ARROW-17872: [C++][CI] Reduce macOS CI dependencies (apache#14310)

Pin LLVM and Clang-format version to 14 because it's pre-installed on the runner image.

Authored-by: Jin Shang <[email protected]>
Signed-off-by: Sutou Kouhei <[email protected]>

Author: js8544

.env

@@ -53,7 +53,7 @@ FEDORA=35
 UBUNTU=20.04
 
 # Default versions for various dependencies
-CLANG_TOOLS=12
+CLANG_TOOLS=14
 CUDA=11.0.3
 DASK=latest
 DOTNET=6.0
@@ -64,7 +64,7 @@ HDFS=3.2.1
 JDK=8
 KARTOTHEK=latest
 # LLVM 12 and GCC 11 reports -Wmismatched-new-delete.
-LLVM=13
+LLVM=14
 MAVEN=3.5.4
 NODE=16
 NUMBA=latest

.github/workflows/cpp.yml

@@ -182,10 +182,7 @@ jobs:
           key: cpp-ccache-macos-${{ hashFiles('cpp/**') }}
           restore-keys: cpp-ccache-macos-
       - name: Build
-        # use brew version of clang, to be consistent with LLVM lib, see ARROW-17790.
         run: |
-          export CC=$(brew --prefix llvm)/bin/clang
-          export CXX=$(brew --prefix llvm)/bin/clang++
           ci/scripts/cpp_build.sh $(pwd) $(pwd)/build
       - name: Test
         shell: bash

.github/workflows/python.yml

@@ -167,10 +167,7 @@ jobs:
             -r python/requirements-test.txt
       - name: Build
         shell: bash
-        # use brew version of clang, to be consistent with LLVM lib, see ARROW-17790.
         run: |
-          export CC=$(brew --prefix llvm)/bin/clang
-          export CXX=$(brew --prefix llvm)/bin/clang++
           export PYTHON=python3
           ci/scripts/cpp_build.sh $(pwd) $(pwd)/build
           ci/scripts/python_build.sh $(pwd) $(pwd)/build

.github/workflows/ruby.yml

@@ -169,10 +169,7 @@ jobs:
           key: ruby-ccache-macos-${{ hashFiles('cpp/**') }}
           restore-keys: ruby-ccache-macos-
       - name: Build C++
-        # use brew version of clang, to be consistent with LLVM lib, see ARROW-17790.
         run: |
-          export CC=$(brew --prefix llvm)/bin/clang
-          export CXX=$(brew --prefix llvm)/bin/clang++
           ci/scripts/cpp_build.sh $(pwd) $(pwd)/build
       - name: Build GLib
         run: |

cpp/Brewfile

@@ -28,8 +28,7 @@ brew "git"
 brew "glog"
 brew "googletest"
 brew "grpc"
-brew "llvm"
-brew "llvm@12"
+brew "llvm@14"
 brew "lz4"
 brew "ninja"
 brew "numpy"

cpp/CMakeLists.txt

@@ -138,9 +138,6 @@ set(ARROW_LLVM_VERSIONS
     "9"
     "8"
     "7")
-list(GET ARROW_LLVM_VERSIONS 0 ARROW_LLVM_VERSION_PRIMARY)
-string(REGEX REPLACE "^([0-9]+)(\\..+)?" "\\1" ARROW_LLVM_VERSION_PRIMARY_MAJOR
-                     "${ARROW_LLVM_VERSION_PRIMARY}")
 
 file(READ ${CMAKE_CURRENT_SOURCE_DIR}/../.env ARROW_ENV)
 string(REGEX MATCH "CLANG_TOOLS=[^\n]+" ARROW_ENV_CLANG_TOOLS_VERSION "${ARROW_ENV}")

cpp/cmake_modules/FindLLVMAlt.cmake

@@ -40,35 +40,30 @@ if(DEFINED LLVM_ROOT)
 endif()
 
 if(NOT LLVM_FOUND)
-  set(LLVM_HINTS ${LLVM_ROOT} ${LLVM_DIR} /usr/lib /usr/share)
-  if(APPLE)
-    find_program(BREW brew)
-    if(BREW)
-      execute_process(COMMAND ${BREW} --prefix "llvm@${ARROW_LLVM_VERSION_PRIMARY_MAJOR}"
-                      OUTPUT_VARIABLE LLVM_BREW_PREFIX
-                      OUTPUT_STRIP_TRAILING_WHITESPACE)
-      if(NOT LLVM_BREW_PREFIX)
-        execute_process(COMMAND ${BREW} --prefix llvm
+  foreach(ARROW_LLVM_VERSION ${ARROW_LLVM_VERSIONS})
+    set(LLVM_HINTS ${LLVM_ROOT} ${LLVM_DIR} /usr/lib /usr/share)
+
+    if(APPLE)
+      find_program(BREW brew)
+      if(BREW)
+        string(REGEX REPLACE "^([0-9]+)(\\..+)?" "\\1" ARROW_LLVM_VERSION_MAJOR
+                             "${ARROW_LLVM_VERSION}")
+        execute_process(COMMAND ${BREW} --prefix "llvm@${ARROW_LLVM_VERSION_MAJOR}"
                         OUTPUT_VARIABLE LLVM_BREW_PREFIX
                         OUTPUT_STRIP_TRAILING_WHITESPACE)
-      endif()
-      if(LLVM_BREW_PREFIX)
         list(APPEND LLVM_HINTS ${LLVM_BREW_PREFIX})
       endif()
     endif()
-  endif()
 
-  foreach(HINT ${LLVM_HINTS})
-    foreach(ARROW_LLVM_VERSION ${ARROW_LLVM_VERSIONS})
-      find_package(LLVM
-                   ${ARROW_LLVM_VERSION}
-                   CONFIG
-                   HINTS
-                   ${HINT})
-      if(LLVM_FOUND)
-        break()
-      endif()
-    endforeach()
+    find_package(LLVM
+                 ${ARROW_LLVM_VERSION}
+                 CONFIG
+                 HINTS
+                 ${LLVM_HINTS})
+
+    if(LLVM_FOUND)
+      break()
+    endif()
   endforeach()
 endif()
 

cpp/src/arrow/util/bpacking.cc

@@ -153,13 +153,14 @@ struct Unpack32DynamicFunction {
   using FunctionType = decltype(&unpack32_default);
 
   static std::vector<std::pair<DispatchLevel, FunctionType>> implementations() {
-    return {
-      { DispatchLevel::NONE, unpack32_default }
+    return {{DispatchLevel::NONE, unpack32_default}
 #if defined(ARROW_HAVE_RUNTIME_AVX2)
-      , { DispatchLevel::AVX2, unpack32_avx2 }
+            ,
+            {DispatchLevel::AVX2, unpack32_avx2}
 #endif
 #if defined(ARROW_HAVE_RUNTIME_AVX512)
-      , { DispatchLevel::AVX512, unpack32_avx512 }
+            ,
+            {DispatchLevel::AVX512, unpack32_avx512}
 #endif
     };
   }

cpp/src/gandiva/GandivaConfig.cmake.in

@@ -27,7 +27,6 @@
 @PACKAGE_INIT@
 
 set(ARROW_LLVM_VERSIONS "@ARROW_LLVM_VERSIONS@")
-set(ARROW_LLVM_VERSION_PRIMARY_MAJOR "@ARROW_LLVM_VERSION_PRIMARY_MAJOR@")
 
 include(CMakeFindDependencyMacro)
 find_dependency(Arrow)

cpp/src/parquet/level_comparison.cc

@@ -44,10 +44,10 @@ struct GreaterThanDynamicFunction {
   using FunctionType = decltype(&GreaterThanBitmap);
 
   static std::vector<std::pair<DispatchLevel, FunctionType>> implementations() {
-    return {
-      { DispatchLevel::NONE, standard::GreaterThanBitmapImpl }
+    return {{DispatchLevel::NONE, standard::GreaterThanBitmapImpl}
 #if defined(ARROW_HAVE_RUNTIME_AVX2)
-      , { DispatchLevel::AVX2, GreaterThanBitmapAvx2 }
+            ,
+            {DispatchLevel::AVX2, GreaterThanBitmapAvx2}
 #endif
     };
   }
@@ -57,10 +57,10 @@ struct MinMaxDynamicFunction {
   using FunctionType = decltype(&FindMinMax);
 
   static std::vector<std::pair<DispatchLevel, FunctionType>> implementations() {
-    return {
-      { DispatchLevel::NONE, standard::FindMinMaxImpl }
+    return {{DispatchLevel::NONE, standard::FindMinMaxImpl}
 #if defined(ARROW_HAVE_RUNTIME_AVX2)
-      , { DispatchLevel::AVX2, FindMinMaxAvx2 }
+            ,
+            {DispatchLevel::AVX2, FindMinMaxAvx2}
 #endif
     };
   }

dev/tasks/verify-rc/github.macos.amd64.yml

@@ -71,8 +71,4 @@ jobs:
           USE_CONDA: 1
         {% endif %}
         run: |
-        {% if not use_conda %}
-          export CC=$(brew --prefix llvm)/bin/clang
-          export CXX=$(brew --prefix llvm)/bin/clang++
-        {% endif %}
           arrow/dev/release/verify-release-candidate.sh {{ release|default("") }} {{ rc|default("") }}

python/CMakeLists.txt

@@ -43,6 +43,12 @@ include(CMakeParseArguments)
 # https://www.cmake.org/cmake/help/latest/policy/CMP0054.html
 cmake_policy(SET CMP0054 NEW)
 
+# find_package() uses <PackageName>_ROOT variables.
+# https://cmake.org/cmake/help/latest/policy/CMP0074.html
+if(POLICY CMP0074)
+  cmake_policy(SET CMP0074 NEW)
+endif()
+
 # Use the first Python installation on PATH, not the newest one
 set(Python3_FIND_STRATEGY "LOCATION")
 # On Windows, use registry last, not first