xrpl-dev-portal/.github/workflows/selenium-link-check.yml

name: Selenium link check

on:
  ## Currently too slow to run on PRs — eventually, we should have it run
  ## on just the changed files in a PR.
  # pull_request:
  #   branches: [master]
  schedule:
    - cron: '0 8 * * 0'    # Sundays 00:00 PST / 01:00 PDT
  workflow_dispatch:

concurrency:
  group: selenium-link-check-${{ github.ref }}
  cancel-in-progress: true

jobs:
  link-check:
    runs-on: ubuntu-latest
    timeout-minutes: 360
    permissions:
      contents: read
      pull-requests: write
    env:
      PORT: 4000
      BASE_URL: http://localhost:4000

    steps:
      - uses: actions/checkout@v6

      - uses: actions/setup-node@v6
        with:
          node-version: lts/*
          cache: npm

      - uses: actions/setup-python@v6
        with:
          python-version: '3.12'

      - run: npm ci

      - name: Build CSS
        run: npm run build-css

      - name: Install Python dependencies
        run: pip install selenium beautifulsoup4 markdown requests

      - name: Restore link cache
        uses: actions/cache/restore@v5
        with:
          path: tools/link-cache.json
          key: selenium-link-cache-${{ github.ref_name }}-${{ github.run_id }}
          restore-keys: |
            selenium-link-cache-${{ github.ref_name }}-
            selenium-link-cache-master-

      - name: Start Realm dev server
        run: |
          npx realm develop --port "$PORT" > realm.log 2>&1 &
          echo $! > realm.pid

      - name: Wait for server
        run: npx --yes wait-on "$BASE_URL" --timeout 180000

      - name: Run Selenium link checker
        run: |
          set -eo pipefail
          python tools/check-links.py -t 2 docs/ 2>&1 | tee selenium-link-check.log

      - name: Stop Realm server
        if: always()
        run: |
          if [ -f realm.pid ]; then kill "$(cat realm.pid)" 2>/dev/null || true; fi

      - name: Strip ANSI color codes from realm.log
        if: always()
        run: |
          if [ -f realm.log ]; then
            sed -i 's/\x1b\[[0-9;]*m//g' realm.log
          fi

      - name: Upload report
        id: upload
        if: always()
        uses: actions/upload-artifact@v7
        with:
          name: selenium-link-check-report
          retention-days: 14
          path: |
            selenium-link-check.log
            realm.log
            tools/broken-links-report.md
            tools/link-cache.json
          if-no-files-found: warn

      - name: Post link-check status to PR
        if: always() && github.event_name == 'pull_request'
        uses: actions/github-script@v9
        with:
          script: |
            const fs = require('fs');
            const marker = '<!-- selenium-link-check-report -->';
            const reportFile = 'tools/broken-links-report.md';
            const artifactUrl = '${{ steps.upload.outputs.artifact-url }}';

            const { data: comments } = await github.rest.issues.listComments({
              owner: context.repo.owner,
              repo: context.repo.repo,
              issue_number: context.issue.number,
            });
            const existing = comments.find(c => c.body?.startsWith(marker));

            let body;
            if (fs.existsSync(reportFile)) {
              const report = fs.readFileSync(reportFile, 'utf-8');
              body = [marker, report, '', `_[Download report zip](${artifactUrl})._`].join('\n');
            } else {
              // If no report file, determine whether the script actually completed
              // by checking the log for the final summary line (no crash or cancellation).
              let scriptCompleted = false;
              try {
                const log = fs.readFileSync('selenium-link-check.log', 'utf-8');
                scriptCompleted = /^\d+ broken links found among \d+ total links\.$/m.test(log);
              } catch (e) {
                // Log file is missing entirely
              }

              if (!scriptCompleted) {
                core.info('Script did not complete (cancelled or crashed) — preserving existing comment.');
                return;
              }

              if (existing) {
                body = [
                  marker,
                  '# Broken links report',
                  '',
                  '✅ Broken links fixed',
                  '',
                  `_[Download report zip](${artifactUrl})._`,
                ].join('\n');
              } else {
                core.info('No broken links and no prior sticky comment — skipping.');
                return;
              }
            }

            if (existing) {
              await github.rest.issues.updateComment({
                owner: context.repo.owner,
                repo: context.repo.repo,
                comment_id: existing.id,
                body,
              });
            } else {
              await github.rest.issues.createComment({
                owner: context.repo.owner,
                repo: context.repo.repo,
                issue_number: context.issue.number,
                body,
              });
            }

      - name: Save link cache
        if: always()
        uses: actions/cache/save@v5
        with:
          path: tools/link-cache.json
          key: selenium-link-cache-${{ github.ref_name }}-${{ github.run_id }}

      - name: Fail if broken links were found
        if: always()
        run: |
          if [ -f tools/broken-links-report.md ]; then
            echo "::error::Broken links detected — see PR comment or tools/broken-links-report.md artifact"
            cat tools/broken-links-report.md
            exit 1
          fi