数据结构化与保存
一、结构化
1、单条新闻的详情字典：news
import requests from bs4 import BeautifulSoup from datetime import datetime import re import pandas import sqlite3 def getclick(url): m = re.search(r'_(.*).html',url) newsid = m.group(1)[5:] clickurl = 'http://oa.gzcc.cn/api.php?op=count&id={}&modelid=80'.format(newsid) resc = requests.get(clickurl).text r = re.search(r"hits(.*)",resc).group(1) click = r.lstrip("').html('").rstrip("');") return(int(click)) def getdetail(url): resd = requests.get(url) resd.encoding='ute-8' soupd = BeautifulSoup(resd.text,'html.parser') news = {} news['url'] = url news['title'] = soupd.select('.show-title')[0].text info = soupd.select('.show-info')[0].text #news['dt'] = datetime.strptime(info.lstrip('发布时间:')[0:19],'%Y-%m-%d %H:%M') #news['source'] = re.search('来源:(.*)点击',info).group(1).strip() news['content'] = soupd.select('.show-content')[0].text.strip() news['click']=getclick(url) return(news) def onepage(pageurl): res = requests.get(pageurl) res.encoding = 'utf-8' soup = BeautifulSoup(res.text,'html.parser') newsls = [] for news in soup.select('li'): if len(news.select('.news-list-title'))>0: newsls.append(getdetail(news.select('a')[0]['href'])) break return(newsls) print(onepage('http://news.gzcc.cn/html/xiaoyuanxinwen/'))
2、一个列表页所有单条新闻汇总列表：newsls.append(news)
import requests from bs4 import BeautifulSoup from datetime import datetime import re import pandas import sqlite3 def getclick(url): m = re.search(r'_(.*).html',url) newsid = m.group(1)[5:] clickurl = 'http://oa.gzcc.cn/api.php?op=count&id={}&modelid=80'.format(newsid) resc = requests.get(clickurl).text r = re.search(r"hits(.*)",resc).group(1) click = r.lstrip("').html('").rstrip("');") return(int(click)) def getdetail(url): resd = requests.get(url) resd.encoding='ute-8' soupd = BeautifulSoup(resd.text,'html.parser') news = {} news['url'] = url news['title'] = soupd.select('.show-title')[0].text info = soupd.select('.show-info')[0].text #news['dt'] = datetime.strptime(info.lstrip('发布时间:')[0:19],'%Y-%m-%d %H:%M') #news['source'] = re.search('来源:(.*)点击',info).group(1).strip() news['content'] = soupd.select('.show-content')[0].text.strip() news['click']=getclick(url) return(news) def onepage(pageurl): res = requests.get(pageurl) res.encoding = 'utf-8' soup = BeautifulSoup(res.text,'html.parser') newsls = [] for news in soup.select('li'): if len(news.select('.news-list-title'))>0: newsls.append(getdetail(news.select('a')[0]['href'])) return(newsls) print(onepage('http://news.gzcc.cn/html/xiaoyuanxinwen/'))
3、所有列表页的所有新闻汇总列表：newstotal.extend(newsls)
import requests from bs4 import BeautifulSoup from datetime import datetime import re import pandas import sqlite3 def getclick(url): m = re.search(r'_(.*).html',url) newsid = m.group(1)[5:] clickurl = 'http://oa.gzcc.cn/api.php?op=count&id={}&modelid=80'.format(newsid) resc = requests.get(clickurl).text r = re.search(r"hits(.*)",resc).group(1) click = r.lstrip("').html('").rstrip("');") return(int(click)) def getdetail(url): resd = requests.get(url) resd.encoding='ute-8' soupd = BeautifulSoup(resd.text,'html.parser') news = {} news['路径'] = url news['标题'] = soupd.select('.show-title')[0].text info = soupd.select('.show-info')[0].text #news['dt'] = datetime.strptime(info.lstrip('发布时间:')[0:19],'%Y-%m-%d %H:%M') #news['source'] = re.search('来源:(.*)点击',info).group(1).strip() #news['content'] = soupd.select('.show-content')[0].text.strip() news['点击数']=getclick(url) return(news) def onepage(pageurl): res = requests.get(pageurl) res.encoding = 'utf-8' soup = BeautifulSoup(res.text,'html.parser') newsls = [] for news in soup.select('li'): if len(news.select('.news-list-title'))>0: newsls.append(getdetail(news.select('a')[0]['href'])) return(newsls) newstotal = [] gzccurl = 'http://news.gzcc.cn/html/xiaoyuanxinwen/' newstotal.extend(onepage(gzccurl)) res = requests.get(gzccurl) res.encoding='utf-8' soup = BeautifulSoup(res.text,'html.parser') n = int(soup.select('.a1')[0].text.rstrip('条')) pages=n//10+1 for i in range(2,3): listurl='http://news.gzcc.cn/html/xiaoyuanxinwen/{}.html'.format(i) newstotal.extend(onepage(listurl)) print(newstotal)
二、转换成pandas的数据结构DataFrame
import requests from bs4 import BeautifulSoup from datetime import datetime import re import pandas import sqlite3 def getclick(url): m = re.search(r'_(.*).html',url) newsid = m.group(1)[5:] clickurl = 'http://oa.gzcc.cn/api.php?op=count&id={}&modelid=80'.format(newsid) resc = requests.get(clickurl).text r = re.search(r"hits(.*)",resc).group(1) click = r.lstrip("').html('").rstrip("');") return(int(click)) def getdetail(url): resd = requests.get(url) resd.encoding='ute-8' soupd = BeautifulSoup(resd.text,'html.parser') news = {} news['路径'] = url news['标题'] = soupd.select('.show-title')[0].text info = soupd.select('.show-info')[0].text #news['dt'] = datetime.strptime(info.lstrip('发布时间:')[0:19],'%Y-%m-%d %H:%M') #news['source'] = re.search('来源:(.*)点击',info).group(1).strip() #news['content'] = soupd.select('.show-content')[0].text.strip() news['点击数']=getclick(url) return(news) def onepage(pageurl): res = requests.get(pageurl) res.encoding = 'utf-8' soup = BeautifulSoup(res.text,'html.parser') newsls = [] for news in soup.select('li'): if len(news.select('.news-list-title'))>0: newsls.append(getdetail(news.select('a')[0]['href'])) return(newsls) newstotal = [] gzccurl = 'http://news.gzcc.cn/html/xiaoyuanxinwen/' newstotal.extend(onepage(gzccurl)) res = requests.get(gzccurl) res.encoding='utf-8' soup = BeautifulSoup(res.text,'html.parser') n = int(soup.select('.a1')[0].text.rstrip('条')) pages=n//10+1 for i in range(2,3): listurl='http://news.gzcc.cn/html/xiaoyuanxinwen/{}.html'.format(i) newstotal.extend(onepage(listurl)) df = pandas.DataFrame(newstotal) print(df.head())
三、从DataFrame保存到excel
import requests from bs4 import BeautifulSoup from datetime import datetime import re import pandas import sqlite3 def getclick(url): m = re.search(r'_(.*).html',url) newsid = m.group(1)[5:] clickurl = 'http://oa.gzcc.cn/api.php?op=count&id={}&modelid=80'.format(newsid) resc = requests.get(clickurl).text r = re.search(r"hits(.*)",resc).group(1) click = r.lstrip("').html('").rstrip("');") return(int(click)) def getdetail(url): resd = requests.get(url) resd.encoding='ute-8' soupd = BeautifulSoup(resd.text,'html.parser') news = {} news['路径'] = url news['标题'] = soupd.select('.show-title')[0].text info = soupd.select('.show-info')[0].text #news['dt'] = datetime.strptime(info.lstrip('发布时间:')[0:19],'%Y-%m-%d %H:%M') #news['source'] = re.search('来源:(.*)点击',info).group(1).strip() news['内容'] = soupd.select('.show-content')[0].text.strip() news['点击数']=getclick(url) return(news) def onepage(pageurl): res = requests.get(pageurl) res.encoding = 'utf-8' soup = BeautifulSoup(res.text,'html.parser') newsls = [] for news in soup.select('li'): if len(news.select('.news-list-title'))>0: newsls.append(getdetail(news.select('a')[0]['href'])) return(newsls) newstotal = [] gzccurl = 'http://news.gzcc.cn/html/xiaoyuanxinwen/' newstotal.extend(onepage(gzccurl)) res = requests.get(gzccurl) res.encoding='utf-8' soup = BeautifulSoup(res.text,'html.parser') n = int(soup.select('.a1')[0].text.rstrip('条')) pages=n//10+1 for i in range(2,3): listurl='http://news.gzcc.cn/html/xiaoyuanxinwen/{}.html'.format(i) newstotal.extend(onepage(listurl)) df = pandas.DataFrame(newstotal) df.to_excel('gzccnews.xlsx')
四、从DataFrame保存到sqlite3数据库
import requests from bs4 import BeautifulSoup from datetime import datetime import re import pandas import sqlite3 def getclick(url): m = re.search(r'_(.*).html',url) newsid = m.group(1)[5:] clickurl = 'http://oa.gzcc.cn/api.php?op=count&id={}&modelid=80'.format(newsid) resc = requests.get(clickurl).text r = re.search(r"hits(.*)",resc).group(1) click = r.lstrip("').html('").rstrip("');") return(int(click)) def getdetail(url): resd = requests.get(url) resd.encoding='ute-8' soupd = BeautifulSoup(resd.text,'html.parser') news = {} news['路径'] = url news['标题'] = soupd.select('.show-title')[0].text #info = soupd.select('.show-info')[0].text #news['dt'] = datetime.strptime(info.lstrip('发布时间:')[0:19],'%Y-%m-%d %H:%M') #news['source'] = re.search('来源:(.*)点击',info).group(1).strip() info = soupd.select('.show-info')[0].text.split('\xa0\xa0') news['发布时间'] = info[0].lstrip('发布时间:').strip() #news['dt'] = datetime.strptime(info[0].lstrip('发布时间:'),'%Y-%m-%d %H:%M:%S') news['来源'] = info[4].lstrip('来源：').strip() news['内容'] = soupd.select('.show-content')[0].text.strip() news['点击数']=getclick(url) return(news) def onepage(pageurl): res = requests.get(pageurl) res.encoding = 'utf-8' soup = BeautifulSoup(res.text,'html.parser') newsls = [] for news in soup.select('li'): if len(news.select('.news-list-title'))>0: newsls.append(getdetail(news.select('a')[0]['href'])) return(newsls) newstotal = [] gzccurl = 'http://news.gzcc.cn/html/xiaoyuanxinwen/' newstotal.extend(onepage(gzccurl)) res = requests.get(gzccurl) res.encoding='utf-8' soup = BeautifulSoup(res.text,'html.parser') n = int(soup.select('.a1')[0].text.rstrip('条')) pages=n//10+1 for i in range(2,3): listurl='http://news.gzcc.cn/html/xiaoyuanxinwen/{}.html'.format(i) newstotal.extend(onepage(listurl)) df = pandas.DataFrame(newstotal) with sqlite3.connect('gzccnews.sqlite') as db: df.to_sql('gzccnews',con=db)
