backfill_artists.yml

name: Backfill Artists

on:
  schedule:
    - cron: "0 1 1 * *" # 매달 1일 KST 오전 10:00 실행 (UTC+9 → UTC 01:00) — 최근 35일치만 증분 반영
  workflow_dispatch:
    inputs:
      since_days:
        description: "최근 N일 이내 등록된 곡만 대상으로 함 (비워두면 전체 곡 대상 백필)"
        required: false

jobs:
  run-npm-task:
    runs-on: ubuntu-latest

    steps:
      - uses: actions/checkout@v4

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

      - name: Install pnpm
        uses: pnpm/action-setup@v2
        with:
          version: 9
          run_install: false

      - name: Install dependencies
        working-directory: packages/crawling
        run: pnpm install

      - name: Create .env file
        working-directory: packages/crawling
        run: |
          echo "SUPABASE_URL=${{ secrets.SUPABASE_URL }}" >> .env
          echo "SUPABASE_KEY=${{ secrets.SUPABASE_KEY }}" >> .env

      # 스케줄 실행은 지난 한 달치만 증분 반영, 수동 실행은 입력값을 그대로 쓴다(비워두면 전체 백필).
      - name: Determine backfill window
        id: window
        run: |
          if [ "${{ github.event_name }}" = "schedule" ]; then
            echo "since_days=35" >> "$GITHUB_OUTPUT"
          else
            echo "since_days=${{ inputs.since_days }}" >> "$GITHUB_OUTPUT"
          fi

      - name: run backfill script - backfillArtists.ts
        working-directory: packages/crawling
        env:
          ARTIST_BACKFILL_SINCE_DAYS: ${{ steps.window.outputs.since_days }}
        run: pnpm run backfill-artists

      # 로그 파일은 러너 파일시스템에만 남고 잡이 끝나면 사라지므로, 실패해도(if: always()) 확인할 수 있게 올려둔다.
      - name: Upload backfill log
        if: always()
        uses: actions/upload-artifact@v4
        with:
          name: artist-backfill-log-${{ github.run_id }}
          path: packages/crawling/src/assets/artistBackfillLog.txt
          if-no-files-found: ignore
          retention-days: 30