update-candidates.yml

name: Update Candidates Pool

on:
  schedule:
    - cron: '0 19 * * 0'  # 每周日凌晨 3 点（UTC+8）
  workflow_dispatch:
    inputs:
      batch_size:
        description: '验证批次大小'
        required: false
        default: '200'
      concurrency:
        description: '验证并发数'
        required: false
        default: '20'
      timeout:
        description: '验证超时时间（秒）'
        required: false
        default: '10'

jobs:
  update-candidates:
    runs-on: ubuntu-latest
    timeout-minutes: 60

    steps:
      - name: Checkout repository
        uses: actions/checkout@v4
        with:
          fetch-depth: 0

      - name: Set up Python
        uses: actions/setup-python@v5
        with:
          python-version: '3.11'
          cache: 'pip'

      - name: Install dependencies
        run: |
          pip install -r scripts/requirements.txt

      - name: Update candidates pool
        id: update
        run: |
          set +e

          # 读取原始书源
          echo "读取原始书源..."
          RAW_COUNT=$(python3 -c "import json; print(len(json.load(open('sources/legado/pool/raw.json'))))" 2>&1)
          if [ $? -ne 0 ]; then
            echo "::error::统计原始书源失败: $RAW_COUNT"
            RAW_COUNT="0"
          fi
          echo "原始书源数量: $RAW_COUNT"

          # 快速过滤（使用独立脚本）
          echo "执行快速过滤..."
          python3 scripts/quick_filter.py \
            --input sources/legado/pool/raw.json \
            --output sources/legado/temp/filtered.json

          FILTER_EXIT_CODE=$?
          if [ $FILTER_EXIT_CODE -ne 0 ]; then
            echo "::error::快速过滤失败，退出码: $FILTER_EXIT_CODE"
            echo "status=failed" >> $GITHUB_OUTPUT
            exit $FILTER_EXIT_CODE
          fi

          # 分批验证
          echo "执行分批验证..."
          python3 scripts/batch_validator.py \
            --input sources/legado/temp/filtered.json \
            --output-valid sources/legado/pool/candidates.json \
            --output-invalid sources/legado/pool/invalid.json \
            --batch-size ${{ github.event.inputs.batch_size || '200' }} \
            --concurrency ${{ github.event.inputs.concurrency || '20' }} \
            --timeout ${{ github.event.inputs.timeout || '10' }} \
            --checkpoint-dir sources/legado/temp/checkpoints

          EXIT_CODE=$?
          set -e

          if [ $EXIT_CODE -eq 0 ]; then
            echo "status=success" >> $GITHUB_OUTPUT

            # 统计结果
            CANDIDATES_COUNT=$(python3 -c "import json; print(len(json.load(open('sources/legado/pool/candidates.json'))))" 2>&1)
            if [ $? -ne 0 ]; then
              echo "::warning::统计候选书源失败: $CANDIDATES_COUNT"
              CANDIDATES_COUNT="0"
            fi
            echo "candidates_count=$CANDIDATES_COUNT" >> $GITHUB_OUTPUT
          else
            echo "status=failed" >> $GITHUB_OUTPUT
            exit $EXIT_CODE
          fi

      - name: Commit changes
        if: steps.update.outputs.status == 'success'
        run: |
          git config --local user.email "github-actions[bot]@users.noreply.github.com"
          git config --local user.name "github-actions[bot]"

          git add sources/legado/pool/candidates.json
          git add sources/legado/pool/invalid.json

          git commit -m "chore: 更新候选池 - 当前 ${{ steps.update.outputs.candidates_count }} 个候选书源" || echo "No changes to commit"

      - name: Push changes
        if: steps.update.outputs.status == 'success'
        run: |
          set +e
          for i in 1 2 3; do
            git push && break || {
              echo "Push attempt $i failed, retrying in $((i * 5)) seconds..."
              sleep $((i * 5))
            }
          done
          set -e

      - name: Upload artifacts on failure
        if: failure()
        uses: actions/upload-artifact@v4
        with:
          name: update-candidates-logs
          path: |
            sources/legado/temp/checkpoints/
          retention-days: 7

      - name: Create issue on failure
        if: failure()
        uses: actions/github-script@v7
        with:
          script: |
            github.rest.issues.create({
              owner: context.repo.owner,
              repo: context.repo.repo,
              title: '候选池更新失败',
              body: `候选池更新在 ${new Date().toISOString()} 失败。\n\n请检查工作流日志：${context.payload.repository.html_url}/actions/runs/${context.runId}`,
              labels: ['bug', 'automation']
            })

      - name: Cleanup
        if: ${{ !cancelled() }}
        run: |
          rm -rf sources/legado/temp/filtered.json || true
          rm -rf sources/legado/temp/checkpoints/*.json || true