#!/bin/bash
# 等待CDN解封后批量修复正文
MAX_ATTEMPTS=24
ATTEMPT=1

while [ $ATTEMPT -le $MAX_ATTEMPTS ]; do
    echo "[$ATTEMPT/$MAX_ATTEMPTS] $(date) 检查 luzhou.gov.cn..."
    RESULT=$(timeout 20 curl -sL --max-time 15 -A "Mozilla/5.0" "https://www.luzhou.gov.cn/xw/gsgg" 2>/dev/null | head -1)

    if [ -z "$RESULT" ] || echo "$RESULT" | grep -q "1.94.217.116"; then
        echo "  被封禁, 等待5分钟..."
        sleep 300
        ATTEMPT=$((ATTEMPT + 1))
    else
        echo "  站点已恢复!"
        echo ""
        echo "=== 1/2 泸州: 清理旧数据 ==="
        echo 'DELETE FROM gov_raw WHERE site_name="泸州市人民政府-公示公告"' | sqlite3 /root/search.db
        echo 'DELETE FROM gov_search WHERE rowid IN (SELECT id FROM gov_raw WHERE site_name="泸州市人民政府-公示公告")' | sqlite3 /root/search.db 2>/dev/null

        echo "=== 1/2 泸州: 全量爬取(5页) ==="
        cd /root/gov_crawler
        python3 crawl_luzhou.py 5

        echo ""
        echo "=== 2/2 泸县: 强制重爬已有记录(修复表格) ==="
        cd /root/gov_crawler
        python3 crawl_luxian.py refetch

        echo ""
        echo "=== 重启 search_app ==="
        systemctl restart search_app
        echo ""
        echo "=== 全部完成! $(date) ==="
        exit 0
    fi
done
echo "重试达上限, 放弃"
exit 1