#coding:utf8
import urllib2,sys,re
import threading,os
import time,datetime
'''''
這里沒有使用隊(duì)列 只是采用多線程分發(fā)對代理量不大的網(wǎng)頁還行但是幾百幾千性能就很差了
'''
def get_proxy_page(url):
'''''解析代理頁面 獲取所有代理地址'''
proxy_list = []
p = re.compile(r'''''<div>(.+?)<span class="Apple-tab-span" style="white-space:pre">.*?</span>(.+?)<span class="Apple-tab-span" style="white-space:pre">.+?</span>(.+?)(<span.+?)?</div>''')
try:
res = urllib2.urlopen(url)
except urllib2.URLError:
print 'url Error'
sys.exit(1)
pageinfo = res.read()
res = p.findall(pageinfo) #取出所有的
#組合成所有代理服務(wù)器列表成一個符合規(guī)則的list
for i in res:
ip = i[0]
port = i[1]
addr = i[2]
l = (ip,port,addr)
proxy_list.append(l)
return proxy_list
#同步鎖裝飾器
lock = threading.Lock()
def synchronous(f):
def call(*args, **kw):
lock.acquire()
try:
return f(*args, **kw)
finally:
lock.release()
return call
#時間計(jì)算器
def sumtime(f):
def call(*args, **kw):
t1 = time.time()
try:
return f(*args, **kw)
finally:
print u'總共用時 %s' % (time.time() - t1)
return call
proxylist = []
reslist = []
#獲取單個代理并處理
@synchronous
def getoneproxy():
global proxylist
if len(proxylist)>0:
return proxylist.pop()
else:
return ''
#添加驗(yàn)證成功的代理
@synchronous
def getreslist(proxy):
global reslist
if not (proxy in reslist):
reslist.append(proxy)
def handle():
timeout = 10
test_url = r'http://www.baidu.com'
test_str = '030173'
while 1:
proxy = getoneproxy()
#最后一個返回是空
if not proxy:
return
print u"正在驗(yàn)證 : %s" %proxy[0]
#第一步啟用 cookie
cookies = urllib2.HTTPCookieProcessor()
proxy_server = r'http://%s:%s' %(proxy[0],proxy[1])
#第二步 裝載代理
proxy_hander = urllib2.ProxyHandler({"http":proxy_server})
#第三步 組合request
try:
opener = urllib2.build_opener(cookies, proxy_hander)
pass
except urllib2.URLError:
print u'url設(shè)置錯誤'
continue
#配置request
opener.addheaders = [('User-Agent','Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.1 (KHTML, like Gecko) Chrome/21.0.1180.89 Safari/537.1')]
#發(fā)送請求
urllib2.install_opener(opener)
t1 = time.time()
try:
req = urllib2.urlopen(test_url,timeout=timeout)
result = req.read()
pos = result.find(test_str)
timeused = time.time() - t1
if pos>1:
#保存到列表中
getreslist((proxy[0],proxy[1],proxy[2],timeused))
print u'成功采集',proxy[0],timeused
else:
continue
except Exception,e:
print u'采集失敗 %s ：timeout' %proxy[0]
continue
def save(reslist):
path = os.getcwd()
filename = path + '/Proxy-'+datetime.datetime.now().strftime(r'%Y%m%d%H%M%S')+'.txt'
f = open(filename,'w+')
for proxy in reslist:
f.write('%s %s %s %s \r\n'%(proxy[0],proxy[1],proxy[2],proxy[3]))
f.close()
@sumtime
def main():
url = r'http://www.free998.net/daili/httpdaili/8949.html'
global proxylist,reslist
#獲取所有線程
proxylist = get_proxy_page(url)
print u'一共獲取 %s 個代理' %len(proxylist)
#print proxylist
print '*'*80
#線程創(chuàng)建和分發(fā)任務(wù)
print u'開始創(chuàng)建線程處理.....'
threads = []
proxy_num = len(proxylist)
for i in range(proxy_num):
th = threading.Thread(target=handle, args=())
threads.append(th)
for thread in threads:
thread.start()
for thread in threads:
threading.Thread.join(thread)
print u'獲取有效代理 %s 個,現(xiàn)在開始排序和保存 '%len(reslist)
reslist = sorted(reslist,cmp=lambda x,y:cmp(x[3],y[3]))
save(reslist)
if __name__=='__main__':
main()