Python 爬虫实现增加播客访问量的方法实现

#!/usr/bin/env python
# -*- coding: utf-8 -*-
# @Author: Nick
# @Date:  2019-10-24 15:40:58
# @Last Modified by:  Nick
# @Last Modified time: 2019-10-24 16:54:31
import random
import re
import time
import urllib
import requests

from bs4 import BeautifulSoup
from fake_useragent import UserAgent

try:
  from lxml import etree
except Exception as e:
  import lxml.html
  # 实例化一个etree对象（解决通过from lxml import etree导包失败）
  etree = lxml.html.etree

# 实例化UserAgent对象，用于产生随机UserAgent
ua = UserAgent()


class BlogSpider(object):
  """
  Increase the number of CSDN blog visits.
  """

  def __init__(self):
    self.url = "https://blog.csdn.net/PY0312/article/list/{}"
    self.headers = {
      "Referer": "https://blog.csdn.net/PY0312/",
      "User-Agent": ua.random
    }
    self.firefoxHead = {
      "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:61.0) Gecko/20100101 Firefox/61.0"}
    self.IPRegular = r"(([1-9]?\d|1\d{2}|2[0-4]\d|25[0-5]).){3}([1-9]?\d|1\d{2}|2[0-4]\d|25[0-5])"

  def send_request(self, num):
    """
    模拟浏览器发起请求
    :param num: num
    :return: html_str
    """
    html_str = requests.get(self.url.format(
      num), headers=self.headers).content.decode()
    # print(html_str)

    return html_str

  def parse_data(self, html_str):
    """
    用于解析发起请求返回的数据
    :param html_str:
    :return: each_page_urls
    """
    # 将返回的 html字符串 转换为 element对象，用于xpath操作
    element_obj = etree.HTML(html_str)
    # print(element_obj)

    # 获取每一页所有blog的url
    each_page_urls = element_obj.xpath(
      '//*[@id="mainBox"]/main/div[2]/div/h5/a/@href')
    # print(each_page_urls)

    return each_page_urls

  def parseIPList(self, url="http://www.xicidaili.com/"):
    """
    爬取最新代理ip，来源：西刺代理
    注意：西刺代理容易被封，如遇到IP被封情况，采用以下两种方法即可解决：
    方法一：请参考我上一篇博客《Python 实现快代理IP爬虫》 ===> 喜欢研究的同学，可参考对接此接口
    方法二：直接屏蔽掉此接口，不使用代理也能正常使用
    :param url: "http://www.xicidaili.com/"
    :return: 代理IP列表ips
    """
    ips = []
    request = urllib.request.Request(url, headers=self.firefoxHead)
    response = urllib.request.urlopen(request)
    soup = BeautifulSoup(response, "lxml")
    tds = soup.find_all("td")
    for td in tds:
      string = str(td.string)
      if re.search(self.IPRegular, string):
        ips.append(string)
    # print(ips)
    return ips

  def main(self, total_page, loop_times, each_num):
    """
    调度方法
    :param total_page: 设置博客总页数
    :param loop_times: 设置循环次数
    :param each_num: 设置每一页要随机挑选文章数
    :return:
    """
    i = 0
    # 根据设置次数，打开循环
    while i < loop_times:
      # 遍历，得到每一页的页码
      for j in range(total_page):
        # 拼接每一页的url，并模拟发送请求, 返回响应数据
        html_str = self.send_request(j + 1)

        # 解析响应数据，得到每一页所有博文的url
        each_page_urls = self.parse_data(html_str)

        # 调用parseIPList随机产生代理IP，防反爬
        # ips = self.parseIPList()
        # proxies = {"http": "{}:8080".format(
        #   ips[random.randint(0, 40)])}

        # 遍历，每一页随机挑选each_num篇文章
        for x in range(each_num):
          # 随机抽取每一页的一篇博文进行访问，防反爬
          current_url = random.choice(each_page_urls)
          status = True if requests.get(
            current_url, headers=self.headers).content.decode() else False
          print("当前正在访问的文章是：{}，访问状态：{}".format(current_url, status))
          time.sleep(1)  # 延时1秒，防反爬
        time.sleep(1)  # 延时1秒，防反爬
      i += 1


if __name__ == '__main__':
  bs = BlogSpider()
  bs.main(7, 200, 3) # 参数参照main方法说明，酌情设置