Python-爬取某条微博下的图片评论

替换账号密码，模拟微博登录并爬取评论下的所有图片评论
写着玩的，用的是selenium，还没来得及加phantomjs，没用函数，一顺写下来的，写的比较乱，效率也不是太高，见谅

纯粹瞎搞，凑活能用

# -*- coding: utf-8 -*-
"""
Created on Tue Nov 27 16:25:29 2018

@author: Administrator
"""
from selenium import webdriver
from bs4 import BeautifulSoup
import time
import re
import os
import urllib.request

#模拟登录
username = '你的账号'
password = '你的密码'
url1 = 'http://login.sina.com.cn/'
chromedriver_dir = r'C:\Users\Administrator\AppData\Local\CentBrowser\Application\chromedriver.exe' # 注意，这里是你的安装路径
browser = webdriver.Chrome(chromedriver_dir)
browser.get(url1)
elem_user = browser.find_element_by_name("username")
elem_user.send_keys(username) #用户名
elem_pwd = browser.find_element_by_name("password")
elem_pwd.send_keys(password)  #密码

time.sleep(20)
        
elem_sub = browser.find_element_by_xpath("//input[@class='W_btn_a btn_34px']")
elem_sub.click()              #点击登陆 因无name属性
time.sleep(2)

#打开需要爬的页面
url2 = 'https://weibo.com/1740806873/Guauhitiu?refer_flag=1001030103_&type=comment#_rnd1543317436501'
browser.get(url2)
#滚动到底部并加载更多页面
n=1;
try:
    while n <= 100:
        js='var q=document.documentElement.scrollTop=100000'  
        browser.execute_script(js)  
        time.sleep(5)
        while n >= 3:
            browser.find_element_by_xpath("//span[@class='more_txt']").click() #点击查看更多
            time.sleep(5)
        n=n+1
except:
    print('加载完成')

os.mkdir('wbpic') #创建文件夹
os.chdir('wbpic') #在文件夹内操作
#获取图片链接
data = browser.page_source
soup1 = BeautifulSoup(data, 'lxml')
imglist = soup1.find_all('img', {"src":re.compile("\/\/.*thumb180.*\.jpg")})  #发现html中带img标签的数据，输出格式为<img xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx，存入集合
lenth = len(imglist)  #计算集合的个数
for i in range(lenth):
    picaddress = imglist[i].attrs['src'] #抓取img中属性为src的信息,例如<img src="123456" xxxxxxxxxxxxxxxx,则输出为123456
    picaddress = picaddress.replace('//','https://')
    picaddress = picaddress.replace('thumb180','bmiddle')
    #print(picaddress)
    filename = picaddress.split('/')[-1] #切片，从/开始到结尾
    with open(filename,'wb') as f:
        img = urllib.request.urlopen(picaddress).read() #下载进程
        f.write(img)

browser.close()