deploy.yml

name: Deploy

on:
  workflow_call:
    inputs:
      environment:
        description: "Deployment environment"
        required: true
        type: string
        default: "staging"
      version:
        description: "Version to deploy (tag or commit)"
        required: false
        type: string
      skip-health-check:
        description: "Skip post-deployment health check"
        required: false
        default: false
        type: boolean
  workflow_dispatch:
    inputs:
      environment:
        description: "Deployment environment"
        required: true
        type: string
        default: "staging"
      version:
        description: "Version to deploy (tag or commit)"
        required: false
        type: string
      skip-health-check:
        description: "Skip post-deployment health check"
        required: false
        default: false
        type: boolean

env:
  CARGO_TERM_COLOR: always
  RUST_BACKTRACE: 1
  CARGO_INCREMENTAL: 0
  CARGO_NET_RETRY: 10
  RUSTUP_MAX_RETRIES: 10
  REGISTRY: ghcr.io
  IMAGE_NAME: terraphim/terraphim-ai

jobs:
  # Validate deployment parameters
  validate:
    name: Validate Deployment
    runs-on: [self-hosted, Linux, X64]
    timeout-minutes: 3
    outputs:
      environment: ${{ steps.env.outputs.environment }}
      version: ${{ steps.version.outputs.version }}
      is-production: ${{ steps.env.outputs.is-production }}
      deployment-url: ${{ steps.env.outputs.deployment-url }}

    steps:
      - name: Checkout
        uses: actions/checkout@v6
        with:
          fetch-depth: 0

      - name: Validate environment
        id: env
        run: |
          ENVIRONMENT="${{ github.event.inputs.environment || inputs.environment }}"

          case "$ENVIRONMENT" in
            staging|dev|development)
              echo "environment=staging" >> $GITHUB_OUTPUT
              echo "is-production=false" >> $GITHUB_OUTPUT
              echo "deployment-url=https://staging.terraphim.ai" >> $GITHUB_OUTPUT
              ;;
            production|prod)
              echo "environment=production" >> $GITHUB_OUTPUT
              echo "is-production=true" >> $GITHUB_OUTPUT
              echo "deployment-url=https://app.terraphim.ai" >> $GITHUB_OUTPUT
              ;;
            *)
              echo "Invalid environment: $ENVIRONMENT"
              echo "Valid environments: staging, production"
              exit 1
              ;;
          esac

      - name: Resolve version
        id: version
        run: |
          VERSION="${{ github.event.inputs.version || inputs.version || github.sha }}"

          # If it's a commit, get the short hash
          if [[ "$VERSION" =~ ^[0-9a-f]{7,40}$ ]]; then
            VERSION=$(git rev-parse --short "$VERSION")
          fi

          echo "version=$VERSION" >> $GITHUB_OUTPUT
          echo "Deploying version: $VERSION to ${{ steps.env.outputs.environment }}"

  # Prepare deployment artifacts
  prepare:
    name: Prepare Artifacts
    runs-on: [self-hosted, Linux, X64]
    timeout-minutes: 15
    needs: validate

    steps:
      - name: Checkout
        uses: actions/checkout@v6
        with:
          ref: ${{ needs.validate.outputs.version }}

      - name: Download CI artifacts
        uses: actions/download-artifact@v4
        continue-on-error: true  # Artifacts may not exist for custom versions
        with:
          pattern: rust-binaries-*
          path: artifacts/binaries
          merge-multiple: true

      - name: Download Docker image
        if: needs.validate.outputs.version != github.sha
        run: |
          # Pull existing Docker image for the version
          docker pull ${{ env.REGISTRY }}/${{ env.IMAGE_NAME }}:${{ needs.validate.outputs.version }} || \
          docker pull ${{ env.REGISTRY }}/${{ env.IMAGE_NAME }}:latest || \
          echo "Docker image not found, will build from source"

      - name: Install Rust toolchain
        if: steps.download.outcome == 'failure'
        uses: dtolnay/rust-toolchain@stable
        with:
          components: rustfmt, clippy

      - name: Cache Cargo (self-hosted)
        if: steps.download.outcome == 'failure'
        uses: actions/cache@v4
        with:
          path: |
            /opt/cargo-cache/registry
            /opt/cargo-cache/git
            ~/.cargo/registry
            ~/.cargo/git
            target
          key: deploy-build-${{ needs.validate.outputs.version }}-${{ hashFiles('**/Cargo.lock') }}
        env:
          CARGO_HOME: /opt/cargo-cache

      - name: Build binaries
        if: steps.download.outcome == 'failure'
        run: |
          cargo build --release --package terraphim_server --package terraphim_mcp_server

          # Create binaries directory
          mkdir -p artifacts/binaries
          cp target/release/terraphim* artifacts/binaries/

      - name: Upload prepared artifacts
        uses: actions/upload-artifact@v4
        with:
          name: deployment-binaries
          path: artifacts/binaries/
          retention-days: 7

  # Deploy to staging
  deploy-staging:
    name: Deploy to Staging
    runs-on: [self-hosted, Linux, X64]
    timeout-minutes: 10
    needs: [validate, prepare]
    if: needs.validate.outputs.environment == 'staging'
    environment:
      name: staging
      url: ${{ needs.validate.outputs.deployment-url }}

    steps:
      - name: Download deployment artifacts
        uses: actions/download-artifact@v4
        with:
          name: deployment-binaries
          path: ./binaries

      - name: Setup SSH
        uses: webfactory/ssh-agent@v0.9.0
        with:
          ssh-private-key: ${{ secrets.STAGING_SSH_KEY }}

      - name: Deploy to staging server
        run: |
          # Configuration
          STAGING_HOST="${{ secrets.STAGING_HOST }}"
          STAGING_USER="${{ secrets.STAGING_USER }}"
          STAGING_PATH="${{ secrets.STAGING_PATH || '/opt/terraphim' }}"

          # Create deployment package
          tar -czf deployment.tar.gz -C binaries .

          # Copy to staging server
          scp -o StrictHostKeyChecking=no deployment.tar.gz $STAGING_USER@$STAGING_HOST:/tmp/

          # Deploy
          ssh -o StrictHostKeyChecking=no $STAGING_USER@$STAGING_HOST << 'EOF'
            set -e

            # Create backup
            if [[ -d "/opt/terraphim" ]]; then
              sudo cp -r /opt/terraphim /opt/terraphim.backup.$(date +%s)
            fi

            # Extract deployment
            cd /tmp
            tar -xzf deployment.tar.gz

            # Stop service
            sudo systemctl stop terraphim-server || true

            # Deploy files
            sudo mkdir -p /opt/terraphim/bin
            sudo cp -r /tmp/* /opt/terraphim/bin/
            sudo chown -R terraphim:terraphim /opt/terraphim/
            sudo chmod +x /opt/terraphim/bin/terraphim*

            # Start service
            sudo systemctl start terraphim-server
            sudo systemctl enable terraphim-server

            # Cleanup
            rm -f /tmp/deployment.tar.gz /tmp/terraphim*
          EOF

      - name: Health check
        if: github.event.inputs.skip-health-check != 'true' && inputs.skip-health-check != 'true'
        run: |
          echo "Performing health check..."

          # Wait for service to start
          for i in {1..30}; do
            if curl -f "${{ needs.validate.outputs.deployment-url }}/health" 2>/dev/null; then
              echo "✅ Health check passed"
              break
            fi
            echo "Waiting for service... ($i/30)"
            sleep 5
          done

          # Final health check
          curl -f "${{ needs.validate.outputs.deployment-url }}/health" || {
            echo "❌ Health check failed"
            exit 1
          }

  # Deploy to production
  deploy-production:
    name: Deploy to Production
    runs-on: [self-hosted, Linux, X64]
    timeout-minutes: 15
    needs: [validate, prepare]
    if: needs.validate.outputs.environment == 'production'
    environment:
      name: production
      url: ${{ needs.validate.outputs.deployment-url }}

    steps:
      - name: Download deployment artifacts
        uses: actions/download-artifact@v4
        with:
          name: deployment-binaries
          path: ./binaries

      - name: Deploy via Docker Compose (Production)
        run: |
          echo "Deploying to production using Docker Compose"

          # Create docker-compose override for production
          cat > docker-compose.override.yml << EOF
          version: '3.8'
          services:
            terraphim:
              image: ${{ env.REGISTRY }}/${{ env.IMAGE_NAME }}:${{ needs.validate.outputs.version }}
              restart: always
              environment:
                - NODE_ENV=production
                - LOG_LEVEL=info
              ports:
                - "80:8080"
              deploy:
                replicas: 2
                resources:
                  limits:
                    memory: 2G
                    cpus: '1.0'
          EOF

      - name: Production health check
        if: github.event.inputs.skip-health-check != 'true' && inputs.skip-health-check != 'true'
        run: |
          echo "Performing production health check..."

          # Extended health check for production
          for i in {1..60}; do
            if curl -f "${{ needs.validate.outputs.deployment-url }}/health" 2>/dev/null; then
              echo "✅ Production health check passed"
              break
            fi
            echo "Waiting for production service... ($i/60)"
            sleep 5
          done

          # Additional production checks
          curl -f "${{ needs.validate.outputs.deployment-url }}/health" && \
          curl -f "${{ needs.validate.outputs.deployment-url }}/config" || {
            echo "❌ Production health check failed"
            exit 1
          }

  # Deploy Docker image
  deploy-docker:
    name: Deploy Docker Image
    runs-on: [self-hosted, Linux, X64]
    timeout-minutes: 10
    needs: [validate]
    if: needs.validate.outputs.environment == 'production'

    steps:
      - name: Set up Docker Buildx
        uses: docker/setup-buildx-action@v3

      - name: Log in to Container Registry
        uses: docker/login-action@v3
        with:
          registry: ${{ env.REGISTRY }}
          username: ${{ github.actor }}
          password: ${{ secrets.GITHUB_TOKEN }}

      - name: Tag and push production image
        run: |
          # Pull source image
          docker pull ${{ env.REGISTRY }}/${{ env.IMAGE_NAME }}:${{ needs.validate.outputs.version }}

          # Tag for production
          docker tag ${{ env.REGISTRY }}/${{ env.IMAGE_NAME }}:${{ needs.validate.outputs.version }} \
                     ${{ env.REGISTRY }}/${{ env.IMAGE_NAME }}:production

          # Push production tag
          docker push ${{ env.REGISTRY }}/${{ env.IMAGE_NAME }}:production

  # Post-deployment notifications
  notify:
    name: Deployment Notifications
    runs-on: [self-hosted, Linux, X64]
    timeout-minutes: 5
    needs: [validate, deploy-staging, deploy-production]
    if: always()

    steps:
      - name: Notify on success
        if: needs.deploy-staging.result == 'success' || needs.deploy-production.result == 'success'
        run: |
          echo "🚀 Deployment successful!"
          echo "Environment: ${{ needs.validate.outputs.environment }}"
          echo "Version: ${{ needs.validate.outputs.version }}"
          echo "URL: ${{ needs.validate.outputs.deployment-url }}"

      - name: Notify on failure
        if: needs.deploy-staging.result == 'failure' || needs.deploy-production.result == 'failure'
        run: |
          echo "❌ Deployment failed!"
          echo "Environment: ${{ needs.validate.outputs.environment }}"
          echo "Version: ${{ needs.validate.outputs.version }}"
          exit 1

      - name: Update deployment status
        if: github.event_name == 'workflow_call'
        run: |
          # Update any external deployment tracking systems
          echo "Updating deployment status for ${{ needs.validate.outputs.environment }}"

  # Rollback on failure
  rollback:
    name: Rollback on Failure
    runs-on: [self-hosted, Linux, X64]
    timeout-minutes: 8
    needs: [validate, deploy-staging, deploy-production]
    if: always() && (needs.deploy-staging.result == 'failure' || needs.deploy-production.result == 'failure') && needs.validate.outputs.is-production == 'true'

    steps:
      - name: Rollback deployment
        run: |
          echo "🔄 Rolling back deployment..."
          echo "This would typically involve:"
          echo "- Restoring previous Docker image"
          echo "- Reverting database migrations if needed"
          echo "- Restoring configuration files"
          echo "- Restarting services"

          # Placeholder for actual rollback logic
          echo "Rollback completed"