前言

一些感悟：果然做过anti的人最擅长作弊

Version-2.0

工作之余更新了一版，更新如下： 1.增加了对用户多页文章对获取 2.增加了点击顺序的随机性 3.增加了点击行为时间的随机性 4.增加了点击内容的随机性 5.增加了点击的轮次

缺陷与不足

1.UA还未设置 2.未有多个IP

import requests from bs4 import BeautifulSoup import time import random #对用户多页文章对获取 def get_writer_article_list(base_url,page_num): all_article_list = [] for i in range(page_num): index = i +1 print('cur index is ' + str(index)) cur_page_url = base_url + str(index) all_article_list = get_article_list(cur_page_url) + all_article_list return all_article_list #获取单页所有文章url def get_article_list(base_url): web_data = requests.get(base_url) soup = BeautifulSoup(web_data.text,'lxml') divs = soup.find_all('div', class_='article-item-box csdn-tracking-statistics') url_list = [] for div in divs: label = div.find_all('a') url = label[0].get('href') url_list.append(url) return url_list #生成每一轮点击的随机list def click_random(url_list,min_random_rate): new_url_list = [] max_url_count = len(url_list) min_url_count = int(max_url_count*min_random_rate) term_url_count = random.randint(min_url_count,max_url_count) for i in range(term_url_count): ramdom_index = random.randint(0,max_url_count-1) new_url_list.append(url_list[ramdom_index]) return new_url_list #多轮点击 def click_article_url(term_num,click_random_start,click_random_end,term_random_start,term_random_end,all_list): for i in range(term_num): term_url_list = click_random(all_list,0.7) for url in term_url_list: requests.get(url) print('click for '+url) click_sleep_time = random.randint(click_random_start,click_random_end) time.sleep(click_sleep_time) print('sleep for '+str(click_sleep_time)) term_num = i +1 print('finish the term of '+str(term_num)) term_sleep_time = random.randint(term_random_start,term_random_end) time.sleep(term_sleep_time) print('sleep for the term '+str(term_sleep_time)) base_url1 = "https://blog.csdn.net/xxx1/article/list/" base_url2 = "https://blog.csdn.net/xxx2/article/list/" url_list_1 = get_writer_article_list(base_url1,2) url_list_2 = get_writer_article_list(base_url2,2) all_list = url_list_1 + url_list_2 click_article_url(200,8,50,30,60,all_list)

Version-1.0

利用午休的时间，用Python写了一个刷Blog中文章阅读量的小玩具，还不成熟，最近会迭代修改。

当前的整体思路：获取用户的文章页，解析html，获取该页面每一篇文章的url，存入list，然后训练访问list中的url，达到点击的目的。

import requests from bs4 import BeautifulSoup import time //爬取我的粉丝 周英俊 同学的blog base_url = "https://blog.csdn.net/qq_38835878/article/list/1" web_data = requests.get(base_url) soup = BeautifulSoup(web_data.text,'lxml') divs = soup.find_all('div', class_='article-item-box csdn-tracking-statistics') url_list = [] for div in divs: label = div.find_all('a') url = label[0].get('href') url_list.append(url) for url in url_list: requests.get(url) print('request for '+url) time.sleep(61) print('sleep for 61s')

缺陷与不足

没有加入翻页功能没有设计代理Ip没有设置UA行为规律性较强：session时长无偏差点击行为pattern明显：顺序访问所有文章。