用python实现的抓取腾讯视频所有电影的爬虫

1. [代码]用python实现的抓取腾讯视频所有电影的爬虫     跳至 [1] [全屏预览]

001	# -*- coding: utf-8 -*-

002	# by awakenjoys. my site: www.dianying.at

003

import re

004	import urllib2

005	from bs4 import BeautifulSoup

006	import string, time

007	import pymongo

008

009	NUM     = 0         #全局变量,电影数量

010	m_type  = u''       #全局变量,电影类型

011	m_site  = u'qq' #全局变量,电影网站

012

013	#根据指定的URL获取网页内容

014	def gethtml(url):

015	req = urllib2.Request(url)

016	response = urllib2.urlopen(req)

017	html = response.read()

018	return html

019

020	#从电影分类列表页面获取电影分类

021	def gettags(html):

022	global m_type

023	soup = BeautifulSoup(html)      #过滤出分类内容

024	#print soup

025	#<ul class="clearfix _group" gname="mi_type" gtype="1">

026	tags_all = soup.find_all('ul', {'class' : 'clearfix _group' , 'gname' : 'mi_type'})

027	#print len(tags_all), tags_all

028	#print str(tags_all[1]).replace('\n', '')

029

030	#<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>

031	re_tags = r'<a _hot=\"tag\.sub\" class=\"_gtag _hotkey\" href=\"(.+?)\" title=\"(.+?)\" tvalue=\"(.+?)\">.+?</a>'

032	p = re.compile(re_tags, re.DOTALL)

033

034	tags = p.findall(str(tags_all[0]))

035

if tags:

036	tags_url = {}

037	#print tags

038	for tag in tags:

039	tag_url = tag[0].decode('utf-8')

040	#print tag_url

041	m_type = tag[1].decode('utf-8')

042	tags_url[m_type] = tag_url

043

044

else:

045	print "Not Find"

046	return tags_url

047

048	#获取每个分类的页数

049	def get_pages(tag_url):

050	tag_html = gethtml(tag_url)

051	#div class="paginator

052	soup = BeautifulSoup(tag_html)      #过滤出标记页面的html

053	#print soup

054	#<div class="mod_pagenav" id="pager">

055	div_page = soup.find_all('div', {'class' : 'mod_pagenav', 'id' : 'pager'})

056	#print div_page #len(div_page), div_page[0]

057

058	#<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>

059	re_pages = r'<a class=.+?><span>(.+?)</span></a>'

060	p = re.compile(re_pages, re.DOTALL)

061	pages = p.findall(str(div_page[0]))

062	#print pages

063	if len(pages) > 1:

064	return pages[-2]

065

else:

066

return 1

067

068

069	def getmovielist(html):

070	soup = BeautifulSoup(html)

071

072	#<ul class="mod_list_pic_130">

073	divs = soup.find_all('ul', {'class' : 'mod_list_pic_130'})

074	#print divs

075	for div_html in divs:

076	div_html = str(div_html).replace('\n', '')

077	#print div_html

078	getmovie(div_html)

079

080

081	def getmovie(html):

082	global NUM

083	global m_type

084	global m_site

085

086

#<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>

087

088	re_movie = r'<li><a class=\"mod_poster_130\" href=\"(.+?)\" target=\"_blank\" title=\"(.+?)\"><img.+?</li>'

089	p = re.compile(re_movie, re.DOTALL)

090	movies = p.findall(html)

091	if movies:

092	conn = pymongo.Connection('localhost', 27017)

093	movie_db = conn.dianying

094	playlinks = movie_db.playlinks

095	#print movies

096	for movie in movies:

097	#print movie

098

NUM += 1

099	print "%s : %d" % ("=" * 70, NUM)

100	values = dict(

101	movie_title = movie[1],

102	movie_url   = movie[0],

103	movie_site      = m_site,

104	movie_type      = m_type

105

)

106	print values

107	playlinks.insert(values)

108	print "_" * 70

109

NUM += 1

110	print "%s : %d" % ("=" * 70, NUM)

111

112

#else:

113	#   print "Not Find"

114

115	def getmovieinfo(url):

116	html = gethtml(url)

117	soup = BeautifulSoup(html)

118

119	#pack pack_album album_cover

120	divs = soup.find_all('div', {'class' : 'pack pack_album album_cover'})

121	#print divs[0]

122

123	#<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>

124	re_info = r'<a href=\"(.+?)\" target=\"new\" title=\"(.+?)\" wl=\".+?\"> </a>'

125	p_info = re.compile(re_info, re.DOTALL)

126	m_info = p_info.findall(str(divs[0]))

127	if m_info:

128	return m_info

129

else:

130	print "Not find movie info"

131

132	return m_info

133

134

135	def insertdb(movieinfo):

136	global conn

137	movie_db = conn.dianying_at

138	movies = movie_db.movies

139	movies.insert(movieinfo)

140

141	if __name__ == "__main__":

142	global conn

143

144	tags_url = "http://v.qq.com/list/1_-1_-1_-1_1_0_0_20_0_-1_0.html"

145	#print tags_url

146	tags_html = gethtml(tags_url)

147	#print tags_html

148	tag_urls = gettags(tags_html)

149	#print tag_urls

150

151

152	for url in tag_urls.items():

153	print  str(url[1]).encode('utf-8') #,url[0]

154	maxpage = int(get_pages(str(url[1]).encode('utf-8')))

155	print maxpage

156

157	for x in range(0, maxpage):

158	#http://v.qq.com/list/1_0_-1_-1_1_0_0_20_0_-1_0.html

159	m_url = str(url[1]).replace('0_20_0_-1_0.html', '')

160	movie_url = "%s%d_20_0_-1_0.html" % (m_url, x)

161	print movie_url

162	movie_html = gethtml(movie_url.encode('utf-8'))

163	#print movie_html

164	getmovielist(movie_html)

165	time.sleep(0.1)