GH-43951: [CI][Python] Use GitHub Packages for vcpkg cache (#44644)

### Rationale for this change We're using only Docker level cache for vcpkg used for wheels. If we have any vcpkg related changes, all vcpkg ports are rebuilt. It's time consuming. ### What changes are included in this PR? Enable NuGet + GitHub Packages based cache. It's port level cache. So we don't need to rebuild all ports when we have any vcpkg related changes. See also: https://learn.microsoft.com/en-us/vcpkg/consume/binary-caching-github-packages NuGet + GitHub Packages based cache isn't enabled with manylinux2014 + aarch64. Because EPEL for CentOS 7 + aarch64 provides old Mono. (FYI: EPEL for CentOS 7 + x86_64 provides newer Mono.) We can't use old Mono to run NuGet on Linux. ### Are these changes tested? Yes. ### Are there any user-facing changes? No. * GitHub Issue: #43951 Lead-authored-by: Sutou Kouhei <[email protected]> Co-authored-by: Sutou Kouhei <[email protected]> Co-authored-by: Raúl Cumplido <[email protected]> Signed-off-by: Sutou Kouhei <[email protected]>
apache · Nov 15, 2024 · df40f7a · df40f7a
1 parent 736d706
commit df40f7a
Show file tree

Hide file tree

Showing 7 changed files with 117 additions and 47 deletions.
diff --git a/ci/docker/python-wheel-manylinux.dockerfile b/ci/docker/python-wheel-manylinux.dockerfile
@@ -69,36 +69,47 @@ RUN /arrow/ci/scripts/install_ccache.sh ${ccache} /usr/local
 ARG vcpkg
 COPY ci/vcpkg/*.patch \
      ci/vcpkg/*linux*.cmake \
+     ci/vcpkg/vcpkg.json \
      arrow/ci/vcpkg/
 COPY ci/scripts/install_vcpkg.sh \
      arrow/ci/scripts/
 ENV VCPKG_ROOT=/opt/vcpkg
 ARG build_type=release
 ENV CMAKE_BUILD_TYPE=${build_type} \
-    VCPKG_FORCE_SYSTEM_BINARIES=1 \
-    VCPKG_OVERLAY_TRIPLETS=/arrow/ci/vcpkg \
+    PATH="${PATH}:${VCPKG_ROOT}" \
     VCPKG_DEFAULT_TRIPLET=${arch_short}-linux-static-${build_type} \
-    VCPKG_FEATURE_FLAGS="manifests"
-
-RUN arrow/ci/scripts/install_vcpkg.sh ${VCPKG_ROOT} ${vcpkg}
-ENV PATH="${PATH}:${VCPKG_ROOT}"
-
-COPY ci/vcpkg/vcpkg.json arrow/ci/vcpkg/
-# cannot use the S3 feature here because while aws-sdk-cpp=1.9.160 contains
-# ssl related fixes as well as we can patch the vcpkg portfile to support
-# arm machines it hits ARROW-15141 where we would need to fall back to 1.8.186
-# but we cannot patch those portfiles since vcpkg-tool handles the checkout of
-# previous versions => use bundled S3 build
-RUN vcpkg install \
+    VCPKG_FEATURE_FLAGS="manifests" \
+    VCPKG_FORCE_SYSTEM_BINARIES=1 \
+    VCPKG_OVERLAY_TRIPLETS=/arrow/ci/vcpkg
+# For --mount=type=secret: The GITHUB_TOKEN is the only real secret but we use
+# --mount=type=secret for GITHUB_REPOSITORY_OWNER and
+# VCPKG_BINARY_SOURCES too because we don't want to store them
+# into the built image in order to easily reuse the built image cache.
+#
+# For vcpkg install: cannot use the S3 feature here because while
+# aws-sdk-cpp=1.9.160 contains ssl related fixes as well as we can
+# patch the vcpkg portfile to support arm machines it hits ARROW-15141
+# where we would need to fall back to 1.8.186 but we cannot patch
+# those portfiles since vcpkg-tool handles the checkout of previous
+# versions => use bundled S3 build
+RUN --mount=type=secret,id=github_repository_owner \
+    --mount=type=secret,id=github_token \
+    --mount=type=secret,id=vcpkg_binary_sources \
+      export GITHUB_REPOSITORY_OWNER=$(cat /run/secrets/github_repository_owner); \
+      export GITHUB_TOKEN=$(cat /run/secrets/github_token); \
+      export VCPKG_BINARY_SOURCES=$(cat /run/secrets/vcpkg_binary_sources); \
+      arrow/ci/scripts/install_vcpkg.sh ${VCPKG_ROOT} ${vcpkg} && \
+      vcpkg install \
         --clean-after-build \
         --x-install-root=${VCPKG_ROOT}/installed \
         --x-manifest-root=/arrow/ci/vcpkg \
-        --x-feature=azure \ 
+        --x-feature=azure \
         --x-feature=flight \
         --x-feature=gcs \
         --x-feature=json \
         --x-feature=parquet \
-        --x-feature=s3
+        --x-feature=s3 && \
+      rm -rf ~/.config/NuGet/
 
 # Make sure auditwheel is up-to-date
 RUN pipx upgrade auditwheel

diff --git a/ci/scripts/install_vcpkg.sh b/ci/scripts/install_vcpkg.sh
@@ -17,7 +17,7 @@
 # specific language governing permissions and limitations
 # under the License.
 
-set -e
+set -eu
 
 if [ "$#" -lt 1 ]; then
   echo "Usage: $0 ``<target-directory> [<vcpkg-version> [<vcpkg-ports-patch>]]"
@@ -42,7 +42,7 @@ pushd ${vcpkg_destination}
 
 git checkout "${vcpkg_version}"
 
-if [[ "$OSTYPE" == "msys" ]]; then
+if [[ "${OSTYPE:-}" == "msys" ]]; then
   ./bootstrap-vcpkg.bat -disableMetrics
 else
   ./bootstrap-vcpkg.sh -disableMetrics
@@ -53,4 +53,33 @@ if [ -f "${vcpkg_ports_patch}" ]; then
   echo "Patch successfully applied to the VCPKG port files!"
 fi
 
+if [ -n "${GITHUB_TOKEN:-}" ] && \
+     [ -n "${GITHUB_REPOSITORY_OWNER:-}" ] && \
+     [ "${VCPKG_BINARY_SOURCES:-}" = "clear;nuget,GitHub,readwrite" ] ; then
+  if type dnf 2>/dev/null; then
+    dnf install -y epel-release
+    dnf install -y mono-complete
+    curl \
+      --location \
+      --output "${vcpkg_destination}/nuget" \
+      https://dist.nuget.org/win-x86-commandline/latest/nuget.exe
+  fi
+  PATH="${vcpkg_destination}:${PATH}"
+  nuget_url="https://nuget.pkg.github.com/${GITHUB_REPOSITORY_OWNER}/index.json"
+  nuget="$(vcpkg fetch nuget | tail -n 1)"
+  if type mono 2>/dev/null; then
+    nuget="mono ${nuget}"
+  fi
+  ${nuget} \
+    sources add \
+    -source "${nuget_url}" \
+    -storepasswordincleartext \
+    -name "GitHub" \
+    -username "${GITHUB_REPOSITORY_OWNER}" \
+    -password "${GITHUB_TOKEN}"
+  ${nuget} \
+    setapikey "${GITHUB_TOKEN}" \
+    -source "${nuget_url}"
+fi
+
 popd
diff --git a/dev/tasks/java-jars/github.yml b/dev/tasks/java-jars/github.yml
@@ -19,6 +19,9 @@
 
 {{ macros.github_header() }}
 
+permissions:
+  packages: write
+
 jobs:
 
   build-cpp-ubuntu:
@@ -51,7 +54,14 @@ jobs:
       - name: Build C++ libraries
         env:
         {{ macros.github_set_sccache_envvars()|indent(8) }}
+          GITHUB_TOKEN: {{ '${{ secrets.GITHUB_TOKEN }}' }}
         run: |
+          if [ "${ARCH}" = "arm64v8" ]; then
+            # We can't use NuGet on manylinux2014_aarch64 because Mono is old.
+            :
+          else
+            export VCPKG_BINARY_SOURCES="clear;nuget,GitHub,readwrite"
+          fi
           archery docker run \
             -e ARROW_JAVA_BUILD=OFF \
             -e ARROW_JAVA_TEST=OFF \

diff --git a/dev/tasks/python-wheels/github.linux.yml b/dev/tasks/python-wheels/github.linux.yml
@@ -19,6 +19,9 @@
 
 {{ macros.github_header() }}
 
+permissions:
+  packages: write
+
 jobs:
   build:
     name: "Build wheel for manylinux {{ manylinux_version }}"
@@ -49,7 +52,18 @@ jobs:
 
       - name: Build wheel
         shell: bash
-        run: archery docker run -e SETUPTOOLS_SCM_PRETEND_VERSION={{ arrow.no_rc_version }} python-wheel-manylinux-{{ manylinux_version }}
+        env:
+          GITHUB_TOKEN: {{ '${{ secrets.GITHUB_TOKEN }}' }}
+        run: |
+          if [ "{{ manylinux_version }}" = "2014" ] && [ "{{ arch }}" = "arm64" ]; then
+            # We can't use NuGet on manylinux2014_aarch64 because Mono is old.
+            :
+          else
+            export VCPKG_BINARY_SOURCES="clear;nuget,GitHub,readwrite"
+          fi
+          archery docker run \
+            -e SETUPTOOLS_SCM_PRETEND_VERSION={{ arrow.no_rc_version }} \
+            python-wheel-manylinux-{{ manylinux_version }}
 
       - uses: actions/upload-artifact@v4
         with:

diff --git a/dev/tasks/python-wheels/github.osx.yml b/dev/tasks/python-wheels/github.osx.yml
@@ -37,6 +37,9 @@
   VCPKG_OVERLAY_TRIPLETS: {{ "${{ github.workspace }}/arrow/ci/vcpkg" }}
   VCPKG_ROOT: {{ "${{ github.workspace }}/vcpkg" }}
 
+permissions:
+  packages: write
+
 jobs:
   build:
     name: Build wheel for Python {{ python_version }} on macOS
@@ -69,27 +72,13 @@ jobs:
           echo "VCPKG_VERSION=$vcpkg_version" >> $GITHUB_ENV
 
       - name: Install Vcpkg
+        env:
+          GITHUB_TOKEN: {{ '${{ secrets.GITHUB_TOKEN }}' }}
         run: arrow/ci/scripts/install_vcpkg.sh $VCPKG_ROOT $VCPKG_VERSION
 
       - name: Add Vcpkg to PATH
         run: echo ${VCPKG_ROOT} >> $GITHUB_PATH
 
-      - name: Setup NuGet Credentials
-        env:
-          GITHUB_TOKEN: {{ '${{ secrets.GITHUB_TOKEN }}' }}
-        run: |
-          mono $(vcpkg fetch nuget | tail -n 1) \
-            sources add \
-            -source "https://nuget.pkg.github.com/$GITHUB_REPOSITORY_OWNER/index.json" \
-            -storepasswordincleartext \
-            -name "GitHub" \
-            -username "$GITHUB_REPOSITORY_OWNER" \
-            -password "$GITHUB_TOKEN" \
-
-          mono $(vcpkg fetch nuget | tail -n 1) \
-            setapikey "$GITHUB_TOKEN" \
-            -source "https://nuget.pkg.github.com/$GITHUB_REPOSITORY_OWNER/index.json"
-
       - name: Install Packages
         run: |
           vcpkg install \

diff --git a/dev/tasks/python-wheels/github.windows.yml b/dev/tasks/python-wheels/github.windows.yml
@@ -33,7 +33,9 @@ jobs:
       # note that we don't run docker build since there wouldn't be a cache hit
       # and rebuilding the dependencies takes a fair amount of time
       REPO: ghcr.io/ursacomputing/arrow
-      # BuildKit isn't really supported on Windows for now
+      # BuildKit isn't really supported on Windows for now.
+      # NuGet + GitHub Packages based vcpkg cache is also disabled for now.
+      # Because secret mount requires BuildKit.
       DOCKER_BUILDKIT: 0
 
     steps:

diff --git a/docker-compose.yml b/docker-compose.yml
@@ -53,26 +53,31 @@
 #
 # See more in cpp/build-support/run-test.sh::print_coredumps
 
-x-common: &common
-  GITHUB_ACTIONS:
-
 x-ccache: &ccache
   CCACHE_COMPILERCHECK: content
   CCACHE_COMPRESS: 1
   CCACHE_COMPRESSLEVEL: 6
   CCACHE_MAXSIZE: 1G
   CCACHE_DIR: /ccache
 
+x-common: &common
+  GITHUB_ACTIONS:
+
+x-cpp: &cpp
+  ARROW_RUNTIME_SIMD_LEVEL:
+  ARROW_SIMD_LEVEL:
+
 x-sccache: &sccache
   AWS_ACCESS_KEY_ID:
   AWS_SECRET_ACCESS_KEY:
   SCCACHE_BUCKET:
   SCCACHE_REGION:
   SCCACHE_S3_KEY_PREFIX: ${SCCACHE_S3_KEY_PREFIX:-sccache}
 
-x-cpp: &cpp
-  ARROW_RUNTIME_SIMD_LEVEL:
-  ARROW_SIMD_LEVEL:
+x-vcpkg-build-secrets: &vcpkg-build-secrets
+  - github_repository_owner
+  - github_token
+  - vcpkg_binary_sources
 
 # CPU/memory limit presets to pass to Docker.
 #
@@ -1123,14 +1128,15 @@ services:
         arch: ${ARCH}
         arch_short: ${ARCH_SHORT}
         base: quay.io/pypa/manylinux2014_${ARCH_ALIAS}:2024-08-03-32dfa47
-        vcpkg: ${VCPKG}
+        manylinux: 2014
         python: ${PYTHON}
         python_abi_tag: ${PYTHON_ABI_TAG}
-        manylinux: 2014
+        vcpkg: ${VCPKG}
       context: .
       dockerfile: ci/docker/python-wheel-manylinux.dockerfile
       cache_from:
         - ${REPO}:${ARCH}-python-${PYTHON}-wheel-manylinux-2014-vcpkg-${VCPKG}
+      secrets: *vcpkg-build-secrets
     environment:
       <<: [*common, *ccache]
     volumes:
@@ -1147,14 +1153,15 @@ services:
         arch: ${ARCH}
         arch_short: ${ARCH_SHORT}
         base: quay.io/pypa/manylinux_2_28_${ARCH_ALIAS}:2024-08-03-32dfa47
-        vcpkg: ${VCPKG}
+        manylinux: 2_28
         python: ${PYTHON}
         python_abi_tag: ${PYTHON_ABI_TAG}
-        manylinux: 2_28
+        vcpkg: ${VCPKG}
       context: .
       dockerfile: ci/docker/python-wheel-manylinux.dockerfile
       cache_from:
         - ${REPO}:${ARCH}-python-${PYTHON}-wheel-manylinux-2-28-vcpkg-${VCPKG}
+      secrets: *vcpkg-build-secrets
     environment:
       <<: [*common, *ccache]
     volumes:
@@ -1239,8 +1246,8 @@ services:
     image: ${REPO}:python-${PYTHON}-wheel-windows-vs2019-vcpkg-${VCPKG}-${PYTHON_WHEEL_WINDOWS_IMAGE_REVISION}
     build:
       args:
-        vcpkg: ${VCPKG}
         python: ${PYTHON}
+        vcpkg: ${VCPKG}
       context: .
       dockerfile: ci/docker/python-wheel-windows-vs2019.dockerfile
       # This should make the pushed images reusable, but the image gets rebuilt.
@@ -2119,3 +2126,11 @@ services:
       /bin/bash -c "
         git config --global --add safe.directory /arrow &&
         /arrow/dev/release/verify-release-candidate.sh $${VERIFY_VERSION} $${VERIFY_RC}"
+
+secrets:
+  github_repository_owner:
+    environment: GITHUB_REPOSITORY_OWNER
+  github_token:
+    environment: GITHUB_TOKEN
+  vcpkg_binary_sources:
+    environment: VCPKG_BINARY_SOURCES