我搜集了国内10几个电影网站的数据,里面近几十W条记录,用文本没法存,mongodb学习成本非常低,安装、下载、运行起来不会花你5分钟时间。

# -*- coding: utf-8 -*-
# by awakenjoys. my site: www.dianying.at
import re
import urllib2
from bs4 import BeautifulSoup
import string, time
import pymongo
 
NUM  = 0   #全局变量,电影数量
m_type = u''  #全局变量,电影类型
m_site = u'qq' #全局变量,电影网站
 
#根据指定的URL获取网页内容
def gethtml(url):
 req = urllib2.Request(url) 
 response = urllib2.urlopen(req) 
 html = response.read()
 return html
 
#从电影分类列表页面获取电影分类
def gettags(html):
 global m_type
 soup = BeautifulSoup(html)  #过滤出分类内容
 #print soup
 #<ul class="clearfix _group" gname="mi_type" gtype="1">
 tags_all = soup.find_all('ul', {'class' : 'clearfix _group' , 'gname' : 'mi_type'})
 #print len(tags_all), tags_all
 #print str(tags_all[1]).replace('\n', '')
 
 #<a _hot="tag.sub" class="_gtag _hotkey" href="http://v.qq.com/list/1_0_-1_-1_1_0_0_20_0_-1_0.html" title="动作" tvalue="0">动作</a>
 re_tags = r'<a _hot=\"tag\.sub\" class=\"_gtag _hotkey\" href=\"(.+" title=\"(.+" tvalue=\"(.+">.+"Not Find"
 return tags_url
 
#获取每个分类的页数
def get_pages(tag_url):
 tag_html = gethtml(tag_url)
 #div class="paginator
 soup = BeautifulSoup(tag_html)  #过滤出标记页面的html
 #print soup
 #<div class="mod_pagenav" id="pager">
 div_page = soup.find_all('div', {'class' : 'mod_pagenav', 'id' : 'pager'})
 #print div_page #len(div_page), div_page[0]
 
 #<a class="c_txt6" href="http://v.qq.com/list/1_2_-1_-1_1_0_24_20_0_-1_0.html" title="25"><span>25</span></a>
 re_pages = r'<a class=.+?><span>(.+"mod_list_pic_130">
 divs = soup.find_all('ul', {'class' : 'mod_list_pic_130'})
 #print divs
 for div_html in divs:
  div_html = str(div_html).replace('\n', '')
  #print div_html
  getmovie(div_html)
 
 
def getmovie(html):
 global NUM
 global m_type
 global m_site
 
 #<h6 class="caption"> <a href="http://www.tudou.com/albumcover/Z7eF_40EL4I.html" target="_blank" title="徒步旅行队">徒步旅行队</a> </h6> <ul class="info"> <li class="desc">法国卖座喜剧片</li> <li class="cast"> </li> </ul> </div> <div class="ext ext_last"> <div class="ext_txt"> <h3 class="ext_title">徒步旅行队</h3> <div class="ext_info"> <span class="ext_area">地区: 法国</span> <span class="ext_cast">导演: </span> <span class="ext_date">年代: 2009</span> <span class="ext_type">类型: 喜剧</span> </div> <p class="ext_intro">理查德·达奇拥有一家小的旅游公司,主要经营法国游客到非洲大草原的旅游服务。六个法国游客决定参加理查德·达奇组织的到非洲的一...</p>
 
 re_movie = r'<li><a class=\"mod_poster_130\" href=\"(.+" target=\"_blank\" title=\"(.+"><img.+"%s : %d" % ("=" * 70, NUM)
   values = dict(
    movie_title = movie[1],
    movie_url = movie[0],
    movie_site  = m_site,
    movie_type  = m_type
    )
   print values
   playlinks.insert(values)
   print "_" * 70
   NUM += 1
   print "%s : %d" % ("=" * 70, NUM)
 
 #else:
 # print "Not Find"
 
def getmovieinfo(url):
 html = gethtml(url)
 soup = BeautifulSoup(html)
 
 #pack pack_album album_cover
 divs = soup.find_all('div', {'class' : 'pack pack_album album_cover'})
 #print divs[0]
 
 #<a href="http://www.tudou.com/albumplay/9NyofXc_lHI/32JqhiKJykI.html" target="new" title="《血滴子》独家纪录片" wl="1"> </a> 
 re_info = r'<a href=\"(.+" target=\"new\" title=\"(.+" wl=\".+"> </a>'
 p_info = re.compile(re_info, re.DOTALL)
 m_info = p_info.findall(str(divs[0]))
 if m_info:
  return m_info
 else:
  print "Not find movie info"
 
 return m_info
 
 
def insertdb(movieinfo):
 global conn
 movie_db = conn.dianying_at
 movies = movie_db.movies
 movies.insert(movieinfo)
 
if __name__ == "__main__":
 global conn
 
 tags_url = "http://v.qq.com/list/1_-1_-1_-1_1_0_0_20_0_-1_0.html"
 #print tags_url
 tags_html = gethtml(tags_url)
 #print tags_html
 tag_urls = gettags(tags_html)
 #print tag_urls
 
 
 for url in tag_urls.items():
  print str(url[1]).encode('utf-8') #,url[0]
  maxpage = int(get_pages(str(url[1]).encode('utf-8')))
  print maxpage
 
  for x in range(0, maxpage):
   #http://v.qq.com/list/1_0_-1_-1_1_0_0_20_0_-1_0.html
   m_url = str(url[1]).replace('0_20_0_-1_0.html', '')
   movie_url = "%s%d_20_0_-1_0.html" % (m_url, x)
   print movie_url
   movie_html = gethtml(movie_url.encode('utf-8'))
   #print movie_html
   getmovielist(movie_html)
   time.sleep(0.1)

标签:
腾讯视频,所有电影,python爬虫

免责声明:本站文章均来自网站采集或用户投稿,网站不提供任何软件下载或自行开发的软件! 如有用户或公司发现本站内容信息存在侵权行为,请邮件告知! 858582#qq.com