Scrapy模拟登录赶集网的实现代码

import scrapy
import re

class GanjiSpider(scrapy.Spider):
  name = 'ganji'
  allowed_domains = ['ganji.com']
  start_urls = ['https://passport.ganji.com/login.php']

  def parse(self, response):
    hash_code = re.search(r'"__hash__":"(.+)"}', response.text).group(1) # 正则获取哈希
    img_url = 'https://passport.ganji.com/ajax.php?dir=captcha&module=login_captcha' # 验证码url
    yield scrapy.Request(img_url, callback=self.do_formdata, meta={'hash_code': hash_code}) # 发送获取验证码请求并保存验证码到本地

  def do_formdata(self, response):
    with open('yzm.jpg', 'wb') as f:
      f.write(response.body)
      # 验证码三种方案：1，保存下来手动输入，2，云打码，3 tesseract模块，在这里我们手动输入
    code = input('请输入验证码：')
    # 创建表单
    formdata = {
      'username': 'your_username',
      'password': 'your_password',
      'setcookie': '14',
      'checkCode': code,
      'next': '',
      'source': 'passport',
      '__hash__': response.request.meta['hash_code'] # meta是在respose.request中
    }
    login_url = "https://passport.ganji.com/login.php"
    yield scrapy.FormRequest(url=login_url, formdata=formdata, callback=self.after_login) # 发送登录请求

  def after_login(self, response):
    print(response.text)