1. 程式人生 > >實例:模擬登陸豆瓣

實例:模擬登陸豆瓣

保存 TP spider orm utf attr sta com parse

# -*- coding: utf-8 -*-
import scrapy
import urllib.request

# https://accounts.douban.com/login

class DoubanSpider(scrapy.Spider):
  name = ‘douban‘
  allowed_domains = [‘www.douban.com‘, ‘accounts.douban.com‘]
  start_urls = [‘https://accounts.douban.com/login‘]

  def parse(self, response):
   # 查找驗證碼圖片,看有沒有驗證碼
   image = response.xpath(‘//img[@id="captcha_image"]/@src‘)
   # 判斷image這個列表是否為空,如果為空,就是沒有驗證碼
   if len(image) == 0:
    print(‘不帶驗證碼的‘ * 10)
    # 不帶驗證碼的
    formdata = {
    ‘source‘: ‘index_nav‘,
    ‘form_email‘: ‘[email protected]‘,
    ‘form_password‘: ‘lizhibin666‘,
    }
   else:
    print(‘帶驗證碼的‘ * 10)
    # 通過屬性選擇器獲取得到
    captchaid = response.css(‘input[name="captcha-id"]::attr(value)‘).extract_first()
    # 獲取驗證碼鏈接
    image_url = image.extract_first()
    # print(‘*‘ * 50)
    # print(captchaid)
    # print(image_url)
    # print(‘*‘ * 50)
    urllib.request.urlretrieve(image_url, ‘code.png‘)
    code = input(‘請輸入驗證碼:‘)
    # 帶驗證碼的
    formdata = {
    ‘source‘: ‘None‘,
    ‘redir‘: ‘https://www.douban.com/‘,
    ‘form_email‘: ‘[email protected]‘,
    ‘form_password‘: ‘lizhibin666‘,
    ‘captcha-solution‘: code,
    ‘captcha-id‘: captchaid,
    ‘login‘: ‘登錄‘,
    }

    post_url = ‘https://accounts.douban.com/login‘
    # 發送post請求
    yield scrapy.FormRequest(url=post_url, formdata=formdata, callback=self.lala)
  #保存文件,查看是否登錄
  def lala(self, response):
    print(‘*‘ * 50)
    with open(‘douban.html‘, ‘wb‘) as fp:
    fp.write(response.body)
    print(‘*‘ * 50)

實例:模擬登陸豆瓣