build-org-statistics.yml

name: Build Organization Language Statistics

on:
  workflow_dispatch:
  schedule:
    - cron: '0 9 * * 1'

jobs:
  language-stats:
    runs-on: ubuntu-latest
    permissions:
      contents: write

    steps:
      - name: Checkout
        uses: actions/checkout@v6

      - name: Build and Update Stats
        uses: nick-fields/retry@v3
        env:
          GH_TOKEN: ${{ secrets.ORG_STATS_PAT }}
          ORG: ${{ github.repository_owner }}
        with:
          max_attempts: 3
          timeout_seconds: 30
          command: |
            gh repo list $ORG --limit 1000 --json name --jq '.[].name' | while read repo; do
              gh api repos/$ORG/$repo/languages 2>/dev/null || echo '{}'
              sleep 0.1
            done | jq -s '
              reduce .[] as $lang (
                {};
                reduce ($lang | to_entries[]) as $entry (
                  .;
                  .[$entry.key] += $entry.value
                )
              )
              | to_entries
              | sort_by(-.value)
              | (map(.value) | add) as $total
              | map({language: .key, bytes: .value, percentage: (.value / $total * 100 | (. * 10 | round) / 10)})
            ' > stats.json

            UPDATED_AT=$(date -u +"%Y-%m-%d")
            export UPDATED_AT
            python3 - <<'PYEOF'
            import json, os

            updated_at = os.environ['UPDATED_AT']

            with open('stats.json') as f:
                stats = json.load(f)

            def bar(pct):
              return f"![{pct}%](https://progress-bar.xyz/{int(pct)}/?width=200&progress_color=58A6FF)"

            rows = []
            for i, s in enumerate(stats, 1):
                rows.append(
                    f"| {i} | {s['language']} | {s['bytes']:,} | {bar(s['percentage'])} |"
                )

            table = (
                "| # | Language | Bytes | Share |\n"
                "|--:|----------|------:|:------|\n"
                + "\n".join(rows)
            )

            note = (
                f"\n\n_Generated automatically by "
                f"[`.github/workflows/build-org-statistics.yml`](.github/workflows/build-org-statistics.yml)"
                f" · Last updated: {updated_at}_"
            )

            with open('README.md') as f:
                content = f.read()

            section_header = "## Technology Stack"
            idx = content.find(section_header)
            if idx == -1:
                print("Section not found in README.md")
                exit(1)

            end_of_header = content.find('\n', idx) + 1
            rest = content[end_of_header:]

            next_section = -1
            for pattern in ['\n## ', '\n---']:
                pos = rest.find(pattern)
                if pos != -1 and (next_section == -1 or pos < next_section):
                    next_section = pos

            after_section = rest[next_section:] if next_section != -1 else ""
            new_content = content[:end_of_header] + "\n" + table + note + "\n" + after_section

            with open('README.md', 'w') as f:
                f.write(new_content)

            print("README.md updated successfully")
            PYEOF

            git config user.name "github-actions[bot]"
            git config user.email "github-actions[bot]@users.noreply.github.com"
            git add README.md
            git diff --staged --quiet || git commit -m "Update technology stack statistics"
            git push
          
        
      - name: Debug stats.json
        run: cat stats.json