# Phase 10 — Automated data freshness monitoring
# REC-019 + REC-030 from Genspark review
#
# This workflow runs two Phase 10 Python scripts on a schedule:
#   1. monitor_law_updates.py — fetches Ratchakitcha RSS, inserts new
#      law/regulation notifications into law_update_notifications table
#   2. scrape_judgments.py — scrapes Supreme Court judgments for years
#      after the latest in DB, inserts into case_judgments table
#
# Both scripts are idempotent and write to Turso production DB.
# Failed runs enqueue retry jobs in ingestion_queue table.
#
# Schedule:
#   - RSS monitor: every weekday 09:00 ICT (02:00 UTC, Mon-Fri)
#   - Judgment scraper: every Saturday 02:00 ICT (19:00 UTC Friday)
#     (weekly to avoid overloading Supreme Court servers)
#
# Required GitHub secrets:
#   - TURSO_DATABASE_URL: libsql://panya-ai-siriwat08.aws-ap-northeast-1.turso.io
#   - TURSO_AUTH_TOKEN: eyJ... (long-lived token from Turso dashboard)
#
# Manual trigger: Actions tab → "Phase 10 Data Freshness" → Run workflow

name: Phase 10 Data Freshness

on:
  schedule:
    # RSS monitor: every weekday 09:00 ICT (02:00 UTC, Mon-Fri)
    - cron: '0 2 * * 1-5'
    # Judgment scraper: every Saturday 02:00 ICT (19:00 UTC Friday)
    - cron: '0 19 * * 5'
  workflow_dispatch:
    # Allow manual trigger from GitHub Actions UI
    inputs:
      script:
        description: 'Which script to run (both|rss|scraper)'
        required: false
        default: 'both'
        type: choice
        options:
          - both
          - rss
          - scraper

permissions:
  contents: read

jobs:
  rss-monitor:
    name: RSS Law Update Monitor (REC-030)
    runs-on: ubuntu-latest
    if: github.event_name == 'schedule' || github.event.inputs.script == 'both' || github.event.inputs.script == 'rss'
    steps:
      - name: Checkout
        uses: actions/checkout@v4

      - name: Setup Python
        uses: actions/setup-python@v5
        with:
          python-version: '3.12'

      - name: Run RSS monitor
        run: python3 scripts/monitor_law_updates.py
        env:
          TURSO_URL: ${{ secrets.TURSO_DATABASE_URL }}
          TURSO_TOKEN: ${{ secrets.TURSO_AUTH_TOKEN }}

      - name: Summary
        if: always()
        run: |
          echo "## RSS Monitor Summary" >> $GITHUB_STEP_SUMMARY
          echo "✓ Checked Ratchakitcha RSS feed for new laws/regulations" >> $GITHUB_STEP_SUMMARY
          echo "✓ New items inserted into law_update_notifications table" >> $GITHUB_STEP_SUMMARY
          echo "✓ Failed fetches enqueued in ingestion_queue for retry" >> $GITHUB_STEP_SUMMARY

  judgment-scraper:
    name: Supreme Court Judgment Scraper (REC-019)
    runs-on: ubuntu-latest
    if: github.event_name == 'schedule' || github.event.inputs.script == 'both' || github.event.inputs.script == 'scraper'
    steps:
      - name: Checkout
        uses: actions/checkout@v4

      - name: Setup Python
        uses: actions/setup-python@v5
        with:
          python-version: '3.12'

      - name: Run judgment scraper
        run: python3 scripts/scrape_judgments.py
        env:
          TURSO_URL: ${{ secrets.TURSO_DATABASE_URL }}
          TURSO_TOKEN: ${{ secrets.TURSO_AUTH_TOKEN }}

      - name: Summary
        if: always()
        run: |
          echo "## Judgment Scraper Summary" >> $GITHUB_STEP_SUMMARY
          echo "✓ Detected latest judgment year in DB" >> $GITHUB_STEP_SUMMARY
          echo "✓ Scraped Supreme Court for years after latest" >> $GITHUB_STEP_SUMMARY
          echo "✓ New judgments inserted into case_judgments table" >> $GITHUB_STEP_SUMMARY
          echo "✓ Failed years enqueued in ingestion_queue for retry" >> $GITHUB_STEP_SUMMARY