Python が Web ページをクロールして urllib2 経由でシードをダウンロードする例

通过urllib2、re模块抓种子

思路

1.用程序登录论坛(如果需要登录才能访问的版块)

2.访问指定版块

3.遍历帖子（先取指定页，再遍历页面所有帖子的url）

4.循环访问所有帖子url，从帖子页面代码中取种子下载地址（通过正则表达式或第三方页面解析库）

5.访问种子页面下载种子

复制代码 代码如下:

import urllib
import urllib2
import cookielib
import re
import sys
import os

# site is website address | fid is part id
site = "http://xxx.yyy.zzz/"
source = "thread0806.php?fid=x&search=&page="

btSave = "./clyzwm/"
if os.path.isdir(btSave):
 print btSave + " existing"
else:
 os.mkdir(btSave)

logfile = "./clyzwm/down.log"
errorfile = "./clyzwm/error.log"
sucfile = "./clyzwm/sucess.log"

headers = {'User-Agent' : 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_9_1) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/32.0.1700.77 Safari/537.36', 
           'Referer' : 'http://xxx.yyy.zzz/'}

def btDown(url, dirPath):
 logger(logfile, "download file : " + url)
 try:
  #pageCode = urllib2.urlopen(url).read()
  #print pageCode
  btStep1 = re.findall('http://[\w]+\.[\w]+\.[\w]{0,4}/[\w]{2,6}\.php\?[\w]{2,6}=([\w]+)', url, re.I)
  #print btStep1
  if len(btStep1)>0:
   ref = btStep1[0]
   downsite = ""
   downData = {}
   if len(ref)>20:
    downsite = re.findall('http://www.[\w]+\.[\w]+/', url)[0]
    downsite = downsite + "download.php"
    reff = re.findall('input\stype=\"hidden\"\sname=\"reff\"\svalue=\"([\w=]+)\"', urllib2.urlopen(url).read(), re.I)[0]
    downData = {'ref': ref, 'reff':reff, 'submit':'download'}
   else:
    downsite = "http://www.downhh.com/download.php"
    downData = {'ref': ref, 'rulesubmit':'download'}
   #print "bt site - " +  downsite + "\n downData:"
   #print downData
   downData = urllib.urlencode(downData)
   downReq = urllib2.Request(downsite, downData)
   downReq.add_header('User-Agent','Mozilla/5.0 (Macintosh; Intel Mac OS X 10_9_1) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/32.0.1700.77 Safari/537.36')
   downPost = urllib2.urlopen(downReq)
   stream = downPost.read(-1)
   if (len(stream) > 1000):
    downPost.close()
    name = btStep1[0]+ ".torrent"
    fw = open(dirPath + name, 'w')
    fw.write(stream)
    fw.close()
    logger(sucfile, url+"\n")
   else:
    logger(errorfile, url+"\n")
 except urllib2.URLError, e:
  print e.reason

def logger(logfile, msg):
 print msg
 fw = open(logfile, 'a')
 fw.write(msg)
 fw.close()

for i in range(1, 1000):
 logger(logfile, "\n\n\n@ page " + str(i) + " ...")
 part = site + source + str(i)

 content = urllib2.urlopen(part).read()
 content = content.decode('gbk').encode('utf8')
 #print content

 pages = re.findall('', content,re.I)
 #print pages

ページ内のページの場合:
page = site + page;
#logger(logfile, "n# visit " + page + " ...")
pageCode = urllib2.urlopen(page) .read()
#print pageCode
zzJump = re.findall('http://www.viidii.info/?http://[w]+/[w]+?[w]{2 ,6}=[w]+' ,pageCode)
#zzJump = re.findall('http://www.viidii.info/?http://[w/?=]*', pageCode)
if len(zzJump) > 0:
zzJump = zzJump[0]
#print "- ページジャンプ - " + zzJump
pageCode = urllib2.urlopen(page).read()
zzPage = re.findall('http ://[w]+.[w]+.[w]+/link[w]?.php?[w]{2,6}=[w]+' ,pageCode)
if len(zzPage ）＞ 0:
zzPage = zzPage[0]
logger(logfile, "n- zhongzi page -" + zzPage)
btDown(zzPage, btSave)
else:
logger(logfile, " n. 見つかりません。")
else:
logger(logfile, "n... 見つかりません ...")
zzPage = re.findall('http://[w]+. [w]+.[w]+/link[w]?.php?ref=[w]+' ,pageCode)