osc/research/pdf_downloader/set_raw_title_kcna.py

import pymysql
from typing import Dict, List, Tuple, Optional

# ================== 配置区 ==================

DB_CONFIG = {
    'host': '47.113.231.200',
    'port': 28089,
    'user': 'root',
    'password': 'passok123A',
    'database': 'dsp',
    'charset': 'utf8mb4',
}

# 仅用于指定哪些 es_srcname 的记录需要处理（值可为空，因为不再做替换）
TARGET_SRCNAMES: List[str] = [
    "http://www.kcna.kp/cn/category/articles/q/5394b80bdae203fadef02522cfb578c0.kcmsf",
    # 添加你需要处理的站点名
]


# ================== 工具函数 ==================

def get_suffix_32(url: str) -> Optional[str]:
    """获取 URL 最后 32 个字符，不足则返回 None"""
    if not url or len(url) < 32:
        return None
    return url[-32:]


def find_foreign_by_suffix(cursor, suffix: str, exclude_id: int) -> Optional[Tuple[str, str]]:
    """
    根据后缀查找外文记录（排除自身）
    """
    query = """
        SELECT es_urltitle, es_urlcontent 
        FROM indeximos 
        WHERE 
            es_sid != %s 
            AND es_urlname IS NOT NULL 
            AND CHAR_LENGTH(es_urlname) >= 32 
            AND RIGHT(es_urlname, 32) = %s
        LIMIT 1
    """
    cursor.execute(query, (exclude_id, suffix))
    result = cursor.fetchone()
    return result if result else None


def update_chinese_record(cursor, record_id: int, title: str, content: str):
    """更新中文记录的 es_title 和 es_content"""
    update_query = """
        UPDATE indeximos 
        SET es_title = %s, es_content = %s 
        WHERE es_sid = %s
    """
    cursor.execute(update_query, (title, content, record_id))


# ================== 主逻辑 ==================

def main():
    if not TARGET_SRCNAMES:
        print("⚠️ 未指定任何目标 es_srcname，程序退出。")
        return

    conn = pymysql.connect(**DB_CONFIG)
    cursor = conn.cursor()

    try:
        # 获取所有目标站点的中文记录
        placeholders = ','.join(['%s'] * len(TARGET_SRCNAMES))
        query = f"""
            SELECT es_sid, es_srcname, es_urlname 
            FROM indeximos 
            WHERE es_srcname IN ({placeholders})
              AND es_urlname IS NOT NULL 
              AND es_urlname != '' 
              AND es_loadtime > '2026-01-16 10:40:00'
        """
        cursor.execute(query, TARGET_SRCNAMES)
        records = cursor.fetchall()
        total = len(records)
        print(f"共加载 {total} 条来自 {TARGET_SRCNAMES} 的记录用于匹配...")

        updated_count = 0
        skipped_short = 0

        for idx, (record_id, es_srcname, es_urlname) in enumerate(records, 1):
            suffix = get_suffix_32(es_urlname)
            if suffix is None:
                skipped_short += 1
                continue

            foreign_data = find_foreign_by_suffix(cursor, suffix, record_id)
            if foreign_data:
                title, content = foreign_data
                update_chinese_record(cursor, record_id, title, content)
                updated_count += 1
                print(f"[{idx}/{total}] ✅ 已更新 ID={record_id} | src={es_srcname}")

        conn.commit()
        print("\n" + "=" * 50)
        print(f"✅ 匹配完成！")
        print(f"  - 成功更新: {updated_count} 条")
        print(f"  - 因 URL 长度 <32 跳过: {skipped_short} 条")
        print(f"  - 总处理: {total} 条")

    except Exception as e:
        conn.rollback()
        print(f"❌ 发生错误，已回滚: {e}")
        raise
    finally:
        cursor.close()
        conn.close()


if __name__ == "__main__":
    main()