背景：
- 1.续上次被封号后，随着经验+3，我今天刚解的号又被封了。。。
- 2.本次v2版更新内容点：
- （1）回复语句多样化；
- （2）增加键盘输入功能；
1.需求分析：
-
1.需求：经验+3
-
2.分析：封号机制是什么？
- 2.1.大量回复固定的词 ->不是这个，词库我都准备好了，还是被封了
- 2.2.回复时间的间隔 ->可能有关，我10s一回复能回复三页，2s一回复20句不到就被封了
-
3.当弹出验证码框时如何操作？
- 这个还没刷出来就被封了
2.源码展示：
from lxml import etree
from selenium import webdriver
import time
import random
from pynput.keyboard import Controller
chrome_driver=r"C:\Program Files (x86)\Google\Chrome\chromedriver.exe"
header = {"User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/80.0.3987.122 Safari/537.36"}
count_num = 0
count_page = 0
host = "https://tieba.baidu.com"
reply_list = [
"经验加三！",
"哈哈，不懂",
"卧槽！牛批",
"我过去就是一个滑稽#(滑稽)",
"/手动滑稽#(滑稽)",
"#(滑稽)","#(滑稽)",
"秀我那瓜子了哈哈",
"哈哈",
"不懂撒",
"这是什么","gkd",
"？？？",
"不知道说什么#(滑稽)",
"你知道的，我只想水一波经验#(滑稽)",
"经验加三，告辞！！#(滑稽)",
"#(滑稽)"
]
keyboard = Controller() # 控制键对象
opt = webdriver.ChromeOptions()
opt.headless = False # 是否隐藏浏览器 True为隐藏
browser = webdriver.Chrome(executable_path=chrome_driver, options=opt)
def browser_of(host_idx, idx):
global host
global count_num
global reply_list
print("本章："+ host_idx, end=" --- ")
browser.get(host_idx)
re = browser.page_source
html = etree.HTML(re)
urls = html.xpath('//div[@class="threadlist_title pull_left j_th_tit "]/a/@href')
time.sleep(5) # 给扫描时间
print("扫描有{}个贴，准备开始水！".format(len(urls)))
j = 0
for url in urls:
url = host + url
print(url)
browser.get(url)
try:
browser.find_element_by_id("ueditor_replace").click() # 鼠标点击
time.sleep(2)
input_box = browser.find_element_by_id("ueditor_replace")
list_len = len(reply_list)
ran_idx = random.randint(0, list_len)
if opt.headless: # 判断浏览器是否显示状态，显示就逐个打印
time.sleep(4)
input_box.send_keys(reply_list[ran_idx] + "经验+3!")
time.sleep(4)
else:
input_box.send_keys(reply_list[ran_idx])
# ---------------每句后面增加经典回复------------------
keyboard.press("e")
time.sleep(1)
keyboard.press("x")
time.sleep(1)
keyboard.press("p")
time.sleep(1)
keyboard.press("e")
time.sleep(1)
keyboard.press("r")
time.sleep(1)
keyboard.press("i")
time.sleep(1)
keyboard.press("e")
time.sleep(1)
keyboard.press("n")
time.sleep(1)
keyboard.press("c")
time.sleep(1)
keyboard.press("e")
#-----------------------------------------
j += 1
count_num += 1
print('{0} - 第{1}页 第{2}条 回复成功：经验+3！'.format(count_num, idx, j))
browser.find_element_by_css_selector(".ui_btn.ui_btn_m.j_submit.poster_submit").click()
except Exception as e:
print(e)
print('fail')
time.sleep(10)
return True
def get_to_next_page(host_idx):
global count_page
count_page += 1
print("本页为 = "+host_idx)
browser.get(host_idx)
re = browser.page_source # 获取网页源码
html = etree.HTML(re)
next_url = html.xpath('//div[@id="frs_list_pager"]/a/@href')[-2] # 获取下一页链接
print("下一页 = "+next_url)
# "file://tieba.baidu.com/f?kw=%E6%BB%91%E7%A8%BD&ie=utf-8&pn=50"
if browser_of(host_idx, count_page): # 执行完成
get_to_next_page("https:" + next_url)
browser.close()
if __name__ == '__main__':
host_idx = "https://tieba.baidu.com/f?kw=%E6%BB%91%E7%A8%BD&ie=utf-8&pn=0" # 放置贴吧内第一页url
get_to_next_page(host_idx)
3.源码说明：
关于键盘输入：如果要控制其他程序，建议把opt.headless改为True，光标在哪就在哪敲字
把opt.headless改为True又扫描登录不到，哈哈，可以自行查找selenium 保存账号和密码的操作
4.运行结果：
5.关于：
1.大量水贴有风险
2.记得要登录沃
3.转载请说明出处！！！