python 会计师事务所_正则、bs4爬取会计师事务所新闻信息——以利安达会计师事务所为例...

预期效果

可以看到，该事务所有1625条新闻纪录，一个一个点开查看判断，再下载所需要的新闻信息(发布时间、标题、内容、链接)，比较麻烦，用爬虫可以较快实现这一过程。

代码实现

import requests

from bs4 import BeautifulSoup

import re

import csv

n=0

f= open('news6', 'w', encoding='utf-8-sig', newline='')

writer = csv.writer(f)

head=['time','title','content','url','key']

writer.writerow(head)

for x in range(41,4169):

url1='http://ltd.reanda.com/dongtai/news_list.asp?'

url=url1+'id='+str(x)

html=requests.get(url)

html.encoding = 'GBK'

if html.status_code==200:

wrong=re.findall('数(据库出)错',html.text)

if wrong==[]:

soup = BeautifulSoup(html.text, 'lxml')

ftitle = soup.find_all(attrs={'class': 'big'})

title = ftitle[0].text

for key in ['调研','莅临','访','研讨','邀','到','视察','接见']:

result = re.search(key, title)

if result:

a=soup.find_all(attrs={'class':'neirong'})

ftime=a[1].text

print(ftime)

time=re.findall('发布时间：(.*)阅读',ftime)

str2=time[0]

x=str2.split('-')

time='{}/{}/{}'.format(x[0],x[1],x[2])

n+=1

print(n)

print('爬取筛选出第{}条新闻，发布时间是{}'.format(n,time))

content=a[0].text

# print(content)

print('关键字为{}'.format(key))

row=[time,title,content,url,key]

writer.writerow(row)

f.close()