小米有品关键词商品搜索接口实战：结构化分页与数据归一化方案

import requests
import time

class YouPinSearch:
    def __init__(self):
        self.base_url = "https://m.xiaomiyoupin.com/api/search/goods"
        self.headers = {
            "User-Agent": "Mozilla/5.0 (iPhone; CPU iPhone OS 16 like Mac OS X) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/16 Mobile/15E148 Safari/604.1",
            "Referer": "https://m.xiaomiyoupin.com/search"
        }
        self.session = requests.Session()

    def search_goods(self, keyword, page=1, page_size=20, sort=0):
        """
        sort:0综合，1销量，2价格升序，3价格降序
        """
        params = {
            "keyword": keyword,
            "page": page,
            "limit": page_size,
            "sort": sort
        }
        try:
            resp = self.session.get(self.base_url, headers=self.headers, params=params, timeout=8)
            resp.raise_for_status()
            res_json = resp.json()
            if res_json.get("code") != 0:
                return []
            raw_list = res_json.get("data", {}).get("list", [])
            result = []
            for item in raw_list:
                result.append({
                    "item_id": item.get("itemId"),
                    "title": item.get("title"),
                    "price": item.get("price"),
                    "sale": item.get("saleNum"),
                    "pic": item.get("picUrl"),
                    "shop_name": item.get("shopName")
                })
            return result
        except Exception as e:
            print(f"请求异常：{e}")
            return []

    def loop_search(self, keyword, max_page=3):
        all_data = []
        for p in range(1, max_page + 1):
            page_data = self.search_goods(keyword, page=p)
            if not page_data:
                break
            all_data.extend(page_data)
            time.sleep(1.2)
        return all_data

if __name__ == "__main__":
    sp = YouPinSearch()
    data = sp.loop_search("保温杯", max_page=2)
    for row in data:
        print(row)