2.0/start_run_ybm.py

"""
药帮忙 (YBM) 平台采集入口
使用 commons1 调度框架 —— 从内部 API 拉取任务，由 YbmCrawler 执行采集。
"""
import random
import time
import re
from datetime import datetime

import requests
from commons.scheduler import CrawlerScheduler
from commons.Logger import get_spider_logger
from commons.feishu_webhook import send_text, send_error_card
from spiders.ybm.ybm_crawl import (
    YbmCrawler,
    PLATFORM_ID,
    YBM_DEVICE_ID,
    load_city_mapping,
)

logger = get_spider_logger("ybm_runner")

PLATFORM_NAME = "药帮忙"
IDLE_SECONDS_MIN = 180
IDLE_SECONDS_MAX = 300


def notify_result(task, crawl_count, success):
    """发送飞书通知"""
    try:
        drug_name = task.get("product_name", "")
        spec = task.get("product_specs", "") or ""
        spec_items = [item.strip() for item in re.split(r"[|、，,\n\r]+", spec) if item.strip()]
        display_spec = "、".join(spec_items)
        notice_time = datetime.now().strftime("%Y-%m-%d %H:%M:%S")
        text_msg = (
            f"{notice_time} 通知:\n"
            f"平台: {PLATFORM_NAME}, 药品: {drug_name}, 品规: {display_spec}, 爬取数据: {crawl_count}条"
        )
        if success:
            send_text(text_msg)
        else:
            send_error_card(
                task_name=drug_name,
                err_msg=f"task_id={task.get('id')}, company_id={task.get('company_id')}, crawl_count={crawl_count}",
                mention_all=False,
            )
    except Exception as e:
        logger.warning("飞书通知发送失败：%s", e)


def main_loop():
    """循环拉取任务并执行采集"""
    load_city_mapping()
    scheduler = CrawlerScheduler(YBM_DEVICE_ID, str(PLATFORM_ID))
    idle_seconds = random.randint(IDLE_SECONDS_MIN, IDLE_SECONDS_MAX)
    logger.info("药帮忙采集启动，每轮间隔 %s 秒", idle_seconds)

    while True:
        try:
            # 先同步上报心跳，确保服务器登记后再拉任务
            requests.post(
                scheduler.heartbeat_url,
                json={"platform": scheduler.platform, "username": scheduler.username},
                headers={'X-Crawler-Token': 'zhijiayun_crawler_2026'},
                timeout=5
            )
            scheduler.start()
            task = scheduler.get_task() or {}
            if not task:
                logger.info("%s 暂无任务，等待 %s 秒后重试", PLATFORM_NAME, idle_seconds)
                time.sleep(idle_seconds)
                continue

            logger.info("开始执行%s爬虫任务 task_id=%s", PLATFORM_NAME, task.get("id"))
            crawler = YbmCrawler(task)
            crawl_count, is_success = crawler.run()
            logger.info("%s爬虫任务执行完成 task_id=%s count=%s success=%s",
                        PLATFORM_NAME, task.get("id"), crawl_count, is_success)

            notify_result(task, crawl_count, is_success)

        except Exception as e:
            logger.error("%s 爬虫任务执行失败: %s", PLATFORM_NAME, e, exc_info=True)

        if __name__ == '__main__':
            main_loop()
        time.sleep(idle_seconds)