網絡爬蟲--14.【糗事百科實戰】

一. 要求

爬取糗事百科段子，假設頁面的URL是 http://www.qiushibaike.com/8hr/page/1

1. 使用requests獲取頁面信息，用XPath / re 做數據提取

2. 獲取每個帖子裏的用戶頭像鏈接、用戶姓名、段子內容、點贊次數和評論次數

3. 保存到 json 文件內

二. 參考代碼

#coding=utf-8
import requests
from retrying import retry
from lxml import etree

class Qiubai_spider():
    def __init__(self):
        self.url = "http://www.qiushibaike.com/8hr/page/{}/"
        self.headers = {
            "User-Agent":"Mozilla/5.0 (compatible; MSIE 9.0; Windows NT 6.1 Trident/5.0;"
        }

    @retry(stop_max_attempt_number=5) #調用retry，當assert出錯時候，重複請求5次
    def parse_url(self,url):
        response = requests.get(url,timeout=10,headers=self.headers) #請求url
        assert response.status_code==200  #當響應碼不是200時候，做斷言報錯處理
        print(url)
        return etree.HTML(response.text) #返回etree之後的html

    def parse_content(self,html):
        item_temp = html.xpath("//div[@class='article block untagged mb15']")
        print(len(item_temp))
        for item in item_temp:
            #獲取用戶頭像地址
            avatar = item.xpath("./div[1]/a[1]/img/@src")[0] if len(item.xpath("./div[1]/a[1]/img/@src"))>0 else None
            #爲頭像地址添加前綴
            if avatar is not None and not avatar.startswith("http:"):
                avatar = "http:"+avatar
            print(avatar)
            name = item.xpath("./div[1]/a[2]/h2/text()")[0] #獲取用戶名
            print(name)
            content = item.xpath("./a[@class='contentHerf']/div/span/text()")[0] #獲取內容
            print(content)
            star_number = item.xpath("./div[@class='stats']/span[1]/i/text()")[0] #獲取點贊數
            print(star_number)
            comment_number = item.xpath("./div[@class='stats']/span[2]/a/i/text()")[0] #獲取評論數
            print(comment_number)
            print("*"*100)

    def run(self):
        '''函數的主要邏輯實現
        '''
        url = self.url.format(1) #獲取到url
        html = self.parse_url(url) #請求url
        self.parse_content(html) #解析頁面內容並把內容存入內容隊列

if __name__ == "__main__":
    qiubai = Qiubai_spider()
    qiubai.run()