| 123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168 |
- # -*- coding: utf-8 -*-
- # Author : Charley
- # Python : 3.12.10
- # Date : 2026/07/10
- """
- REA (collectrea.com) 增量爬虫(日调度)
- 逻辑(两阶段,与 wheatland 一致):
- 1. GET /search 只解析「最近」板块(RECENT AUCTIONS tab,约 16 场)
- 2. 查库 select distinct auction_key from rea_record,得到已爬过的场次
- 3. 差集 = 新增场次
- 4. 没有新增 → 本轮无数据可抓,结束
- 5. 阶段一 列表:对每个新增场次翻页抓 lot 列表入库(state 默认 0)
- 6. 阶段二 详情:扫库 state != 1 的记录 → 逐条进详情页抓「标题 + 5 字段 + 多图」写回
- 目标网站: https://collectrea.com/search
- 说明:REA 的场次一旦结束即定型,新场次会出现在「最近」板块顶部;因此增量只需盯「最近」,
- 用 auction_key 差集识别新增场次即可(老场次早已在库)。
- """
- import sys
- import time
- import random
- import schedule
- from curl_cffi import requests
- from loguru import logger
- from tenacity import retry, stop_after_attempt, wait_fixed
- from mysql_pool import MySQLConnectionPool
- from rea_core import (
- client_identifier_list,
- crawl_one_auction,
- get_auction_list,
- update_details_for_pending,
- after_log,
- )
- logger.remove()
- logger.add("./logs/{time:YYYYMMDD}.log", encoding="utf-8", rotation="00:00",
- format="[{time:YYYY-MM-DD HH:mm:ss.SSS}] {level} {message}",
- level="DEBUG", retention="7 day")
- def get_existing_auction_keys(log, sql_pool):
- """查库返回已爬过的 auction_key 集合。
- Args:
- log: logger 对象。
- sql_pool: MySQL 连接池;为 None 时返回空集合(视作库内无任何场次)。
- Returns:
- set[str]: 已存在的 auction_key 字符串集合。
- """
- if sql_pool is None:
- log.warning("sql_pool 为 None,视为库内无任何场次(将全量重抓最近板块)")
- return set()
- rows = sql_pool.select_all(
- "select distinct auction_key from rea_record"
- )
- keys = {str(r[0]) for r in rows} if rows else set()
- log.info(f"库中已存在 {len(keys)} 个 auction_key")
- return keys
- def diff_new_auctions(log, recent_auctions, existing_keys):
- """从「最近」场次中筛出库里没有的新增场次。
- Args:
- log: logger 对象。
- recent_auctions (list[dict]): get_auction_list(only_recent=True) 的返回。
- existing_keys (set[str]): 已存在的 auction_key 集合。
- Returns:
- list[dict]: 待抓取的新增场次列表。
- """
- new_list = [a for a in recent_auctions if a["auction_key"] not in existing_keys]
- log.info(f"新增待抓取场次数: {len(new_list)} -> {[a['auction_key'] for a in new_list]}")
- return new_list
- def run_incremental(log, sql_pool):
- """增量抓取主流程。
- Args:
- log: logger 对象。
- sql_pool: MySQL 连接池;为 None 时不入库,仅在内存收集并 print 样本。
- """
- impersonate = random.choice(client_identifier_list)
- with requests.Session() as session:
- try:
- recent = get_auction_list(log, session, impersonate, only_recent=True)
- except Exception as e:
- log.error(f"获取最近场次列表失败: {e}")
- return
- existing_keys = get_existing_auction_keys(log, sql_pool)
- new_auctions = diff_new_auctions(log, recent, existing_keys)
- if not new_auctions:
- log.info("本轮无新增场次,跳过 list 抓取")
- return
- collected = []
- for idx, auc in enumerate(new_auctions, 1):
- log.info(f"========== [{idx}/{len(new_auctions)}] 开始抓场次 {auc['auction_key']} ({auc['auction_name']}) ==========")
- try:
- lots = crawl_one_auction(log, sql_pool, session, impersonate, auc)
- if sql_pool is None:
- collected.extend(lots)
- except Exception as e:
- log.error(f"场次 {auc['auction_key']} 抓取异常: {e}")
- continue
- if sql_pool is None:
- log.info(f"增量抓取结束,共 {len(collected)} 条 lot(未入库)")
- for row in collected[:3]:
- print(row)
- @retry(stop=stop_after_attempt(100), wait=wait_fixed(3600), after=after_log)
- def rea_main(log):
- """日调度主函数:增量 list + 补详情。
- Args:
- log: logger 对象。
- """
- log.info(f"开始运行 {sys._getframe().f_code.co_name} 增量爬虫任务 ...")
- sql_pool = MySQLConnectionPool(log=log)
- if not sql_pool:
- log.error("MySQL数据库连接失败")
- raise Exception("MySQL数据库连接失败")
- try:
- # 阶段一:抓新增场次的列表入库
- try:
- run_incremental(log, sql_pool)
- except Exception as e:
- log.error(f"增量抓取失败: {e}")
- # 阶段二:扫库 state != 1 的记录补抓详情
- if sql_pool is not None:
- try:
- update_details_for_pending(log, sql_pool)
- except Exception as e:
- log.error(f"详情补抓失败: {e}")
- except Exception as e:
- log.error(f"{sys._getframe().f_code.co_name} error: {e}")
- finally:
- log.info(f"爬虫程序 {sys._getframe().f_code.co_name} 运行结束,等待下一轮采集 ...")
- def schedule_task():
- """启动调度器:先即时跑一次,之后每天 05:00 跑一次增量。"""
- # rea_main(log=logger)
- schedule.every().day.at("05:00").do(rea_main, log=logger)
- while True:
- schedule.run_pending()
- time.sleep(1)
- if __name__ == "__main__":
- # 测试时直接跑一次
- # rea_main(log=logger)
- # 上生产再切回 schedule
- schedule_task()
|