apacheGH-39352: [FS][Azure] Enable azure in builds (apache#39971)

### Rationale for this change ### What changes are included in this PR? Enable Azure in linux and mac os wheel builds. Tried to copy GCS Don't enable Azure for windows builds because windows builds where all failing. Failures were a combination of cmake version or `Could not find a package configuration file provided by "wil"`. I think it makes sense to come back to windows builds in another PR. ### Are these changes tested? There is no new functionality to test. ### Are there any user-facing changes? No * Closes: apache#39352 Authored-by: Thomas Newton <[email protected]> Signed-off-by: Sutou Kouhei <[email protected]>
dgreiss · Feb 17, 2024 · f86f122 · f86f122
1 parent 2d957fd
commit f86f122
Show file tree

Hide file tree

Showing 10 changed files with 26 additions and 2 deletions.
diff --git a/ci/docker/python-wheel-manylinux.dockerfile b/ci/docker/python-wheel-manylinux.dockerfile
@@ -82,6 +82,7 @@ RUN vcpkg install \
         --clean-after-build \
         --x-install-root=${VCPKG_ROOT}/installed \
         --x-manifest-root=/arrow/ci/vcpkg \
+        --x-feature=azure \ 
         --x-feature=flight \
         --x-feature=gcs \
         --x-feature=json \

diff --git a/ci/scripts/cpp_build.sh b/ci/scripts/cpp_build.sh
@@ -54,6 +54,7 @@ if [ "${GITHUB_ACTIONS:-false}" = "true" ]; then
 fi
 
 if [ "${ARROW_ENABLE_THREADING:-ON}" = "OFF" ]; then
+  ARROW_AZURE=OFF
   ARROW_FLIGHT=OFF
   ARROW_FLIGHT_SQL=OFF
   ARROW_GCS=OFF

diff --git a/ci/scripts/python_build.sh b/ci/scripts/python_build.sh
@@ -55,6 +55,7 @@ export PYARROW_CMAKE_GENERATOR=${CMAKE_GENERATOR:-Ninja}
 export PYARROW_BUILD_TYPE=${CMAKE_BUILD_TYPE:-debug}
 
 export PYARROW_WITH_ACERO=${ARROW_ACERO:-OFF}
+export PYARROW_WITH_AZURE=${ARROW_AZURE:-OFF}
 export PYARROW_WITH_CUDA=${ARROW_CUDA:-OFF}
 export PYARROW_WITH_DATASET=${ARROW_DATASET:-ON}
 export PYARROW_WITH_FLIGHT=${ARROW_FLIGHT:-OFF}

diff --git a/ci/scripts/python_test.sh b/ci/scripts/python_test.sh
@@ -39,6 +39,7 @@ fi
 
 # By default, force-test all optional components
 : ${PYARROW_TEST_ACERO:=${ARROW_ACERO:-ON}}
+: ${PYARROW_TEST_AZURE:=${ARROW_AZURE:-ON}}
 : ${PYARROW_TEST_CUDA:=${ARROW_CUDA:-ON}}
 : ${PYARROW_TEST_DATASET:=${ARROW_DATASET:-ON}}
 : ${PYARROW_TEST_FLIGHT:=${ARROW_FLIGHT:-ON}}

diff --git a/ci/scripts/python_wheel_macos_build.sh b/ci/scripts/python_wheel_macos_build.sh
@@ -63,6 +63,7 @@ pip install "delocate>=0.10.3"
 
 echo "=== (${PYTHON_VERSION}) Building Arrow C++ libraries ==="
 : ${ARROW_ACERO:=ON}
+: ${ARROW_AZURE:=ON}
 : ${ARROW_DATASET:=ON}
 : ${ARROW_FLIGHT:=ON}
 : ${ARROW_GANDIVA:=OFF}
@@ -95,6 +96,7 @@ pushd ${build_dir}/build
 
 cmake \
     -DARROW_ACERO=${ARROW_ACERO} \
+    -DARROW_AZURE=${ARROW_AZURE} \
     -DARROW_BUILD_SHARED=ON \
     -DARROW_BUILD_STATIC=OFF \
     -DARROW_BUILD_TESTS=OFF \
@@ -148,6 +150,7 @@ export PYARROW_BUNDLE_ARROW_CPP=1
 export PYARROW_CMAKE_GENERATOR=${CMAKE_GENERATOR}
 export PYARROW_INSTALL_TESTS=1
 export PYARROW_WITH_ACERO=${ARROW_ACERO}
+export PYARROW_WITH_AZURE=${ARROW_AZURE}
 export PYARROW_WITH_DATASET=${ARROW_DATASET}
 export PYARROW_WITH_FLIGHT=${ARROW_FLIGHT}
 export PYARROW_WITH_GANDIVA=${ARROW_GANDIVA}

diff --git a/ci/scripts/python_wheel_manylinux_build.sh b/ci/scripts/python_wheel_manylinux_build.sh
@@ -49,6 +49,7 @@ rm -rf /arrow/python/pyarrow/*.so.*
 
 echo "=== (${PYTHON_VERSION}) Building Arrow C++ libraries ==="
 : ${ARROW_ACERO:=ON}
+: ${ARROW_AZURE:=ON}
 : ${ARROW_DATASET:=ON}
 : ${ARROW_FLIGHT:=ON}
 : ${ARROW_GANDIVA:=OFF}
@@ -87,6 +88,7 @@ pushd /tmp/arrow-build
 
 cmake \
     -DARROW_ACERO=${ARROW_ACERO} \
+    -DARROW_AZURE=${ARROW_AZURE} \
     -DARROW_BUILD_SHARED=ON \
     -DARROW_BUILD_STATIC=OFF \
     -DARROW_BUILD_TESTS=OFF \
@@ -141,6 +143,7 @@ export PYARROW_BUNDLE_ARROW_CPP=1
 export PYARROW_CMAKE_GENERATOR=${CMAKE_GENERATOR}
 export PYARROW_INSTALL_TESTS=1
 export PYARROW_WITH_ACERO=${ARROW_ACERO}
+export PYARROW_WITH_AZURE=${ARROW_AZURE}
 export PYARROW_WITH_DATASET=${ARROW_DATASET}
 export PYARROW_WITH_FLIGHT=${ARROW_FLIGHT}
 export PYARROW_WITH_GANDIVA=${ARROW_GANDIVA}

diff --git a/ci/scripts/python_wheel_unix_test.sh b/ci/scripts/python_wheel_unix_test.sh
@@ -28,15 +28,17 @@ fi
 
 source_dir=${1}
 
+: ${ARROW_AZURE:=ON}
 : ${ARROW_FLIGHT:=ON}
-: ${ARROW_SUBSTRAIT:=ON}
-: ${ARROW_S3:=ON}
 : ${ARROW_GCS:=ON}
+: ${ARROW_S3:=ON}
+: ${ARROW_SUBSTRAIT:=ON}
 : ${CHECK_IMPORTS:=ON}
 : ${CHECK_UNITTESTS:=ON}
 : ${INSTALL_PYARROW:=ON}
 
 export PYARROW_TEST_ACERO=ON
+export PYARROW_TEST_AZURE=${ARROW_AZURE}
 export PYARROW_TEST_CYTHON=OFF
 export PYARROW_TEST_DATASET=ON
 export PYARROW_TEST_FLIGHT=${ARROW_FLIGHT}

diff --git a/ci/vcpkg/vcpkg.json b/ci/vcpkg/vcpkg.json
@@ -105,6 +105,16 @@
         }
       ]
     },
+    "azure": {
+      "description": "Azure blob storage support",
+      "dependencies": [
+        "azure-core-cpp",
+        "azure-identity-cpp",
+        "azure-storage-blobs-cpp",
+        "azure-storage-common-cpp",
+        "azure-storage-files-datalake-cpp"
+      ]
+    },
     "orc": {
       "description": "ORC support",
       "dependencies": [

diff --git a/dev/tasks/python-wheels/github.osx.amd64.yml b/dev/tasks/python-wheels/github.osx.amd64.yml
@@ -85,6 +85,7 @@ jobs:
             --clean-after-build \
             --x-install-root=${VCPKG_ROOT}/installed \
             --x-manifest-root=arrow/ci/vcpkg \
+            --x-feature=azure \
             --x-feature=flight \
             --x-feature=gcs \
             --x-feature=json \

diff --git a/dev/tasks/python-wheels/github.osx.arm64.yml b/dev/tasks/python-wheels/github.osx.arm64.yml
@@ -71,6 +71,7 @@ jobs:
             --clean-after-build \
             --x-install-root=${VCPKG_ROOT}/installed \
             --x-manifest-root=arrow/ci/vcpkg \
+            --x-feature=azure \
             --x-feature=flight \
             --x-feature=gcs \
             --x-feature=json \