爬取网站为:http://xiaohua.zol.com.cn/youmo/
查看网页机构,爬取笑话内容时存在如下问题:
1、每页需要进入“查看更多”链接下面网页进行进一步爬取内容每页查看更多链接内容比较多,多任务进行,这里采用线程池的方式,可以有效地控制系统中并发线程的数量。避免当系统中包含有大量的并发线程时,导致系统性能下降,甚至导致 Python 解释器崩溃,引入线程池,花费时间更少,更效率。
- 创建线程 池threadpool.ThreadPool()
- 创建需要线程池处理的任务即threadpool.makeRequests(),makeRequests存放的是要开启多线程的函数,以及函数相关参数和回调函数,其中回调函数可以不写(默认是无)。
- 将创建的多个任务put到线程池中,threadpool.putRequest()
- 等到所有任务处理完毕theadpool.pool()
2、查看链接笑话页内容,div元素内部文本分布比较混乱。有的分布在<p>链接内有的属于div的文本,可采用正则表达式的方式解决。
注意2种获取元素节点的方式:
1)lxml获取节点字符串
res=requests.get(url,headers=headers) html = res.text lxml 获取节点写法 element=etree.HTML(html) divEle=element.xpath("//div[@class='article-text']")[0] # 获取div节点 div= etree.tostring(divEle, encoding = 'utf-8' ).decode('utf-8') # 转换为div字符串
2)正则表达式写法1,过滤回车、制表符和p标签
# 第一种方式:replace content = re.findall('<div class="article-text">(.*?)</div>',html,re.S) content = content[0].replace(' ','').replace(' ','').replace('<p>','').replace('</p>','').strip()
3)正则表达式写法2,过滤回车、制表符和p标签
# 第二种方式:sub for index in range(len(content)): content[index] = re.sub(r'( | |<p>|</p>)+','',content[index]).strip() list = ''.join(content) print(list)
3、完整代码
index.py
import requests import threadpool import time import os,sys import re from lxml import etree from lxml.html import tostring class ScrapDemo(): next_page_url="" #下一页的URL page_num=1 #当前页 detail_url_list=0 #详情页面URL地址list deepth=0 #设置抓取的深度 headers = { "user-agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/68.0.3440.84 Safari/537.36" } fileNum=0 def __init__(self,url): self.scrapyIndex(url) def threadIndex(self,urllist): #开启线程池 if len(urllist) == 0: print("请输入需要爬取的地址") return False ScrapDemo.detail_url_list=len(urllist) pool=threadpool.ThreadPool(len(urllist)) requests=threadpool.makeRequests(self.detailScray,urllist) for req in requests: pool.putRequest(req) time.sleep(0.5) pool.wait() def detailScray(self,url): # 获取html结构 if not url == "": url='http://xiaohua.zol.com.cn/{}'.format(url) res=requests.get(url,headers=ScrapDemo.headers) html=res.text # element=etree.HTML(html) # divEle=element.xpath("//div[@class='article-text']")[0] # Element div self.downloadText(html) def downloadText(self,ele): # 抓取数据并存为txt文件 clist = re.findall('<div class="article-text">(.*?)</div>',ele,re.S) for index in range(len(clist)): ''' 正则表达式:过滤掉回车、制表符和p标签 ''' clist[index]=re.sub(r'( | |<p>|</p>)+','',clist[index]) content="".join(clist) # print(content) basedir=os.path.dirname(__file__) filePath=os.path.join(basedir) filename="xiaohua{0}-{1}.txt".format(ScrapDemo.deepth,str(ScrapDemo.fileNum)) file=os.path.join(filePath,'file_txt',filename) try: f=open(file,"w") f.write(content) if ScrapDemo.fileNum == (ScrapDemo.detail_url_list - 1): print(ScrapDemo.next_page_url) print(ScrapDemo.deepth) if not ScrapDemo.next_page_url == "": self.scrapyIndex(ScrapDemo.next_page_url) except Exception as e: print("Error:%s" % str(e)) ScrapDemo.fileNum=ScrapDemo.fileNum+1 print(ScrapDemo.fileNum) def scrapyIndex(self,url): if not url == "": ScrapDemo.fileNum=0 ScrapDemo.deepth=ScrapDemo.deepth+1 print("开启第{0}页抓取".format(ScrapDemo.page_num)) res=requests.get(url,headers=ScrapDemo.headers) html=res.text element=etree.HTML(html) a_urllist=element.xpath("//a[@class='all-read']/@href") # 当前页所有查看全文 next_page=element.xpath("//a[@class='page-next']/@href") # 获取下一页的url ScrapDemo.next_page_url='http://xiaohua.zol.com.cn/{}'.format(next_page[0]) if not len(next_page) == 0 and ScrapDemo.next_page_url != url: ScrapDemo.page_num=ScrapDemo.page_num+1 self.threadIndex(a_urllist[:]) else: print('下载完成,当前页数为{}页'.format(ScrapDemo.page_num)) sys.exit()
runscrapy.py
from app import ScrapDemo url="http://xiaohua.zol.com.cn/youmo/" ScrapDemo(url)
运行如下:
总共1988个文件,下载完成。