Dynamically select Bazelisk architecture on macOS CI runners to support Apple Silicon (M1/M2 arm64) runners (#66)

* Integrate wheel build and PyPI publish steps into main build job; remove redundant 8-job pypi-publish matrix

* Simplify wheel build command by omitting default python binary arg

* Dynamically select Bazelisk architecture on macOS CI runners to support Apple Silicon (M1/M2 arm64) runners

* Conditionally compile Arrow AVX SIMD optimization sources only when building for x86_64 CPU architecture to prevent Apple Silicon macOS errors

Author: vkarampudi

.bazelrc

@@ -50,7 +50,6 @@ build:sframe_fix --copt=-Wa,--gsframe=no
 build:sframe_fix --cxxopt=-Wa,--gsframe=no
 build:sframe_fix --host_copt=-Wa,--gsframe=no
 build:sframe_fix --host_cxxopt=-Wa,--gsframe=no
-build:linux --config=sframe_fix
 
 build:macos --macos_minimum_os=10.12
 

.github/workflows/build.yml

@@ -38,7 +38,12 @@ jobs:
         if [ "$RUNNER_OS" == "Linux" ]; then
           curl -Lo /tmp/bazelisk https://github.com/bazelbuild/bazelisk/releases/download/v1.20.0/bazelisk-linux-amd64
         elif [ "$RUNNER_OS" == "macOS" ]; then
-          curl -Lo /tmp/bazelisk https://github.com/bazelbuild/bazelisk/releases/download/v1.20.0/bazelisk-darwin-amd64
+          ARCH=$(uname -m)
+          if [ "$ARCH" == "arm64" ]; then
+            curl -Lo /tmp/bazelisk https://github.com/bazelbuild/bazelisk/releases/download/v1.20.0/bazelisk-darwin-arm64
+          else
+            curl -Lo /tmp/bazelisk https://github.com/bazelbuild/bazelisk/releases/download/v1.20.0/bazelisk-darwin-amd64
+          fi
         fi
         chmod +x /tmp/bazelisk
         sudo mv /tmp/bazelisk /usr/local/bin/bazel
@@ -63,6 +68,18 @@ jobs:
       shell: bash -l {0}
       run: bazel build //...
 
+    - name: Build Release Wheel
+      shell: bash -l {0}
+      run: bazel run //:build_pip_package
+
+    - name: Publish Package to PyPI
+      if: github.event_name == 'push' && (startsWith(github.ref, 'refs/tags/v') || startsWith(github.ref, 'refs/heads/r'))
+      uses: pypa/gh-action-pypi-publish@release/v1
+      with:
+        password: ${{ secrets.PYPI_API_TOKEN }}
+        packages-dir: dist/
+        skip-existing: true
+
   docker-serving-build:
     name: Build Docker Serving Image
     if: github.event_name == 'push' && github.ref == 'refs/heads/master'
@@ -77,52 +94,3 @@ jobs:
       run: |
         cd struct2tensor/tools/tf_serving_docker
         docker build -t struct2tensor-serving:latest .
-
-  pypi-publish:
-    name: Build and Publish Wheels to PyPI
-    if: github.event_name == 'push' && (startsWith(github.ref, 'refs/tags/v') || startsWith(github.ref, 'refs/heads/r'))
-    runs-on: ${{ matrix.os }}
-    strategy:
-      matrix:
-        os: [ubuntu-latest, macos-latest]
-        python-version: ["3.10", "3.11", "3.12", "3.13"]
-    steps:
-    - uses: actions/checkout@v4
-
-    - name: Set up conda environment
-      uses: conda-incubator/setup-miniconda@v3
-      with:
-        auto-activate-base: false
-        activate-environment: s2t-env
-        environment-file: environment.yml
-        python-version: ${{ matrix.python-version }}
-        use-mamba: true
-
-    - name: Install Bazel
-      shell: bash -l {0}
-      run: |
-        if [ "$RUNNER_OS" == "Linux" ]; then
-          curl -Lo /tmp/bazelisk https://github.com/bazelbuild/bazelisk/releases/download/v1.20.0/bazelisk-linux-amd64
-        elif [ "$RUNNER_OS" == "macOS" ]; then
-          curl -Lo /tmp/bazelisk https://github.com/bazelbuild/bazelisk/releases/download/v1.20.0/bazelisk-darwin-amd64
-        fi
-        chmod +x /tmp/bazelisk
-        sudo mv /tmp/bazelisk /usr/local/bin/bazel
-
-    - name: Patch Apple CC toolchain (macOS only)
-      if: runner.os == 'macOS'
-      shell: bash -l {0}
-      run: bazel build //:patch_local_config_apple_cc
-
-    - name: Build Release Wheel
-      shell: bash -l {0}
-      run: |
-        ./configure.sh
-        bazel run --config=sframe_fix //:build_pip_package -- --python_bin_path python
-
-    - name: Publish Package to PyPI
-      uses: pypa/gh-action-pypi-publish@release/v1
-      with:
-        password: ${{ secrets.PYPI_API_TOKEN }}
-        packages-dir: dist/
-        skip-existing: true

third_party/arrow.BUILD

@@ -53,6 +53,14 @@ fbs_headers = [
     "cpp/src/generated/Tensor_generated.h",
 ]
 
+arrow_simd_srcs = select({
+    "@platforms//cpu:x86_64": glob([
+        "cpp/src/arrow/compute/**/*_avx2.cc",
+        "cpp/src/arrow/compute/**/*_avx512.cc",
+    ]),
+    "//conditions:default": [],
+})
+
 cc_library(
     name = "arrow",
     srcs = glob(
@@ -87,6 +95,8 @@ cc_library(
             "cpp/src/arrow/util/bpacking_avx512.*",
             "cpp/src/arrow/util/bpacking_neon*",
             "cpp/src/arrow/util/bpacking_simd*",
+            "cpp/src/arrow/compute/**/*_avx2.*",
+            "cpp/src/arrow/compute/**/*_avx512.*",
             "cpp/src/arrow/util/compression_brotli*",
             "cpp/src/arrow/util/compression_bz2*",
             "cpp/src/arrow/util/compression_lz4*",
@@ -106,7 +116,7 @@ cc_library(
             #"cpp/src/arrow/vendored/datetime/**",
             "cpp/src/parquet/encryption/encryption_internal.cc",
         ],
-    ),
+    ) + arrow_simd_srcs,
     hdrs = [
         # Headers from above genrules.
         "cpp/src/arrow/util/config.h",