GH-47167: [C++][Dev] Update clang-format dependency

.pre-commit-config.yaml

@@ -50,7 +50,7 @@ repos:
           )
         types: []
   - repo: https://github.com/pre-commit/mirrors-clang-format
-    rev: v14.0.6
+    rev: v18.1.8
     hooks:
       - id: clang-format
         name: C++ Format
@@ -93,15 +93,15 @@ repos:
           ?^cpp/thirdparty/|
           )
   - repo: https://github.com/pre-commit/mirrors-clang-format
-    rev: v14.0.6
+    rev: v18.1.8
     hooks:
       - id: clang-format
         alias: c-glib
         name: C/GLib Format
         files: >-
           ^c_glib/
   - repo: https://github.com/pre-commit/mirrors-clang-format
-    rev: v14.0.6
+    rev: v18.1.8
     hooks:
       - id: clang-format
         name: MATLAB (C++) Format
@@ -156,7 +156,7 @@ repos:
         files: >-
           ^python/
   - repo: https://github.com/pre-commit/mirrors-clang-format
-    rev: v14.0.6
+    rev: v18.1.8
     hooks:
       - id: clang-format
         alias: python
@@ -213,7 +213,7 @@ repos:
         files: >-
             ^r/.*\.R$
   - repo: https://github.com/pre-commit/mirrors-clang-format
-    rev: v14.0.6
+    rev: v18.1.8
     hooks:
       - id: clang-format
         alias: r

cpp/src/arrow/acero/exec_plan.cc

@@ -137,7 +137,7 @@ struct ExecPlanImpl : public ExecPlan {
             opentelemetry::nostd::shared_ptr<opentelemetry::trace::Span> span =
                 ::arrow::internal::tracing::UnwrapSpan(span_.details.get());
             std::for_each(std::begin(pairs), std::end(pairs),
-                          [span](std::pair<std::string, std::string> const& pair) {
+                          [span](const std::pair<std::string, std::string>& pair) {
                             span->SetAttribute(pair.first, pair.second);
                           });
           }

cpp/src/arrow/compute/kernels/aggregate_test.cc

@@ -3889,8 +3889,7 @@ class TestPrimitiveQuantileKernel : public ::testing::Test {
 #define INTYPE(x) Datum(static_cast<typename TypeParam::c_type>(x))
 #define DOUBLE(x) Datum(static_cast<double>(x))
 // output type per interpolation: linear, lower, higher, nearest, midpoint
-#define O(a, b, c, d, e) \
-  { DOUBLE(a), INTYPE(b), INTYPE(c), INTYPE(d), DOUBLE(e) }
+#define O(a, b, c, d, e) {DOUBLE(a), INTYPE(b), INTYPE(c), INTYPE(d), DOUBLE(e)}
 
 template <typename ArrowType>
 class TestIntegerQuantileKernel : public TestPrimitiveQuantileKernel<ArrowType> {};

cpp/src/arrow/compute/kernels/hash_aggregate.cc

@@ -282,7 +282,7 @@ concept CFloatingPointConcept = std::floating_point<T> || std::same_as<T, util::
 
 template <typename T>
 concept CDecimalConcept = std::same_as<T, Decimal32> || std::same_as<T, Decimal64> ||
-    std::same_as<T, Decimal128> || std::same_as<T, Decimal256>;
+                          std::same_as<T, Decimal128> || std::same_as<T, Decimal256>;
 
 template <typename CType>
 struct AntiExtrema {

cpp/src/arrow/compute/kernels/temporal_internal.h

@@ -45,8 +45,8 @@ using std::chrono::duration_cast;
 using ArrowTimeZone = std::variant<const time_zone*, OffsetZone>;
 
 template <class Duration, class Func>
-auto ApplyTimeZone(const ArrowTimeZone& tz, sys_time<Duration> st, Func&& func)
-    -> decltype(func(zoned_time<Duration>{})) {
+auto ApplyTimeZone(const ArrowTimeZone& tz, sys_time<Duration> st,
+                   Func&& func) -> decltype(func(zoned_time<Duration>{})) {
   return std::visit(
       [&](auto&& zone) {
         if constexpr (std::is_pointer_v<std::decay_t<decltype(zone)> >) {
@@ -60,8 +60,8 @@ auto ApplyTimeZone(const ArrowTimeZone& tz, sys_time<Duration> st, Func&& func)
 
 template <class Duration, class Func>
 auto ApplyTimeZone(const ArrowTimeZone& tz, local_time<Duration> lt,
-                   std::optional<choose> c, Func&& func)
-    -> decltype(func(zoned_time<Duration>{})) {
+                   std::optional<choose> c,
+                   Func&& func) -> decltype(func(zoned_time<Duration>{})) {
   return std::visit(
       [&](auto&& zone) {
         if constexpr (std::is_pointer_v<std::decay_t<decltype(zone)> >) {

cpp/src/arrow/compute/kernels/vector_selection_filter_internal.cc

@@ -368,7 +368,7 @@ class PrimitiveFilterImpl {
           }
         }
       }  // !filter_block.AllSet()
-    }    // while(in_position < values_length_)
+    }  // while(in_position < values_length_)
   }
 
   // Write the next out_position given the selected in_position for the input

cpp/src/arrow/compute/kernels/vector_sort_test.cc

@@ -76,8 +76,8 @@ std::ostream& operator<<(std::ostream& os, NullPlacement null_placement) {
 // Tests for NthToIndices
 
 template <typename ArrayType>
-auto GetLogicalValue(const ArrayType& array, uint64_t index)
-    -> decltype(array.GetView(index)) {
+auto GetLogicalValue(const ArrayType& array,
+                     uint64_t index) -> decltype(array.GetView(index)) {
   return array.GetView(index);
 }
 

cpp/src/arrow/dataset/dataset_writer_test.cc

@@ -613,8 +613,8 @@ TEST_F(DatasetWriterTestFixture, ErrOnExistingData) {
                      fs::File("testdir/chunk-5.arrow"), fs::File("testdir/blah.txt")}));
   filesystem_ = std::dynamic_pointer_cast<MockFileSystem>(fs);
   write_options_.filesystem = filesystem_;
-  ASSERT_RAISES(Invalid, DatasetWriter::Make(
-                             write_options_, scheduler_, [] {}, [] {}, [] {}));
+  ASSERT_RAISES(Invalid,
+                DatasetWriter::Make(write_options_, scheduler_, [] {}, [] {}, [] {}));
   AssertEmptyFiles(
       {"testdir/chunk-0.arrow", "testdir/chunk-5.arrow", "testdir/blah.txt"});
 
@@ -627,8 +627,8 @@ TEST_F(DatasetWriterTestFixture, ErrOnExistingData) {
   filesystem_ = std::dynamic_pointer_cast<MockFileSystem>(fs2);
   write_options_.filesystem = filesystem_;
   write_options_.base_dir = "testdir";
-  ASSERT_RAISES(Invalid, DatasetWriter::Make(
-                             write_options_, scheduler_, [] {}, [] {}, [] {}));
+  ASSERT_RAISES(Invalid,
+                DatasetWriter::Make(write_options_, scheduler_, [] {}, [] {}, [] {}));
   AssertEmptyFiles({"testdir/part-0.arrow"});
 }
 

cpp/src/arrow/dataset/file_test.cc

@@ -594,11 +594,12 @@ class FileSystemWriteTest : public testing::TestWithParam<std::tuple<bool, bool>
 };
 
 TEST_P(FileSystemWriteTest, Write) {
-  auto plan_factory = [](const FileSystemDatasetWriteOptions& write_options,
-                         std::function<Future<std::optional<cp::ExecBatch>>()>*
-                             sink_gen) {
-    return std::vector<acero::Declaration>{{"write", WriteNodeOptions{write_options}}};
-  };
+  auto plan_factory =
+      [](const FileSystemDatasetWriteOptions& write_options,
+         std::function<Future<std::optional<cp::ExecBatch>>()>* sink_gen) {
+        return std::vector<acero::Declaration>{
+            {"write", WriteNodeOptions{write_options}}};
+      };
   TestDatasetWriteRoundTrip(plan_factory, /*has_output=*/false);
 }
 

cpp/src/arrow/device.cc

@@ -42,8 +42,7 @@ Result<std::shared_ptr<Device::SyncEvent>> MemoryManager::WrapDeviceSyncEvent(
   return nullptr;
 }
 
-Device::~Device() {}
-
+Device::~Device(){}
 #define COPY_BUFFER_SUCCESS(maybe_buffer) \
   ((maybe_buffer).ok() && *(maybe_buffer) != nullptr)
 

cpp/src/arrow/engine/substrait/serde_test.cc

@@ -1229,9 +1229,9 @@ TEST(Substrait, ExtensionSetFromPlan) {
        {std::shared_ptr<ExtensionIdRegistry>(), MakeExtensionIdRegistry()}) {
     ExtensionIdRegistry* ext_id_reg = sp_ext_id_reg.get();
     ExtensionSet ext_set(ext_id_reg);
-    ASSERT_OK_AND_ASSIGN(auto sink_decls,
-                         DeserializePlans(
-                             *buf, [] { return kNullConsumer; }, ext_id_reg, &ext_set));
+    ASSERT_OK_AND_ASSIGN(
+        auto sink_decls,
+        DeserializePlans(*buf, [] { return kNullConsumer; }, ext_id_reg, &ext_set));
 
     EXPECT_OK_AND_ASSIGN(auto decoded_null_type, ext_set.DecodeType(42));
     EXPECT_EQ(decoded_null_type.id.uri, kArrowExtTypesUri);
@@ -1665,9 +1665,9 @@ TEST(Substrait, JoinPlanBasic) {
        {std::shared_ptr<ExtensionIdRegistry>(), MakeExtensionIdRegistry()}) {
     ExtensionIdRegistry* ext_id_reg = sp_ext_id_reg.get();
     ExtensionSet ext_set(ext_id_reg);
-    ASSERT_OK_AND_ASSIGN(auto sink_decls,
-                         DeserializePlans(
-                             *buf, [] { return kNullConsumer; }, ext_id_reg, &ext_set));
+    ASSERT_OK_AND_ASSIGN(
+        auto sink_decls,
+        DeserializePlans(*buf, [] { return kNullConsumer; }, ext_id_reg, &ext_set));
 
     auto join_decl = sink_decls[0].inputs[0];
 

cpp/src/arrow/flight/serialization_internal.cc

@@ -36,7 +36,7 @@ struct overloaded : Ts... {
   using Ts::operator()...;
 };
 template <class... Ts>  // CTAD will not be needed for >=C++20
-overloaded(Ts...)->overloaded<Ts...>;
+overloaded(Ts...) -> overloaded<Ts...>;
 
 namespace arrow {
 namespace flight {

cpp/src/arrow/io/memory_test.cc

@@ -580,8 +580,7 @@ class TestTransformInputStream : public ::testing::Test {
     auto stream = std::make_shared<TransformInputStream>(
         std::make_shared<BufferReader>(src), this->transform());
     std::shared_ptr<Buffer> actual;
-    AccumulateReads(
-        stream, [&]() -> int64_t { return chunk_sizes(gen); }, &actual);
+    AccumulateReads(stream, [&]() -> int64_t { return chunk_sizes(gen); }, &actual);
     AssertBufferEqual(*actual, *expected);
   }
 
@@ -613,8 +612,7 @@ class TestTransformInputStream : public ::testing::Test {
 
   void AccumulateReads(const std::shared_ptr<InputStream>& stream, int64_t chunk_size,
                        std::shared_ptr<Buffer>* out) {
-    return AccumulateReads(
-        stream, [=]() { return chunk_size; }, out);
+    return AccumulateReads(stream, [=]() { return chunk_size; }, out);
   }
 
  protected:

cpp/src/arrow/json/test_common.h

@@ -89,7 +89,7 @@ struct GenerateImpl {
   Status Visit(const NullType&) { return OK(writer.Null()); }
 
   Status Visit(const BooleanType&) {
-    return OK(writer.Bool(std::uniform_int_distribution<uint16_t>{}(e)&1));
+    return OK(writer.Bool(std::uniform_int_distribution<uint16_t>{}(e) & 1));
   }
 
   template <typename T>

cpp/src/arrow/scalar.cc

@@ -565,7 +565,7 @@ struct ScalarValidateImpl {
 };
 
 template <typename T, size_t N>
-void FillScalarScratchSpace(void* scratch_space, T const (&arr)[N]) {
+void FillScalarScratchSpace(void* scratch_space, const T (&arr)[N]) {
   static_assert(sizeof(arr) <= internal::kScalarScratchSpaceSize);
   std::memcpy(scratch_space, arr, sizeof(arr));
 }

cpp/src/arrow/util/async_generator.h

@@ -1439,9 +1439,8 @@ class MergedGenerator {
             immediate_inner(next_item.result());
             if (immediate_inner.was_empty) {
               Future<AsyncGenerator<T>> next_source = state->PullSource();
-              if (next_source.TryAddCallback([this] {
-                    return OuterCallback{state, index};
-                  })) {
+              if (next_source.TryAddCallback(
+                      [this] { return OuterCallback{state, index}; })) {
                 // We hit an unfinished future so we can stop looping
                 return;
               }

cpp/src/arrow/util/bpacking.cc

@@ -33,15 +33,15 @@ struct UnpackDynamicFunction {
   using Implementation = std::pair<DispatchLevel, FunctionType>;
 
   static constexpr auto implementations() {
-    return std::array {
-      // Current SIMD unpack algorithm works terribly on SSE4.2 due to lack of variable
-      // rhsift and poor xsimd fallback.
-      Implementation{DispatchLevel::NONE, &unpack_scalar<Uint>},
+    return std::array{
+        // Current SIMD unpack algorithm works terribly on SSE4.2 due to lack of variable
+        // rhsift and poor xsimd fallback.
+        Implementation{DispatchLevel::NONE, &unpack_scalar<Uint>},
 #if defined(ARROW_HAVE_RUNTIME_AVX2)
-          Implementation{DispatchLevel::AVX2, &unpack_avx2<Uint>},
+        Implementation{DispatchLevel::AVX2, &unpack_avx2<Uint>},
 #endif
 #if defined(ARROW_HAVE_RUNTIME_AVX512)
-          Implementation{DispatchLevel::AVX512, &unpack_avx512<Uint>},
+        Implementation{DispatchLevel::AVX512, &unpack_avx512<Uint>},
 #endif
     };
   }

cpp/src/arrow/util/byte_stream_split_internal.cc

@@ -35,9 +35,9 @@ struct ByteStreamSplitDecodeDynamic {
   using Implementation = std::pair<DispatchLevel, FunctionType>;
 
   constexpr static auto implementations() {
-    return std::array {
-      Implementation {
-        DispatchLevel::NONE,
+    return std::array{
+        Implementation{
+            DispatchLevel::NONE,
 #if defined(ARROW_HAVE_NEON)
             // We always expect Neon to be available on Arm64
             &ByteStreamSplitDecodeSimd<xsimd::neon64, kNumStreams>,
@@ -47,13 +47,12 @@ struct ByteStreamSplitDecodeDynamic {
 #else
             &ByteStreamSplitDecodeScalar<kNumStreams>,
 #endif
-      }
-      ,
+        },
 #if defined(ARROW_HAVE_RUNTIME_AVX2)
-          Implementation{
-              DispatchLevel::AVX2,
-              &ByteStreamSplitDecodeSimd<xsimd::avx2, kNumStreams>,
-          },
+        Implementation{
+            DispatchLevel::AVX2,
+            &ByteStreamSplitDecodeSimd<xsimd::avx2, kNumStreams>,
+        },
 #endif
     };
   }
@@ -83,9 +82,9 @@ struct ByteStreamSplitEncodeDynamic {
   using Implementation = std::pair<DispatchLevel, FunctionType>;
 
   constexpr static auto implementations() {
-    return std::array {
-      Implementation {
-        DispatchLevel::NONE,
+    return std::array{
+        Implementation{
+            DispatchLevel::NONE,
 #if defined(ARROW_HAVE_NEON)
             // We always expect Neon to be available on Arm64
             &ByteStreamSplitEncodeSimd<xsimd::neon64, kNumStreams>,
@@ -95,10 +94,9 @@ struct ByteStreamSplitEncodeDynamic {
 #else
             &ByteStreamSplitEncodeScalar<kNumStreams>,
 #endif
-      }
-      ,
+        },
 #if defined(ARROW_HAVE_RUNTIME_AVX2)
-          Implementation{DispatchLevel::AVX2, &ByteStreamSplitEncodeAvx2<kNumStreams>},
+        Implementation{DispatchLevel::AVX2, &ByteStreamSplitEncodeAvx2<kNumStreams>},
 #endif
     };
   }

cpp/src/arrow/util/byte_stream_split_internal.h

@@ -122,8 +122,8 @@ void ByteStreamSplitDecodeSimd(const uint8_t* data, int width, int64_t num_value
 
 // Like xsimd::zip_lo, but zip groups of kNumBytes at once.
 template <typename Arch, int kNumBytes>
-auto zip_lo_n(const xsimd::batch<int8_t, Arch>& a, const xsimd::batch<int8_t, Arch>& b)
-    -> xsimd::batch<int8_t, Arch> {
+auto zip_lo_n(const xsimd::batch<int8_t, Arch>& a,
+              const xsimd::batch<int8_t, Arch>& b) -> xsimd::batch<int8_t, Arch> {
   using arrow::internal::SizedInt;
   using simd_batch = xsimd::batch<int8_t, Arch>;
   // For signed arithmetic
@@ -146,8 +146,8 @@ auto zip_lo_n(const xsimd::batch<int8_t, Arch>& a, const xsimd::batch<int8_t, Ar
 
 // Like xsimd::zip_hi, but zip groups of kNumBytes at once.
 template <typename Arch, int kNumBytes>
-auto zip_hi_n(const xsimd::batch<int8_t, Arch>& a, const xsimd::batch<int8_t, Arch>& b)
-    -> xsimd::batch<int8_t, Arch> {
+auto zip_hi_n(const xsimd::batch<int8_t, Arch>& a,
+              const xsimd::batch<int8_t, Arch>& b) -> xsimd::batch<int8_t, Arch> {
   using simd_batch = xsimd::batch<int8_t, Arch>;
   using arrow::internal::SizedInt;
   // For signed arithmetic

cpp/src/arrow/util/compare.h

@@ -51,8 +51,8 @@ class EqualityComparable {
     }
   };
 
-  friend bool operator==(T const& a, T const& b) { return a.Equals(b); }
-  friend bool operator!=(T const& a, T const& b) { return !(a == b); }
+  friend bool operator==(const T& a, const T& b) { return a.Equals(b); }
+  friend bool operator!=(const T& a, const T& b) { return !(a == b); }
 
  private:
   const T& cast() const { return static_cast<const T&>(*this); }