fix(python): Fix python wheel error by disabling s3 (#826)

* trt

* remove sudo

* test

* try build graphar cpp

* try build graphar cpp

* remove sudo

* remove sudo

* try

* try

* try close s3

* try close s3

* try

* try

* open arrow s3 in macos

* try

* update

* fix smoke test

* only run in cp39

* try fix

* update

* format

* format

Author: yangxk1

.github/workflows/python-wheel-workflow.yml

@@ -130,27 +130,28 @@ jobs:
         fi
         dnf install -y 'dnf-command(config-manager)' || dnf install -y dnf-plugins-core || true
         # Follow official Apache Arrow install instructions for AlmaLinux/RHEL-family
-        dnf install -y epel-release || \
-          dnf install -y oracle-epel-release-el${ALMA_MAJOR} || \
-          dnf install -y https://dl.fedoraproject.org/pub/epel/epel-release-latest-${ALMA_MAJOR}.noarch.rpm
-        dnf install -y https://packages.apache.org/artifactory/arrow/almalinux/${ALMA_MAJOR}/apache-arrow-release-latest.rpm
+        dnf install -y epel-release || dnf install -y oracle-epel-release-el$(cut -d: -f5 /etc/system-release-cpe | cut -d. -f1) || dnf install -y https://dl.fedoraproject.org/pub/epel/epel-release-latest-$(cut -d: -f5 /etc/system-release-cpe | cut -d. -f1).noarch.rpm
+        dnf install -y https://packages.apache.org/artifactory/arrow/almalinux/$(cut -d: -f5 /etc/system-release-cpe | cut -d. -f1)/apache-arrow-release-latest.rpm
         dnf config-manager --set-enabled epel || :
         dnf config-manager --set-enabled powertools || :
         dnf config-manager --set-enabled crb || :
-        dnf config-manager --set-enabled ol${ALMA_MAJOR}_codeready_builder || :
-        dnf config-manager --set-enabled codeready-builder-for-rhel-${ALMA_MAJOR}-rhui-rpms || :
-        subscription-manager repos --enable codeready-builder-for-rhel-${ALMA_MAJOR}-$(arch)-rpms || :
-        # manylinux images may carry older Arrow packages (e.g. arrow1700-*) which
-        # conflict with the newer packages from the Apache Arrow repo (e.g. arrow2200-*).
-        # Remove any preinstalled Arrow/Parquet RPMs so we install a consistent set.
-        dnf remove -y 'arrow*' 'parquet*' || true
-        # Required for GraphAr C++ build via Arrow CMake packages
-        dnf install -y --allowerasing \
-          arrow-devel \
-          arrow-dataset-devel \
-          arrow-acero-devel \
-          parquet-devel \
-          libcurl-devel re2-devel ccache
+        dnf config-manager --set-enabled ol$(cut -d: -f5 /etc/system-release-cpe | cut -d. -f1)_codeready_builder || :
+        dnf config-manager --set-enabled codeready-builder-for-rhel-$(cut -d: -f5 /etc/system-release-cpe | cut -d. -f1)-rhui-rpms || :
+        subscription-manager repos --enable codeready-builder-for-rhel-$(cut -d: -f5 /etc/system-release-cpe | cut -d. -f1)-$(arch)-rpms || :
+        dnf install -y arrow-devel # For C++
+        dnf install -y arrow-glib-devel # For GLib (C)
+        dnf install -y arrow-dataset-devel # For Apache Arrow Dataset C++
+        dnf install -y arrow-dataset-glib-devel # For Apache Arrow Dataset GLib (C)
+        dnf install -y arrow-acero-devel # For Apache Arrow Acero C++
+        dnf install -y arrow-flight-devel # For Apache Arrow Flight C++
+        dnf install -y arrow-flight-glib-devel # For Apache Arrow Flight GLib (C)
+        dnf install -y arrow-flight-sql-devel # For Apache Arrow Flight SQL C++
+        dnf install -y arrow-flight-sql-glib-devel # For Apache Arrow Flight SQL GLib (C)
+        dnf install -y gandiva-devel # For Apache Gandiva C++
+        dnf install -y gandiva-glib-devel # For Apache Gandiva GLib (C)
+        dnf install -y parquet-devel # For Apache Parquet C++
+        dnf install -y parquet-glib-devel # For Apache Parquet GLib (C)
+
     steps:
       - name: Checkout (needed for some tooling)
         uses: actions/checkout@v4
@@ -184,6 +185,7 @@ jobs:
               otool -l "$CONDA/lib/libarrow.dylib" | (grep -A3 -E 'LC_BUILD_VERSION|LC_VERSION_MIN_MACOSX' || true)
             fi
           fi
+
       - name: Install Arrow (Windows)
         if: matrix.os == 'windows'
         shell: pwsh
@@ -224,6 +226,7 @@ jobs:
             exit 1
           fi
           echo "PKGDIR=$PKGDIR" >> "$GITHUB_ENV"
+
       - name: Build wheels
         shell: bash
         run: |
@@ -232,12 +235,24 @@ jobs:
           python -m pip install packaging cibuildwheel
           mkdir -p python/dist
           python -m cibuildwheel --output-dir python/dist "$PKGDIR"
+
       - name: Store artifacts
         uses: actions/upload-artifact@v4
         with:
           name: wheels-${{ matrix.os }}-${{ matrix.platform }}
           path: python/dist/*
 
+      - name: Smoke test wheel
+        shell: bash
+        run: |
+          set -euxo pipefail
+          python -m pip uninstall -y graphar || true
+
+          python -m pip install --find-links python/dist graphar
+
+          python -c "import graphar; import graphar._core; print('GraphAr imported successfully')"
+          graphar --help >/dev/null
+
   upload_test_pypi:
     name: Publish to TestPyPI (auto)
     needs: [build_wheels, build_sdist]

cpp/src/graphar/filesystem.cc

@@ -370,20 +370,26 @@ Result<std::shared_ptr<FileSystem>> FileSystemFromUriOrPath(
 // arrow::fs::InitializeS3 and arrow::fs::FinalizeS3 need arrow_version >= 15
 Status InitializeS3() {
 #if defined(ARROW_VERSION) && ARROW_VERSION > 14000000
+#if defined(ARROW_S3)
   auto options = arrow::fs::S3GlobalOptions::Defaults();
+#endif
 #else
   arrow::fs::S3GlobalOptions options;
   options.log_level = arrow::fs::S3LogLevel::Fatal;
 #endif
 #if defined(ARROW_VERSION) && ARROW_VERSION >= 15000000
+#if defined(ARROW_S3)
   RETURN_NOT_ARROW_OK(arrow::fs::InitializeS3(options));
+#endif
 #endif
   return Status::OK();
 }
 
 Status FinalizeS3() {
 #if defined(ARROW_VERSION) && ARROW_VERSION >= 15000000
+#if defined(ARROW_S3)
   RETURN_NOT_ARROW_OK(arrow::fs::FinalizeS3());
+#endif
 #endif
   return Status::OK();
 }

python/pyproject.toml

@@ -22,7 +22,7 @@ build-backend = "scikit_build_core.build"
 
 [project]
 name = "graphar"
-version = "0.13.0-dev"
+version = "0.13.0.dev1"
 description = "An open source, standard data file format for graph data storage and retrieval."
 readme = "README.md"
 authors = [{ name = "GraphAr community", email = "dev@graphar.apache.org" }]