From 9de8c1bec0e0551871a81eb755ff700276c8542e Mon Sep 17 00:00:00 2001
From: Enrico Minack <github@enrico.minack.dev>
Date: Sat, 14 Dec 2024 22:13:14 +0100
Subject: [PATCH] Move download spark job into workflow

---
 .github/workflows/ci.yml             | 48 ++-----------------------
 .github/workflows/download-spark.yml | 54 ++++++++++++++++++++++++++++
 2 files changed, 57 insertions(+), 45 deletions(-)
 create mode 100644 .github/workflows/download-spark.yml

diff --git a/.github/workflows/ci.yml b/.github/workflows/ci.yml
index 070d094c..1a457860 100644
--- a/.github/workflows/ci.yml
+++ b/.github/workflows/ci.yml
@@ -54,51 +54,9 @@ jobs:
           git diff
         shell: bash
 
-  download:
+  download-spark:
     name: "Spark"
-    runs-on: ubuntu-latest
-    strategy:
-      fail-fast: false
-      matrix:
-        # use spark versions from test-integration.yaml workflow
-        include:
-          - spark-version: '3.0.3'
-            hadoop-version: '2.7'
-          - spark-version: '3.1.3'
-            hadoop-version: '2.7'
-          - spark-version: '3.2.4'
-            hadoop-version: '2.7'
-          - spark-version: '3.3.4'
-            hadoop-version: '3'
-          - spark-version: '3.4.3'
-            hadoop-version: '3'
-          - spark-version: '3.5.3'
-            hadoop-version: '3'
-          - spark-version: '4.0.0-preview2'
-            hadoop-version: '3'
-    steps:
-      - name: Cache Spark Binaries
-        uses: actions/cache@v4
-        with:
-          path: ~/spark
-          key: ${{ runner.os }}-spark-binaries-${{ inputs.spark-version }}-${{ inputs.scala-compat-version }}
-
-      - name: Setup Spark Binaries
-        env:
-          SPARK_PACKAGE: spark-${{ inputs.spark-version }}/spark-${{ inputs.spark-version }}-bin-hadoop${{ inputs.hadoop-version }}.tgz
-        run: |
-          if [[ ! -e ~/spark ]]
-          then
-            wget --progress=dot:giga "https://www.apache.org/dyn/closer.lua/spark/${SPARK_PACKAGE}?action=download" -O - | tar -xzC "${{ runner.temp }}"
-            archive=$(basename "${SPARK_PACKAGE}") bash -c "mv -v "${{ runner.temp }}/\${archive/%.tgz/}" ~/spark"
-          fi
-        shell: bash
-
-      - name: Upload Spark Binaries
-        uses: actions/upload-artifact@v4
-        with:
-          name: Spark-Binaries-${{ inputs.spark-version }}-${{ inputs.hadoop-version }}
-          path: ~/spark
+    uses: "./.github/workflows/download-spark.yml"
 
   build:
     name: "Build"
@@ -175,7 +133,7 @@ jobs:
 
   test-integration:
     name: "Test Integration"
-    needs: [test-dgraph, test-spark]
+    needs: [download-spark, test-dgraph, test-spark]
     uses: "./.github/workflows/test-integration.yml"
 
   delete_binaries:
diff --git a/.github/workflows/download-spark.yml b/.github/workflows/download-spark.yml
new file mode 100644
index 00000000..00150399
--- /dev/null
+++ b/.github/workflows/download-spark.yml
@@ -0,0 +1,54 @@
+name: Build
+
+on:
+  workflow_call:
+
+jobs:
+  build:
+    name: Download (Spark ${{ matrix.spark-version }} Hadoop ${{ matrix.hadoop-version }})
+    runs-on: ubuntu-latest
+
+    strategy:
+      fail-fast: false
+      matrix:
+        # use spark versions from test-integration.yaml workflow
+        include:
+          - spark-version: '3.0.3'
+            hadoop-version: '2.7'
+          - spark-version: '3.1.3'
+            hadoop-version: '2.7'
+          - spark-version: '3.2.4'
+            hadoop-version: '2.7'
+          - spark-version: '3.3.4'
+            hadoop-version: '3'
+          - spark-version: '3.4.3'
+            hadoop-version: '3'
+          - spark-version: '3.5.3'
+            hadoop-version: '3'
+          - spark-version: '4.0.0-preview2'
+            hadoop-version: '3'
+
+    steps:
+      - name: Cache Spark Binaries
+        uses: actions/cache@v4
+        with:
+          path: ~/spark
+          key: ${{ runner.os }}-spark-binaries-${{ matrix.spark-version }}-${{ matrix.scala-compat-version }}
+
+      - name: Setup Spark Binaries
+        env:
+          SPARK_PACKAGE: spark-${{ matrix.spark-version }}/spark-${{ matrix.spark-version }}-bin-hadoop${{ matrix.hadoop-version }}.tgz
+        run: |
+          if [[ ! -e ~/spark ]]
+          then
+            wget --progress=dot:giga "https://www.apache.org/dyn/closer.lua/spark/${SPARK_PACKAGE}?action=download" -O - | tar -xzC "${{ runner.temp }}"
+            archive=$(basename "${SPARK_PACKAGE}") bash -c "mv -v "${{ runner.temp }}/\${archive/%.tgz/}" ~/spark"
+          fi
+        shell: bash
+
+      - name: Upload Spark Binaries
+        uses: actions/upload-artifact@v4
+        with:
+          name: Spark-Binaries-${{ matrix.spark-version }}-${{ matrix.hadoop-version }}
+          path: ~/spark
+