2.0/start_run_yaoex.py

"""
壹药城 (Yaoex) 平台采集入口
使用 commons1 调度框架 —— 从内部 API 拉取任务，由 YaoexCrawler / YaoexSnapshotCrawl 执行采集。
"""
import random
import re
import time
from datetime import datetime

import requests

from commons.Logger import get_spider_logger
from commons.feishu_webhook import send_text, send_error_card
from spiders.yaoex.yaoex_crawl import (
    YaoexCrawler,
    PLATFORM_ID,
    YYC_DEVICE_ID,
    TASK_PULL_URL,
    TASK_HEARTBEAT_URL,
    TASK_API_TOKEN,
)
from spiders.yaoex.yaoex_snapshot_crawl import YaoexSnapshotCrawl

logger = get_spider_logger("yaoex_runner")

PLATFORM_NAME = "壹药城"
IDLE_SECONDS_MIN = 180
IDLE_SECONDS_MAX = 300


def pull_task():
    """从内部调度 API 拉取任务"""
    headers = {'X-Crawler-Token': TASK_API_TOKEN}
    params = {'platform': PLATFORM_ID, 'username': YYC_DEVICE_ID}
    try:
        response = requests.get(TASK_PULL_URL, params=params, headers=headers, timeout=15)
        result = response.json()
        logger.info("拉取任务 API 返回: %s", result)
        if result.get('code') == 'success':
            task = result.get('data', {}).get('task') or {}
            if task and task.get("id"):
                logger.info("拉取任务成功 platform=%s task_id=%s", PLATFORM_ID, task.get("id"))
                return task
            else:
                logger.info("拉取任务成功但无有效任务（task=%s），视为暂无任务", task)
                return {}
        else:
            logger.warning("拉取任务 API 返回非 success: code=%s, msg=%s",
                           result.get('code'), result.get('msg'))
            return {}
    except Exception as e:
        logger.error("拉取任务 API 请求失败: %s", e)
        return {}


def send_heartbeat():
    """向调度 API 上报心跳，告知服务器爬虫在线"""
    headers = {'X-Crawler-Token': TASK_API_TOKEN}
    payload = {"platform": str(PLATFORM_ID), "username": YYC_DEVICE_ID}
    try:
        resp = requests.post(TASK_HEARTBEAT_URL, json=payload, headers=headers, timeout=10)
        logger.debug("心跳上报: HTTP %s, %s", resp.status_code, resp.text[:200])
    except Exception as e:
        logger.warning("心跳上报失败: %s", e)


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"""
    idle_seconds = random.randint(IDLE_SECONDS_MIN, IDLE_SECONDS_MAX)
    logger.info("壹药城采集启动，每轮间隔 %s 秒", idle_seconds)

    while True:
        try:
            # 先上报心跳，告诉服务器爬虫在线
            send_heartbeat()
            task = pull_task()
            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

            # 根据 snapshot_collect_status 选择爬虫类型
            snapshot_status = task.get("snapshot_collect_status", 1)
            if snapshot_status == 0:
                spider_cls = YaoexSnapshotCrawl
                logger.info("snapshot_collect_status=%s → 启用快照模式 YaoexSnapshotCrawl", snapshot_status)
            else:
                spider_cls = YaoexCrawler
                logger.info("snapshot_collect_status=%s → 启用接口模式 YaoexCrawler", snapshot_status)

            logger.info("开始执行%s爬虫任务 task_id=%s", PLATFORM_NAME, task.get("id"))
            crawler = spider_cls(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)

        time.sleep(idle_seconds)


if __name__ == '__main__':
    main_loop()