sitemap-parity.yml

name: Sitemap Parity

# Ensures every URL currently listed in the production sitemap is still
# reachable on the deployment that will be produced by this branch. Runs on:
#   - pull_request  -> waits for the Netlify deploy-preview, then checks
#   - push to main  -> waits for the Netlify production deploy, then checks
#
# In both cases the job blocks until Netlify posts a successful commit status
# for the exact commit under test (Netlify integrates via the GitHub Commit
# Status API, not the Deployments API), then runs the check against the URL
# from that status. If any prod URL is unreachable on the new deploy, the
# check fails and blocks the merge.

on:
  pull_request:
    branches: ['main']
    
permissions:
  contents: read

concurrency:
  group: sitemap-parity-${{ github.event.pull_request.head.sha || github.sha }}
  cancel-in-progress: true

jobs:
  check:
    runs-on: ubuntu-latest
    timeout-minutes: 20

    steps:
      - name: Checkout repository
        uses: actions/checkout@v4
        with:
          # For PRs, check out the PR head commit so the script matches what
          # Netlify actually deployed. For pushes, this is the pushed commit.
          ref: ${{ github.event.pull_request.head.sha || github.sha }}

      - name: Setup Node.js
        uses: actions/setup-node@v4
        with:
          node-version: '20'

      - name: Wait for Netlify deployment
        id: wait
        env:
          GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
          SHA: ${{ github.event.pull_request.head.sha || github.sha }}
          # Total wait budget in seconds (default 15 min) and poll interval.
          WAIT_TIMEOUT: '900'
          WAIT_INTERVAL: '15'
        run: |
          set -euo pipefail
          echo "Waiting for a Netlify commit status on commit $SHA ..."

          # Netlify reports build status via the GitHub Commit Status API
          # (context typically `netlify/<site-name>/deploy-preview`) rather
          # than the Deployments API, so we poll that endpoint.
          deadline=$(( $(date +%s) + WAIT_TIMEOUT ))
          preview_url=""

          while [ "$(date +%s)" -lt "$deadline" ]; do
            status_json=$(gh api "repos/${{ github.repository }}/commits/$SHA/status")

            # Grab the first status whose context looks like a Netlify one.
            # `map(select(...)) | .[0]` normalises the jq stream into a single
            # (possibly null) object so downstream extraction is unambiguous.
            netlify_state=$(printf '%s' "$status_json" | jq -r '
              .statuses
              | map(select(.context | ascii_downcase | startswith("netlify/")))
              | .[0].state // ""')
            netlify_target=$(printf '%s' "$status_json" | jq -r '
              .statuses
              | map(select(.context | ascii_downcase | startswith("netlify/")))
              | .[0].target_url // ""')

            if [ "$netlify_state" = "success" ]; then
              if [ -z "$netlify_target" ] || [ "$netlify_target" = "null" ]; then
                echo "::error::Netlify status is success but has no target_url."
                exit 1
              fi
              preview_url="$netlify_target"
              echo "Found successful Netlify deployment -> $preview_url"
              break
            elif [ "$netlify_state" = "failure" ] || [ "$netlify_state" = "error" ]; then
              echo "::error::Netlify build reported state=$netlify_state for commit $SHA."
              exit 1
            elif [ -n "$netlify_state" ]; then
              echo "Netlify status is currently: $netlify_state. Waiting ${WAIT_INTERVAL}s ..."
            else
              echo "No Netlify status posted yet. Waiting ${WAIT_INTERVAL}s ..."
            fi

            sleep "$WAIT_INTERVAL"
          done

          if [ -z "$preview_url" ]; then
            echo "::error::Timed out after ${WAIT_TIMEOUT}s waiting for a Netlify deployment on $SHA."
            exit 1
          fi

          echo "preview_url=$preview_url" >> "$GITHUB_OUTPUT"

      - name: Check sitemap parity
        env:
          PREVIEW_URL: ${{ steps.wait.outputs.preview_url }}
        run: |
          echo "Preview URL: $PREVIEW_URL"
          node scripts/check-sitemap-parity.mjs \
            --prod-base=https://open-elements.com \
            --preview-base="$PREVIEW_URL" \
            --concurrency=4 \
            --retries=4