使用scrapy抓取大众点评网美食的商家名字，人均消费，地址等等信息

由于大众点评网需要登录才能访问，因此需要先获取登录后的cookie，然后在请求时带上cookie。

以下是一个示例代码，可以抓取大众点评网上某个城市的美食商家的名称、人均消费、地址等信息：

import scrapy
from scrapy.http.cookies import CookieJar


class DianpingSpider(scrapy.Spider):
    name = 'dianping'
    allowed_domains = ['www.dianping.com']
    custom_settings = {
        'USER_AGENT': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/58.0.3029.110 Safari/537.36 Edge/16.16299',
    }

    def start_requests(self):
        # 通过登录获取cookie
        return [scrapy.Request(url='https://account.dianping.com/login?redir=https://www.dianping.com/',
                               callback=self.parse_login)]

    def parse_login(self, response):
        formdata = {
            'username': 'your_username',
            'password': 'your_password',
            'redir': 'https://www.dianping.com/',
            'geetest_challenge': '',
            'geetest_validate': '',
            'geetest_seccode': '',
        }
        # 提交登录表单
        yield scrapy.FormRequest.from_response(
            response,
            formdata=formdata,
            callback=self.after_login
        )

    def after_login(self, response):
        # 验证登录是否成功
        if '我的点评' in response.text:
            # 登录成功，开始抓取数据
            cookie_jar = CookieJar()
            cookie_jar.extract_cookies(response, response.request)
            cookies = {}
            for cookie in cookie_jar:
                cookies[cookie.name] = cookie.value
            # 抓取某个城市的美食商家列表
            city = 'shenzhen'
            url = f'https://www.dianping.com/{city}/ch10'
            yield scrapy.Request(url=url, cookies=cookies, callback=self.parse)

    def parse(self, response):
        # 抓取每个商家的信息
        for item in response.css('div.tit > a'):
            name = item.css('::text').get()
            url = item.css('::attr(href)').get()
            yield scrapy.Request(url=url, cookies=response.request.cookies, callback=self.parse_detail,
                                 meta={'name': name})

    def parse_detail(self, response):
        # 抓取商家的详细信息
        name = response.meta['name']
        avg_price = response.css('span.avg-price > b::text').get()
        address = response.css('div.expand-info.address span::text').get()
        phone = response.css('p.expand-info.tel span::text').get()
        yield {
            'name': name,
            'avg_price': avg_price,
            'address': address,
            'phone': phone,
        }

需要将代码中的`your_username`和`your_password`替换为自己的账号和密码。这里使用了`scrapy.http.cookies.CookieJar`类来获取登录后的cookie，并将cookie在后续的请求中带上。在抓取商家的详细信息时，使用了`meta`参数来传递商家名称，以便后续整合数据。