

@@ -2059,34 +2059,6 @@ René Nyffenegger rene.nyffenegger@adp-gmbh.ch
 
 --------------------------------------------------------------------------------
 
-The file cpp/src/arrow/vendored/optional.hpp has the following license
-
-Boost Software License - Version 1.0 - August 17th, 2003
-
-Permission is hereby granted, free of charge, to any person or organization
-obtaining a copy of the software and accompanying documentation covered by
-this license (the "Software") to use, reproduce, display, distribute,
-execute, and transmit the Software, and to prepare derivative works of the
-Software, and to permit third-parties to whom the Software is furnished to
-do so, all subject to the following:
-
-The copyright notices in the Software and this entire statement, including
-the above license grant, this restriction and the following disclaimer,
-must be included in all copies of the Software, in whole or in part, and
-all derivative works of the Software, unless such copies or derivative
-works are solely in the form of machine-executable object code generated by
-a source language processor.
-
-THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
-IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
-FITNESS FOR A PARTICULAR PURPOSE, TITLE AND NON-INFRINGEMENT. IN NO EVENT
-SHALL THE COPYRIGHT HOLDERS OR ANYONE DISTRIBUTING THE SOFTWARE BE LIABLE
-FOR ANY DAMAGES OR OTHER LIABILITY, WHETHER IN CONTRACT, TORT OR OTHERWISE,
-ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
-DEALINGS IN THE SOFTWARE.
-
---------------------------------------------------------------------------------
-
 This project includes code from Folly.
 
  * cpp/src/arrow/vendored/ProducerConsumerQueue.h

c_glib/arrow-glib/compute.cpp

@@ -938,7 +938,7 @@ garrow_source_node_options_new_record_batch_reader(
     arrow_reader->schema(),
     [arrow_reader]() {
       using ExecBatch = arrow::compute::ExecBatch;
-      using ExecBatchOptional = arrow::util::optional<ExecBatch>;
+      using ExecBatchOptional = std::optional<ExecBatch>;
       auto arrow_record_batch_result = arrow_reader->Next();
       if (!arrow_record_batch_result.ok()) {
         return arrow::AsyncGeneratorEnd<ExecBatchOptional>();
@@ -979,7 +979,7 @@ garrow_source_node_options_new_record_batch(GArrowRecordBatch *record_batch)
     state->record_batch->schema(),
     [state]() {
       using ExecBatch = arrow::compute::ExecBatch;
-      using ExecBatchOptional = arrow::util::optional<ExecBatch>;
+      using ExecBatchOptional = std::optional<ExecBatch>;
       if (!state->generated) {
         state->generated = true;
         return arrow::Future<ExecBatchOptional>::MakeFinished(
@@ -1296,7 +1296,7 @@ garrow_aggregate_node_options_new(GList *aggregations,
 
 
 typedef struct GArrowSinkNodeOptionsPrivate_ {
-  arrow::AsyncGenerator<arrow::util::optional<arrow::compute::ExecBatch>> generator;
+  arrow::AsyncGenerator<std::optional<arrow::compute::ExecBatch>> generator;
   GArrowRecordBatchReader *reader;
 } GArrowSinkNodeOptionsPrivate;
 
@@ -1333,7 +1333,7 @@ garrow_sink_node_options_init(GArrowSinkNodeOptions *object)
 {
   auto priv = GARROW_SINK_NODE_OPTIONS_GET_PRIVATE(object);
   new(&(priv->generator))
-    arrow::AsyncGenerator<arrow::util::optional<arrow::compute::ExecBatch>>();
+    arrow::AsyncGenerator<std::optional<arrow::compute::ExecBatch>>();
 }
 
 static void

ci/scripts/python_wheel_macos_build.sh

@@ -34,7 +34,7 @@ rm -rf ${source_dir}/python/pyarrow/*.so.*
 
 echo "=== (${PYTHON_VERSION}) Set SDK, C++ and Wheel flags ==="
 export _PYTHON_HOST_PLATFORM="macosx-${MACOSX_DEPLOYMENT_TARGET}-${arch}"
-export MACOSX_DEPLOYMENT_TARGET=${MACOSX_DEPLOYMENT_TARGET:-10.13}
+export MACOSX_DEPLOYMENT_TARGET=${MACOSX_DEPLOYMENT_TARGET:-10.14}
 export SDKROOT=${SDKROOT:-$(xcrun --sdk macosx --show-sdk-path)}
 
 if [ $arch = "arm64" ]; then

ci/vcpkg/universal2-osx-static-debug.cmake

@@ -21,6 +21,6 @@ set(VCPKG_LIBRARY_LINKAGE static)
 
 set(VCPKG_CMAKE_SYSTEM_NAME Darwin)
 set(VCPKG_OSX_ARCHITECTURES "x86_64;arm64")
-set(VCPKG_OSX_DEPLOYMENT_TARGET "10.13")
+set(VCPKG_OSX_DEPLOYMENT_TARGET "10.14")
 
 set(VCPKG_BUILD_TYPE debug)

ci/vcpkg/universal2-osx-static-release.cmake

@@ -21,6 +21,6 @@ set(VCPKG_LIBRARY_LINKAGE static)
 
 set(VCPKG_CMAKE_SYSTEM_NAME Darwin)
 set(VCPKG_OSX_ARCHITECTURES "x86_64;arm64")
-set(VCPKG_OSX_DEPLOYMENT_TARGET "10.13")
+set(VCPKG_OSX_DEPLOYMENT_TARGET "10.14")
 
 set(VCPKG_BUILD_TYPE release)

cpp/examples/arrow/compute_register_example.cc

@@ -149,7 +149,7 @@ arrow::Status RunComputeRegister(int argc, char** argv) {
   ARROW_RETURN_NOT_OK(maybe_plan.status());
   ARROW_ASSIGN_OR_RAISE(auto plan, maybe_plan);
 
-  arrow::AsyncGenerator<arrow::util::optional<cp::ExecBatch>> source_gen, sink_gen;
+  arrow::AsyncGenerator<std::optional<cp::ExecBatch>> source_gen, sink_gen;
   ARROW_RETURN_NOT_OK(
       cp::Declaration::Sequence(
           {

cpp/examples/arrow/execution_plan_documentation_examples.cc

@@ -157,11 +157,11 @@ struct BatchesWithSchema {
   std::shared_ptr<arrow::Schema> schema;
   // This method uses internal arrow utilities to
   // convert a vector of record batches to an AsyncGenerator of optional batches
-  arrow::AsyncGenerator<arrow::util::optional<cp::ExecBatch>> gen() const {
+  arrow::AsyncGenerator<std::optional<cp::ExecBatch>> gen() const {
     auto opt_batches = ::arrow::internal::MapVector(
-        [](cp::ExecBatch batch) { return arrow::util::make_optional(std::move(batch)); },
+        [](cp::ExecBatch batch) { return std::make_optional(std::move(batch)); },
         batches);
-    arrow::AsyncGenerator<arrow::util::optional<cp::ExecBatch>> gen;
+    arrow::AsyncGenerator<std::optional<cp::ExecBatch>> gen;
     gen = arrow::MakeVectorGenerator(std::move(opt_batches));
     return gen;
   }
@@ -259,7 +259,7 @@ arrow::Result<BatchesWithSchema> MakeGroupableBatches(int multiplicity = 1) {
 arrow::Status ExecutePlanAndCollectAsTable(
     cp::ExecContext& exec_context, std::shared_ptr<cp::ExecPlan> plan,
     std::shared_ptr<arrow::Schema> schema,
-    arrow::AsyncGenerator<arrow::util::optional<cp::ExecBatch>> sink_gen) {
+    arrow::AsyncGenerator<std::optional<cp::ExecBatch>> sink_gen) {
   // translate sink_gen (async) to sink_reader (sync)
   std::shared_ptr<arrow::RecordBatchReader> sink_reader =
       cp::MakeGeneratorReader(schema, std::move(sink_gen), exec_context.memory_pool());
@@ -312,7 +312,7 @@ arrow::Status ScanSinkExample(cp::ExecContext& exec_context) {
   ARROW_ASSIGN_OR_RAISE(scan,
                         cp::MakeExecNode("scan", plan.get(), {}, scan_node_options));
 
-  arrow::AsyncGenerator<arrow::util::optional<cp::ExecBatch>> sink_gen;
+  arrow::AsyncGenerator<std::optional<cp::ExecBatch>> sink_gen;
 
   ARROW_RETURN_NOT_OK(
       cp::MakeExecNode("sink", plan.get(), {scan}, cp::SinkNodeOptions{&sink_gen}));
@@ -337,7 +337,7 @@ arrow::Status SourceSinkExample(cp::ExecContext& exec_context) {
 
   ARROW_ASSIGN_OR_RAISE(auto basic_data, MakeBasicBatches());
 
-  arrow::AsyncGenerator<arrow::util::optional<cp::ExecBatch>> sink_gen;
+  arrow::AsyncGenerator<std::optional<cp::ExecBatch>> sink_gen;
 
   auto source_node_options = cp::SourceNodeOptions{basic_data.schema, basic_data.gen()};
 
@@ -367,7 +367,7 @@ arrow::Status TableSourceSinkExample(cp::ExecContext& exec_context) {
 
   ARROW_ASSIGN_OR_RAISE(auto table, GetTable());
 
-  arrow::AsyncGenerator<arrow::util::optional<cp::ExecBatch>> sink_gen;
+  arrow::AsyncGenerator<std::optional<cp::ExecBatch>> sink_gen;
   int max_batch_size = 2;
   auto table_source_options = cp::TableSourceNodeOptions{table, max_batch_size};
 
@@ -427,7 +427,7 @@ arrow::Status ScanFilterSinkExample(cp::ExecContext& exec_context) {
                                                  cp::FilterNodeOptions{filter_opt}));
 
   // finally, pipe the filter node into a sink node
-  arrow::AsyncGenerator<arrow::util::optional<cp::ExecBatch>> sink_gen;
+  arrow::AsyncGenerator<std::optional<cp::ExecBatch>> sink_gen;
   ARROW_RETURN_NOT_OK(
       cp::MakeExecNode("sink", plan.get(), {filter}, cp::SinkNodeOptions{&sink_gen}));
 
@@ -470,7 +470,7 @@ arrow::Status ScanProjectSinkExample(cp::ExecContext& exec_context) {
   std::cout << "Schema after projection : \n"
             << project->output_schema()->ToString() << std::endl;
 
-  arrow::AsyncGenerator<arrow::util::optional<cp::ExecBatch>> sink_gen;
+  arrow::AsyncGenerator<std::optional<cp::ExecBatch>> sink_gen;
   ARROW_RETURN_NOT_OK(
       cp::MakeExecNode("sink", plan.get(), {project}, cp::SinkNodeOptions{&sink_gen}));
   auto schema = arrow::schema({arrow::field("a * 2", arrow::int32())});
@@ -496,7 +496,7 @@ arrow::Status SourceScalarAggregateSinkExample(cp::ExecContext& exec_context) {
 
   ARROW_ASSIGN_OR_RAISE(auto basic_data, MakeBasicBatches());
 
-  arrow::AsyncGenerator<arrow::util::optional<cp::ExecBatch>> sink_gen;
+  arrow::AsyncGenerator<std::optional<cp::ExecBatch>> sink_gen;
 
   auto source_node_options = cp::SourceNodeOptions{basic_data.schema, basic_data.gen()};
 
@@ -532,7 +532,7 @@ arrow::Status SourceGroupAggregateSinkExample(cp::ExecContext& exec_context) {
 
   ARROW_ASSIGN_OR_RAISE(auto basic_data, MakeBasicBatches());
 
-  arrow::AsyncGenerator<arrow::util::optional<cp::ExecBatch>> sink_gen;
+  arrow::AsyncGenerator<std::optional<cp::ExecBatch>> sink_gen;
 
   auto source_node_options = cp::SourceNodeOptions{basic_data.schema, basic_data.gen()};
 
@@ -640,7 +640,7 @@ arrow::Status SourceOrderBySinkExample(cp::ExecContext& exec_context) {
 
   std::cout << "basic data created" << std::endl;
 
-  arrow::AsyncGenerator<arrow::util::optional<cp::ExecBatch>> sink_gen;
+  arrow::AsyncGenerator<std::optional<cp::ExecBatch>> sink_gen;
 
   auto source_node_options = cp::SourceNodeOptions{basic_data.schema, basic_data.gen()};
   ARROW_ASSIGN_OR_RAISE(cp::ExecNode * source,
@@ -670,7 +670,7 @@ arrow::Status SourceHashJoinSinkExample(cp::ExecContext& exec_context) {
   ARROW_ASSIGN_OR_RAISE(std::shared_ptr<cp::ExecPlan> plan,
                         cp::ExecPlan::Make(&exec_context));
 
-  arrow::AsyncGenerator<arrow::util::optional<cp::ExecBatch>> sink_gen;
+  arrow::AsyncGenerator<std::optional<cp::ExecBatch>> sink_gen;
 
   cp::ExecNode* left_source;
   cp::ExecNode* right_source;
@@ -714,7 +714,7 @@ arrow::Status SourceKSelectExample(cp::ExecContext& exec_context) {
   ARROW_ASSIGN_OR_RAISE(auto input, MakeGroupableBatches());
   ARROW_ASSIGN_OR_RAISE(std::shared_ptr<cp::ExecPlan> plan,
                         cp::ExecPlan::Make(&exec_context));
-  arrow::AsyncGenerator<arrow::util::optional<cp::ExecBatch>> sink_gen;
+  arrow::AsyncGenerator<std::optional<cp::ExecBatch>> sink_gen;
 
   ARROW_ASSIGN_OR_RAISE(
       cp::ExecNode * source,
@@ -761,7 +761,7 @@ arrow::Status ScanFilterWriteExample(cp::ExecContext& exec_context,
   ARROW_ASSIGN_OR_RAISE(scan,
                         cp::MakeExecNode("scan", plan.get(), {}, scan_node_options));
 
-  arrow::AsyncGenerator<arrow::util::optional<cp::ExecBatch>> sink_gen;
+  arrow::AsyncGenerator<std::optional<cp::ExecBatch>> sink_gen;
 
   std::string root_path = "";
   std::string uri = "file://" + file_path;
@@ -820,7 +820,7 @@ arrow::Status SourceUnionSinkExample(cp::ExecContext& exec_context) {
 
   ARROW_ASSIGN_OR_RAISE(std::shared_ptr<cp::ExecPlan> plan,
                         cp::ExecPlan::Make(&exec_context));
-  arrow::AsyncGenerator<arrow::util::optional<cp::ExecBatch>> sink_gen;
+  arrow::AsyncGenerator<std::optional<cp::ExecBatch>> sink_gen;
 
   cp::Declaration union_node{"union", cp::ExecNodeOptions{}};
   cp::Declaration lhs{"source",

cpp/examples/arrow/join_example.cc

@@ -89,7 +89,7 @@ arrow::Status DoHashJoin() {
   ARROW_ASSIGN_OR_RAISE(std::shared_ptr<cp::ExecPlan> plan,
                         cp::ExecPlan::Make(&exec_context));
 
-  arrow::AsyncGenerator<arrow::util::optional<cp::ExecBatch>> sink_gen;
+  arrow::AsyncGenerator<std::optional<cp::ExecBatch>> sink_gen;
 
   cp::ExecNode* left_source;
   cp::ExecNode* right_source;

cpp/examples/minimal_build/run_static.sh

@@ -102,7 +102,7 @@ echo
 
 rm -rf $EXAMPLE_BUILD_DIR
 mkdir -p $EXAMPLE_BUILD_DIR
-${CXX:-c++} \
+${CXX:-c++} -std=c++17 \
   -o $EXAMPLE_BUILD_DIR/arrow-example \
   $EXAMPLE_DIR/example.cc \
   $(PKG_CONFIG_PATH=$ARROW_BUILD_DIR/lib/pkgconfig \

cpp/gdb_arrow.py

@@ -2175,28 +2175,6 @@ def to_string(self):
             return f"arrow::util::string_view of size {size}, {data}"
 
 
-class OptionalPrinter:
-    """
-    Pretty-printer for arrow::util::optional.
-    """
-
-    def __init__(self, name, val):
-        self.val = val
-
-    def to_string(self):
-        data_type = self.val.type.template_argument(0)
-        # XXX We rely on internal details of our vendored optional<T>
-        # implementation, as inlined methods may not be callable from gdb.
-        if not self.val['has_value_']:
-            inner = "nullopt"
-        else:
-            data_ptr = self.val['contained']['data'].address
-            assert data_ptr
-            inner = data_ptr.reinterpret_cast(
-                data_type.pointer()).dereference()
-        return f"arrow::util::optional<{data_type}>({inner})"
-
-
 class VariantPrinter:
     """
     Pretty-printer for arrow::util::Variant.
@@ -2436,10 +2414,8 @@ def to_string(self):
     "arrow::SimpleTable": TablePrinter,
     "arrow::Status": StatusPrinter,
     "arrow::Table": TablePrinter,
-    "arrow::util::optional": OptionalPrinter,
     "arrow::util::string_view": StringViewPrinter,
     "arrow::util::Variant": VariantPrinter,
-    "nonstd::optional_lite::optional": OptionalPrinter,
     "nonstd::sv_lite::basic_string_view": StringViewPrinter,
 }
 

cpp/src/arrow/array/array_binary.h

@@ -75,7 +75,7 @@ class BaseBinaryArray : public FlatArray {
                              raw_value_offsets_[i + 1] - pos);
   }
 
-  util::optional<util::string_view> operator[](int64_t i) const {
+  std::optional<util::string_view> operator[](int64_t i) const {
     return *IteratorType(*this, i);
   }
 
@@ -240,7 +240,7 @@ class ARROW_EXPORT FixedSizeBinaryArray : public PrimitiveArray {
     return util::string_view(reinterpret_cast<const char*>(GetValue(i)), byte_width());
   }
 
-  util::optional<util::string_view> operator[](int64_t i) const {
+  std::optional<util::string_view> operator[](int64_t i) const {
     return *IteratorType(*this, i);
   }
 

cpp/src/arrow/array/array_primitive.h

@@ -54,7 +54,7 @@ class ARROW_EXPORT BooleanArray : public PrimitiveArray {
 
   bool GetView(int64_t i) const { return Value(i); }
 
-  util::optional<bool> operator[](int64_t i) const { return *IteratorType(*this, i); }
+  std::optional<bool> operator[](int64_t i) const { return *IteratorType(*this, i); }
 
   /// \brief Return the number of false (0) values among the valid
   /// values. Result is not cached.
@@ -111,7 +111,7 @@ class NumericArray : public PrimitiveArray {
   // For API compatibility with BinaryArray etc.
   value_type GetView(int64_t i) const { return Value(i); }
 
-  util::optional<value_type> operator[](int64_t i) const {
+  std::optional<value_type> operator[](int64_t i) const {
     return *IteratorType(*this, i);
   }
 
@@ -152,7 +152,7 @@ class ARROW_EXPORT DayTimeIntervalArray : public PrimitiveArray {
 
   IteratorType end() const { return IteratorType(*this, length()); }
 
-  util::optional<TypeClass::DayMilliseconds> operator[](int64_t i) const {
+  std::optional<TypeClass::DayMilliseconds> operator[](int64_t i) const {
     return *IteratorType(*this, i);
   }
 
@@ -188,7 +188,7 @@ class ARROW_EXPORT MonthDayNanoIntervalArray : public PrimitiveArray {
 
   IteratorType end() const { return IteratorType(*this, length()); }
 
-  util::optional<TypeClass::MonthDayNanos> operator[](int64_t i) const {
+  std::optional<TypeClass::MonthDayNanos> operator[](int64_t i) const {
     return *IteratorType(*this, i);
   }
 

cpp/src/arrow/array/array_test.cc

@@ -2290,7 +2290,7 @@ TEST_F(TestFWBinaryArray, ArrayIndexOperator) {
   auto fsba = checked_pointer_cast<FixedSizeBinaryArray>(arr);
 
   ASSERT_EQ("abc", (*fsba)[0].value());
-  ASSERT_EQ(util::nullopt, (*fsba)[1]);
+  ASSERT_EQ(std::nullopt, (*fsba)[1]);
   ASSERT_EQ("def", (*fsba)[2].value());
 }
 
@@ -3538,7 +3538,7 @@ TYPED_TEST(TestPrimitiveArray, IndexOperator) {
       ASSERT_EQ(this->values_[i], res.value());
     } else {
       ASSERT_FALSE(res.has_value());
-      ASSERT_EQ(res, util::nullopt);
+      ASSERT_EQ(res, std::nullopt);
     }
   }
 }

cpp/src/arrow/compute/exec/asof_join_benchmark.cc

@@ -88,7 +88,7 @@ static void TableJoinOverhead(benchmark::State& state,
     }
     ASSERT_OK_AND_ASSIGN(arrow::compute::ExecNode * join_node,
                          MakeExecNode(factory_name, plan.get(), input_nodes, options));
-    AsyncGenerator<util::optional<ExecBatch>> sink_gen;
+    AsyncGenerator<std::optional<ExecBatch>> sink_gen;
     ASSERT_OK(MakeExecNode("sink", plan.get(), {join_node}, SinkNodeOptions{&sink_gen}));
     state.ResumeTiming();
     ASSERT_FINISHES_OK(StartAndCollect(plan.get(), sink_gen));