1.从新闻url获取新闻详情： 字典,anews
1 # -*- coding: utf-8 -*- 2 """ 3 Spyder Editor 4 5 This is a temporary script file. 6 """ 7 import requests 8 from bs4 import BeautifulSoup 9 from datetime import datetime 10 import re 11 12 def click(url): 13 id = re.findall('(d{1,5})',url)[-1] 14 clickUrl='http://oa.gzcc.cn/api.php?op=count&id={}&modelid=80'.format(id) 15 resClick = requests.get(clickUrl) 16 newsClick = int(resClick.text.split('.html')[-1].lstrip("('").rstrip("');")) 17 return newsClick 18 19 def newsdt(showinfo): 20 newsDate = showinfo.split()[0].split(':')[1] 21 newsTime = showinfo.split()[1] 22 newDT = newsDate+' '+newsTime 23 dt = datetime.strptime(newDT,'%Y-%m-%d %H:%M:%S') 24 return dt 25 26 def anews(url): 27 newsDetail = {} 28 res = requests.get(url) 29 res.encoding = 'utf-8' 30 soup = BeautifulSoup(res.text,'html.parser') 31 newsDetail['nenewsTitle'] = soup.select('.show-title')[0].text 32 showinfo = soup.select('.show-info')[0].text 33 newsDetail['newsDT'] = newsdt(showinfo) 34 newsDetail['newsClick'] = click(newsUrl) 35 return newsDetail 36 37 newsUrl = 'http://news.gzcc.cn/html/2019/xiaoyuanxinwen_0404/11155.html' 38 anews(newsUrl) 39 40 res=requests.get('http://news.gzcc.cn/html/xiaoyuanxinwen') 41 res.encoding='utf-8' 42 soup = BeautifulSoup(res.text,'html.parser') 43 44 for news in soup.select('li'): 45 if len(news.select('.news-list-title'))>0: 46 newsUrl = news.select('a')[0]['href'] 47 print(anews(newsUrl))
2.从列表页的url获取新闻url：列表append(字典) alist
1 listUrl='http://news.gzcc.cn/html/xiaoyuanxinwen/' 2 res = requests.get(listUrl) 3 res.encoding='utf-8' 4 soup = BeautifulSoup(res.text,'html.parser') 5 newsList=[] 6 for news in soup.select('li'): 7 if len(news.select('.news-list-title'))>0: 8 newsUrl=news.select('a')[0]['href'] 9 newsDesc=news.select('.news-list-description')[0].text 10 newsDict=anews(newsUrl) 11 newsDict['description']=newsDesc 12 newsList.append(newsDict) 13 print(newsList)
3.生成所页列表页的url并获取全部新闻 ：列表extend(列表) allnews
*每个同学爬学号尾数开始的10个列表页
1 def alist(url): 2 res=requests.get(listUrl) 3 res.encoding='utf-8' 4 soup = BeautifulSoup(res.text,'html.parser') 5 newsList=[] 6 for news in soup.select('li'): 7 if len(news.select('.news-list-title'))>0: 8 newsURL=news.select('a')[0]['href'] 9 newsDesc=news.select('.news-list-description')[0].text 10 newsDict=anews(newsURL) 11 newsDict['description']=newsDesc 12 newsList.append(newsDict) 13 return newsList 14 listUrl='http://news.gzcc.cn/html/xiaoyuanxinwen/' 15 alist(listUrl) 16 17 allnews=[] 18 for i in range(78,88): 19 listUrl='http://news.gzcc.cn/html/xiaoyuanxinwen/{}.html'.format(i) 20 allnews.extend(alist(listUrl)) 21 len(allnews) 22 print(allnews)
4.设置合理的爬取间隔
1 allnews=[] 2 for i in range(78,88): 3 listUrl='http://news.gzcc.cn/html/xiaoyuanxinwen/{}.html'.format(i) 4 allnews.extend(alist(listUrl)) 5 len(allnews) 6 7 pd.Series(anews) 8 newsdf=pd.DataFrame(allnews) 9 for i in range(5): 10 print(i) 11 time.sleep(random.random()*3) 12 print(newsdf)
5.用pandas做简单的数据处理并保存
保存到csv或excel文件
1 newsdf.to_csv(r'F:hhhh.csv')
保存到数据库
import sqlite3
with sqlite3.connect('gzccnewsdb.sqlite') as db:
newsdf.to_sql('gzccnewsdb',db)
