- 分析
#通过非贪婪模式将格式去除掉
pattern = re.compile('<.*?>')
br标签是换行符
由于此时所有格式均被清洗爬取结果不易读
为了保持爬取的内容格式和网页显示的一样可读,还需要在替换字符之前,将换行符br标签替换成换行符
post = post.replace('<br/>','/n')
- 代码实现
import re
import requests
from lxml import etree
import html
class PostsCrawler:
domain = "https://www.newsmth.net"
#通过非贪婪模式将格式去除掉
pattern = re.compile('<.*?>')
#获取网页内容
def get_content(self,topic_url,page):
querystring = {"ajax": "", "p": str(page)}
url = self.domain + topic_url
r = requests.get(url,params=querystring)
#方便调用
self.html = r.text
self.tree = etree.HTML(r.text)
#获取最大页码
def get_max_page(self):
#浏览器上下栏有两个分页设置,因此选择第一个即可
pages = self.tree.xpath('//ol[@class="page-main"][1]/li')
#只有一页的情况
if len(pages) == 1:
return 1
#页面没有在最后一页时最大的页码
last_page_test = pages[len(pages)-1].xpath('a')[0].text
#如果页面在最后一页
if last_page_test == '>>':
return int(pages[len(pages)-2].xpath('a')[0].text)
return last_page_test
#获取帖子内容
def get_posts(self):
c_elements = self.tree.xpath('//td[@class="a-content"]')
posts = []
for c_element in c_elements:
post = c_element.xpath('//p')[0]
post = etree.tostring(post).decode('GBK')
#增加可读性,在替换字符之前,将换行符br标签替换成换行符
post = post.replace('<br/>','/n')
post = self.pattern.sub('',post)
post = html.unescape(post)
posts.append(post)
return posts
if __name__ == "__main__":
url = "/nForum/article/AutoWorld/1942076572"
post_crawler = PostsCrawler()
#获取网页第一页内容
post = post_crawler.get_content(url,1)
posts = post_crawler.get_posts()
page_count = post_crawler.get_max_page()
print(page_count)
#获取所有网页的内容
if page_count > 1:
for i in range(2,page_count +1):
post = post_crawler.get_content(url, i)
posts += post_crawler.get_posts()
if i>=2:
break
i = 1 #分楼层
for p in posts:
print(p)
print("=============================",i,"=============================")
print("")
i +=1