# coding:utf-8
'''功能:采集百度新闻(http://news.baidu.com/)内容,百度新闻聚合了许多行业网站的新闻,已经帮我们去重筛选了,采集自己行业的新闻数据很不错。
主要思路:1,利用字典把各个网站的网址与正则及网页编码对应起来 2,把采集过得url放到一个文件中,判断是否采集过 3,百度新闻5分钟跟新一次,可以再建个程序每隔几分钟运行一次
'''
import pycurl,StringIO,json,urllib,urllib2,re
import MySQLdb
import time
from warnings import filterwarnings
import MySQLdb as Database
filterwarnings('ignore', category = Database.Warning)
import sys
reload(sys)
sys.setdefaultencoding('utf8')
headers = [
"User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/49.0.2623.112 Safari/537.36",
"Cookie: spversion=20130314; historystock=603158%7C*%7C1A0001%7C*%7C000967%7C*%7C603328; Hm_lvt_78c58f01938e4d85eaf619eae71b4ed1=1467682875,1467682943,1467682974,1468293176; Hm_lpvt_78c58f01938e4d85eaf619eae71b4ed1=1468293226",
]
def curl(url):
c = pycurl.Curl() #通过curl方法构造一个对象
#c.setopt(pycurl.REFERER, 'http://qy.m.58.com/') #设置referer
c.setopt(pycurl.FOLLOWLOCATION, True) #自动进行跳转抓取
c.setopt(pycurl.MAXREDIRS,5) #设置最多跳转多少次
c.setopt(pycurl.CONNECTTIMEOUT, 60) #设置链接超时
c.setopt(pycurl.TIMEOUT,120) #下载超时
c.setopt(pycurl.ENCODING, 'gzip,deflate')
# c.setopt(c.PROXY,ip) # 代理
c.fp = StringIO.StringIO()
c.setopt(pycurl.URL, url) #设置要访问的URL
c.setopt(pycurl.HTTPHEADER,headers) #传入请求头
# c.setopt(pycurl.POST, 1)
# c.setopt(pycurl.POSTFIELDS, data) #传入POST数据
c.setopt(c.WRITEFUNCTION, c.fp.write) #回调写入字符串缓存
c.perform()
code = c.getinfo(c.HTTP_CODE) #返回状态码
html = c.fp.getvalue() #返回源代码
return html
# 通过正则提取元素
def search(req,html):
text = re.search(req,html)
if text:
data = text.group(1)
else:
data = 'no'
return data
# 去除文章url、多余标签等、补全路径等
def content_sort(content):
content = re.sub('<p.*?>','<p>',content,flags=re.I)
content = re.sub('<P.*?>','<p>',content)
content = re.sub('</?span.*?>','',content)
content = re.sub('</?a.*?>','',content)
content = re.sub('<!.*?>','',content)
content = re.sub('</?img.*?>','',content,re.IGNORECASE)
content = re.sub('</?IMG.*?>','',content,re.IGNORECASE)
content = re.sub('</?div.*?>','',content,flags=re.I)
content = re.sub('</?DIV.*?>','',content)
content = re.sub('</?iframe.*?>','',content)
content = re.sub('</?center.*?>','',content)
content = re.sub('</?[fF].*?>','',content)
content = re.sub('<script.*?>[\s\S]*?</script>','',content)
content = re.sub('</?strong.*?>','',content)
content = re.sub('<INPUT.*?>','',content)
content = re.sub('<style.*?>[\s\S]*?</style>','',content)
content = re.sub(' ','',content)
content = re.sub(' ','',content)
content = re.sub(' ','',content)
return content
#域名与正则、编码对应表
req_dict = {
'finance.sina.com.cn':
{'title':'<h1.*?>(.*?)</h1>','content':'<!-- 原始正文start -->([\s\S]*?)<!-- 原始正文end -->','decode':'utf-8'},
'stock.eastmoney.com':
{'title':'<h1.*?>(.*?)</h1>','content':'<div id="ContentBody" class="Body">([\s\S]*?)<div class="BodyEnd">','decode':'gbk'},
'finance.eastmoney.com':
{'title':'<h1.*?>(.*?)</h1>','content':'<div id="ContentBody" class="Body">([\s\S]*?)<div class="BodyEnd">','decode':'gbk'},#ok
'guba.eastmoney.com':
{'title':'<title>(.*?)_.*?</title>','content':'<div id="zwconbody">([\s\S]*?)<div class="zwconbtns clearfix">','decode':'utf-8'},#ok
'stock.jrj.com.cn':
{'title':'<title>(.*?)-','content':'<div class="texttit_m1">([\s\S]*?)<div id="itougu">','decode':'gbk'},
'hk.jrj.com.cn':
{'title':'<title>(.*?)-','content':'<div class="texttit_m1">([\s\S]*?)<div id="itougu">','decode':'gbk'},
'hkstock.cnfol.com':
{'title':'<title>(.*?)_.*?</title>','content':'<div class="ArtM" id="Content">([\s\S]*?)<!--正文结束-->','decode':'utf-8'},#ok
'sc.stock.cnfol.com':
{'title':'<title>(.*?)_.*?</title>','content':'<div class="ArtM" id="Content">([\s\S]*?)<!--正文结束-->','decode':'utf-8'},#ok
'money.163.com':
{'title':'<title>(.*?)_.*?</title>','content':'<div class="post_text".*?">([\s\S]*?)<!--.*?s -->','decode':'utf-8'},
'www.chinastock.com.cn':
{'title':'<div class="d_title">([\s\S]*?)</div>','content':'<div class="d_content" id="Zoom">([\s\S]*?)<div class="dleft_new_attachment">','decode':'utf-8'},
'stock.huagu.com':
{'title':'<h1 id="h1-title">([\s\S]*?)</h1>','content':'<div class="article_con" id="div-article-content">([\s\S]*?)<div class="clear"></div>','decode':'utf-8'},
'stock.sohu.com':
{'title':'<h1 itemprop="headline">([\s\S]*?)</h1>','content':'<div itemprop="articleBody">([\s\S]*?)<div class="original-title"','decode':'gbk'},
'stock.cngold.org':
{'title':'<title>(.*?)-.*?</title>','content':'<div class="det_content" id="zoom">([\s\S]*?)<div class="listPage">','decode':'utf-8'},
'hk.stock.hexun.com':
{'title':'<title>(.*?)[-_|].*?</title>','content':'<div class="art_contextBox">([\s\S]*?)<div class="showAll">','decode':'utf-8'},
'stock.gucheng.com':
{'title':'<title>(.*?)[-_|].*?</title>','content':'<div class="content">([\s\S]*?)</div>','decode':'utf-8'},
'www.cnstock.com':
{'title':'<title>(.*?)-.*?</title>','content':'<div class="content-inner" id="qmt_content_div">([\s\S]*?)</div>','decode':'gbk'},
'www.ccstock.cn':
{'title':'<title>(.*?)-.*?</title>','content':'<div id="newscontent">([\s\S]*?)</div>','decode':'utf-8'},
'news.emoney.cn':
{'title':'<title>(.*?)-.*?</title>','content':'<div class="RL_details_content">([\s\S]*?)<div class="PageNav">','decode':'utf-8'},
'finance.ce.cn':
{'title':'<title>(.*?)</title>','content':'<div class=TRS_Editor>([\s\S]*?)<textarea id="allinfo"','decode':'gbk'},
'www.p5w.net':
{'title':'<title>(.*?)[_-|].*?</title>','content':'<div class="text">([\s\S]*?)<div class="pages">','decode':'gbk'},
'www.nbd.com.cn':
{'title':'<title>(.*?)[_-|][\s\S]*?</title>','content':'<div class="main-left-article">([\s\S]*?)<div style="overflow:','decode':'utf-8'},
'stock.hexun.com':
{'title':'<title>(.*?)[-_|].*?</title>','content':'<div class="art_contextBox">([\s\S]*?)<div class="showAll">','decode':'gbk'},
'stock.caijing.com.cn':
{'title':'<title>(.*?)[-_|].*?</title>','content':'<div id="the_content".*?>([\s\S]*?)<div class="ar_writer"','decode':'utf-8'},
}
def id():
'''获取标题对应id,构建url.我用的是phpcms,前台显示需将url写入数据库'''
con = MySQLdb.connect('localhost','root','','phpcmsv9',charset='utf8')
with con:
cur = con.cursor()
cur.execute("select id from v9_news where title = title")
numrows = int(cur.rowcount)
return numrows+1
def CmsSQL(title,content):
'''写入数据,如何将多个数据写入数据库可参考'''
value1 = []
value1.append(content)
value1.append(idnum)
value2 = []
value2.append(title)
value2.append(urlid)
value2.append(int(time.time()))
value2.append(int(time.time()))
db = MySQLdb.connect('localhost','root','','phpcmsv9',charset='utf8')
cursor = db.cursor()
cursor.execute("insert into v9_news_data (content,id) values(%s,%s)" ,value1)
cursor.execute("insert into v9_news(title,catid,typeid,url,inputtime,updatetime) values(%s,6,0,%s,%s,%s)",value2)
db.commit()
db.close()
url = 'http://news.baidu.com/n?cmd=4&class=stock&tn=rss'
urls = re.findall(r'<link><!\[CDATA\[(.*?)\]\]></link>',curl(url))
urls.reverse()
for url in urls:
with open('urls.txt') as f1 :
if url not in f1.read(): #判断url是否采集过
url.strip()
f1.close()
line = url.split('/')[2]
if req_dict.has_key(line): #通过键位是否存在判断这个网站是否写好的正则
time.sleep(1)
try:
title = search(req_dict[line]['title'],curl(url)).decode(req_dict[line]['decode']) #网址与正则及网页编码对应起来
content = url + search(req_dict[line]['content'],curl(url)).decode(req_dict[line]['decode'])
except:
continue
urlid = 'http://localhost/index.php?m=content&c=index&a=show&catid=6&id=%s' %id()
idnum = int(id())
print id(),content_sort(title)
CmsSQL(content_sort(title),content_sort(content))
f1w =open('urls.txt','a+')
f1w.write(url+'\n')
f1w.close()
else:
print u'正则不存在'
open('requrl','a+').write(url+'\n')
else:
print u'此url在列表中:'