python爬蟲抓取騰訊影片所有電影（程式碼）

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

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(&#39;ul&#39;, {&#39;class&#39; : &#39;clearfix _group&#39; , &#39;gname&#39; : &#39;mi_type&#39;})

    #print len(tags_all), tags_all

    #print str(tags_all[1]).replace(&#39;\n&#39;, &#39;&#39;)

    #<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&#39;<a _hot=\"tag\.sub\" class=\"_gtag _hotkey\" href=\"(.+?)\" title=\"(.+?)\" tvalue=\"(.+?)\">.+?</a>&#39;

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

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

    if tags:

        tags_url = {}

        #print tags

        for tag in tags:

            tag_url = tag[0].decode(&#39;utf-8&#39;)

            #print tag_url

            m_type = tag[1].decode(&#39;utf-8&#39;)

            tags_url[m_type] = tag_url
    else:

            print "Not Find"

    return tags_url

#获取每个分类的页数

def get_pages(tag_url):

    tag_html = gethtml(tag_url)

    #p class="paginator

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

    #print soup

    #<p class="mod_pagenav" id="pager">

    p_page = soup.find_all(&#39;p&#39;, {&#39;class&#39; : &#39;mod_pagenav&#39;, &#39;id&#39; : &#39;pager&#39;})

    #print p_page #len(p_page), p_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&#39;<a class=.+?><span>(.+?)</span></a>&#39;

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

    pages = p.findall(str(p_page[0]))

    #print pages

    if len(pages) > 1:

        return pages[-2]

    else:

        return 1

def getmovielist(html):

    soup = BeautifulSoup(html)

    #<ul class="mod_list_pic_130">

    ps = soup.find_all(&#39;ul&#39;, {&#39;class&#39; : &#39;mod_list_pic_130&#39;})

    #print ps

    for p_html in ps:

        p_html = str(p_html).replace(&#39;\n&#39;, &#39;&#39;)

        #print p_html

        getmovie(p_html)

def getmovie(html):

    global NUM

    global m_type

    global m_site
    re_movie = r&#39;<li><a class=\"mod_poster_130\" href=\"(.+?)\" target=\"_blank\" title=\"(.+?)\"><img.+?</li>&#39;

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

    movies = p.findall(html)

    if movies:

        conn = pymongo.Connection(&#39;localhost&#39;, 27017)

        movie_db = conn.dianying

        playlinks = movie_db.playlinks

        #print movies

        for movie in movies:

            #print movie

            NUM += 1

            print "%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

    ps = soup.find_all(&#39;p&#39;, {&#39;class&#39; : &#39;pack pack_album album_cover&#39;})

    #print ps[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&#39;<a href=\"(.+?)\" target=\"new\" title=\"(.+?)\" wl=\".+?\"> </a>&#39;

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

    m_info = p_info.findall(str(ps[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(&#39;utf-8&#39;) #,url[0]

        maxpage = int(get_pages(str(url[1]).encode(&#39;utf-8&#39;)))

        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(&#39;0_20_0_-1_0.html&#39;, &#39;&#39;)

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

            print movie_url

            movie_html = gethtml(movie_url.encode(&#39;utf-8&#39;))

            #print movie_html

            getmovielist(movie_html)

            time.sleep(0.1)