GH-7: Add macOS, Windows CI (#432)

Fixes #7.
Fixes #8.

.github/workflows/test.yml

@@ -61,7 +61,7 @@ jobs:
         uses: actions/cache@6849a6489940f00c2f30c0fb92c6274307ccb58a # v4.1.2
         with:
           path: .docker
-          key: maven-${{ matrix.jdk }}-${{ matrix.maven }}-${{ hashFiles('**/docker-compose.yml') }}
+          key: maven-${{ matrix.jdk }}-${{ matrix.maven }}-${{ hashFiles('**/docker-compose.yml', '**/pom.xml') }}
           restore-keys: maven-${{ matrix.jdk }}-${{ matrix.maven }}-
       - name: Execute Docker Build
         env:
@@ -72,3 +72,71 @@ jobs:
             -e CI=true \
             -e "DEVELOCITY_ACCESS_KEY=$DEVELOCITY_ACCESS_KEY" \
             ${{ matrix.image }}
+
+  macos:
+    name: ${{ matrix.arch }} macOS ${{ matrix.macos }} Java JDK ${{ matrix.jdk }}
+    runs-on: macos-${{ matrix.macos }}
+    if: ${{ !contains(github.event.pull_request.title, 'WIP') }}
+    timeout-minutes: 30
+    strategy:
+      fail-fast: false
+      matrix:
+        include:
+          - arch: AMD64
+            jdk: 11
+            macos: 13
+          - arch: AArch64
+            jdk: 11
+            macos: latest
+    steps:
+      - name: Set up Java
+        uses: actions/setup-java@v4
+        with:
+          distribution: 'temurin'
+          java-version: ${{ matrix.jdk }}
+      - name: Checkout Arrow
+        uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+          submodules: recursive
+      - name: Build
+        shell: bash
+        env:
+          DEVELOCITY_ACCESS_KEY: ${{ secrets.GE_ACCESS_TOKEN }}
+        run: ci/scripts/java_build.sh $(pwd) $(pwd)/build
+      - name: Test
+        shell: bash
+        env:
+          DEVELOCITY_ACCESS_KEY: ${{ secrets.GE_ACCESS_TOKEN }}
+        run: ci/scripts/java_test.sh $(pwd) $(pwd)/build
+
+  windows:
+    name: AMD64 Windows Server 2022 Java JDK ${{ matrix.jdk }}
+    runs-on: windows-latest
+    if: ${{ !contains(github.event.pull_request.title, 'WIP') }}
+    timeout-minutes: 30
+    strategy:
+      fail-fast: false
+      matrix:
+        jdk: [11]
+    steps:
+      - name: Set up Java
+        uses: actions/setup-java@v4
+        with:
+          java-version: ${{ matrix.jdk }}
+          distribution: 'temurin'
+      - name: Checkout Arrow
+        uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+          submodules: recursive
+      - name: Build
+        shell: bash
+        env:
+          DEVELOCITY_ACCESS_KEY: ${{ secrets.GE_ACCESS_TOKEN }}
+        run: ci/scripts/java_build.sh $(pwd) $(pwd)/build
+      - name: Test
+        shell: bash
+        env:
+          DEVELOCITY_ACCESS_KEY: ${{ secrets.GE_ACCESS_TOKEN }}
+        run: ci/scripts/java_test.sh $(pwd) $(pwd)/build

.gitignore

@@ -26,4 +26,3 @@ cmake_install.cmake
 dependency-reduced-pom.xml
 install_manifest.txt
 target/
-/.mvn/.develocity/

arrow-format/FlightSql.proto

@@ -1857,7 +1857,7 @@ message DoPutPreparedStatementResult {
   // statement should be considered invalid, and all subsequent requests for this prepared
   // statement must use this new handle.
   // The updated handle allows implementing query parameters with stateless services.
-  // 
+  //
   // When an updated handle is not provided by the server, clients should contiue
   // using the previous handle provided by `ActionCreatePreparedStatementResonse`.
   optional bytes prepared_statement_handle = 1;

arrow-format/Schema.fbs

@@ -216,14 +216,14 @@ table Bool {
 
 /// Contains two child arrays, run_ends and values.
 /// The run_ends child array must be a 16/32/64-bit integer array
-/// which encodes the indices at which the run with the value in 
+/// which encodes the indices at which the run with the value in
 /// each corresponding index in the values child array ends.
 /// Like list/struct types, the value array can be of any type.
 table RunEndEncoded {
 }
 
 /// Exact decimal value represented as an integer value in two's
-/// complement. Currently 32-bit (4-byte), 64-bit (8-byte), 
+/// complement. Currently 32-bit (4-byte), 64-bit (8-byte),
 /// 128-bit (16-byte) and 256-bit (32-byte) integers are used.
 /// The representation uses the endianness indicated in the Schema.
 table Decimal {

arrow-format/substrait/extension_types.yaml

@@ -48,7 +48,7 @@
 #  *  Functions have the same meaning when applied to the encoded type
 #
 # Note: if two types have a different range (e.g. string and large_string) then
-# they do not satisfy the above criteria and are not encodings. 
+# they do not satisfy the above criteria and are not encodings.
 #
 # These types will never have a Substrait equivalent.  In the Substrait point
 # of view these are execution details.
@@ -167,4 +167,3 @@ types:
     parameters:
       - name: unit
         type: string
-

ci/scripts/java_build.sh

@@ -44,9 +44,18 @@ mkdir -p "${build_dir}/arrow-format"
 cp -r "${source_dir}/arrow-format" "${build_dir}"
 cp -r "${source_dir}/dev" "${build_dir}"
 
-for source_root in $(find "${source_dir}" -not \( -path "${source_dir}"/build -prune \) -type f -name pom.xml -exec realpath -s --relative-to="${source_dir}" '{}' \; |
-                         awk -F/ '{print $1}' |
-                         sort -u); do
+# Instead of hardcoding the list of directories to copy, find pom.xml and then
+# crawl back up to the top.  GNU realpath has --relative-to but this does not
+# work on macOS
+
+poms=$(find "${source_dir}" -not \( -path "${source_dir}"/build -prune \) -type f -name pom.xml)
+if [[ "$OSTYPE" == "darwin"* ]]; then
+    poms=$(echo "$poms" | xargs -n1 python -c "import sys; import os.path; print(os.path.relpath(sys.argv[1], '${source_dir}'))")
+else
+    poms=$(echo "$poms" | xargs -n1 realpath -s --relative-to="${source_dir}")
+fi
+
+for source_root in $(echo "${poms}" | awk -F/ '{print $1}' | sort -u); do
     cp -r "${source_dir}/${source_root}" "${build_dir}"
 done
 

dev/release/rat_exclude_files.txt

@@ -15,4 +15,5 @@
 # specific language governing permissions and limitations
 # under the License.
 
+.gitmodules
 dataset/src/test/resources/data/student.csv