爬取饿了么某一地址附近全部商家

如下是爬取饿了么某一地址附近全部商家的简单实现,
可细化点:
1.发现同一手机号请求验证码3次后被要求图形验证,还没去研究解决方法.简单加了个KeyError跳过.

import requests
session = requests.session()

headers = {
'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/70.0.3538.110 Safari/537.36'
}
url_1 = 'https://h5.ele.me/restapi/eus/login/mobile_send_code'
tel = input('请输入手机号码：')
data_1 = {'captcha_hash':'',
        'captcha_value':'',
        'mobile':tel,
        'scf':''}

get_token = session.post(url_1, headers=headers, data=data_1).json()

try:
    token = get_token['validate_token']
    url_2 = 'https://h5.ele.me/restapi/eus/login/login_by_mobile'
    code = input('请输入手机验证码：')
    data_2 = {'mobile':tel,
            'scf':'ms',
            'validate_code':code,
            'validate_token':token
            }

    session.post(url_2,headers=headers,data=data_2)


    address_url = 'https://www.ele.me/restapi/v2/pois?'
    place = input('请输入你的收货地址：')
    params = {'extras[]':'count','geohash':'ws105rz9smwm','keyword':place,'limit':'20','type':'nearby'}
    # 这里使用了深圳的geohash

    address_res = requests.get(address_url,params=params)
    address_json = address_res.json()

    print('以下，是与'+place+'相关的位置信息：\n')
    n=0
    for address in address_json:
        print(str(n)+'. '+address['name']+'：'+address['short_address']+'\n')
        n = n+1
    address_num = int(input('请输入您选择位置的序号：'))
    final_address = address_json[address_num]

    restaurants_url = 'https://www.ele.me/restapi/shopping/restaurants?'
    # 使用带有餐馆列表的那个XHR地址。
    params = {
        'extras[]':'activities',
        'geohash':final_address['geohash'],
        'latitude':final_address['latitude'],
        'limit':'24',
        'longitude':final_address['longitude'],
        'offset':'0',
        'terminal':'web'
    }
    #其中geohash和经纬度，来自前面获取到的数据。
    restaurants_res = session.get(restaurants_url,params=params)
    restaurants = restaurants_res.json()
    for restaurant in restaurants:
        print(restaurant['name'])
except KeyError:
    print(get_token)