Review Update Dependency Cache Workflow

You now explore the Update dependency cache workflow.

0001

Check out .github/workflows/update-cache.yml file.

name: Update dependency cache

on:
  push:
    branches: [main]

  workflow_dispatch:

concurrency:
  group: ${{ github.workflow }}-${{ github.ref }}
  cancel-in-progress: true

jobs:
  update-dependency-cache:
    name: Update dependency cache
    runs-on: ubuntu-latest
    steps:
      - name: Checkout
        uses: actions/checkout@v4

      - name: Set up JDK ${{ vars.JAVA_VERSION }}
        uses: actions/setup-java@v4
        with:
          java-version: ${{ vars.JAVA_VERSION }}
          distribution: ${{ vars.JAVA_DISTRIBUTION }}

      - uses: actions/cache@v4
        name: Cache dependencies
        id: cache
        with:
          path: |
            ~/.gradle/caches
            ~/.gradle/wrapper            
          key: ${{ runner.os }}-gradle-${{ hashFiles('**/*.gradle*', '**/gradle-wrapper.properties') }}
          restore-keys: |
            ${{ runner.os }}-gradle-            
          lookup-only: true

      - if: ${{ steps.cache.outputs.cache-hit != 'true' }}
        name: Update cache
        run: |
          chmod +x gradlew
          ./gradlew dependencies          

This GitHub Actions workflow is designed to update the dependency cache for a project, specifically for Java Gradle-based projects. Let’s take a high-level look at key components of the workflow.

Events

push: triggers the workflow when the merge group for a given pull request is successful and the PR is actually merged into the main branch.

workflow_dispatch: allows the workflow to be manually started for debugging without inputs.

Concurrency

Ensures that only one instance of the workflow runs for a specific branch at a time, canceling any in-progress runs if a new one starts (explore more about concurrency group in 14. Experiments With GitHub Actions Concurrency Group).

Job

update-dependency-cache

  • Purpose: updates dependency cache in the main branch if needed.
  • Steps
    • Checkout the code.
    • Set up Java (JDK).
    • If a cache entry already exists, the process of restoring dependency is skipped.
    • This step executes only if the cache is not hit, after which the cache will be refreshed.