最近写了一个爬取代{过}{滤}理的python脚本,参考了一下别人的代码,有了大量代{过}{滤}理就方便了,这是v1.0版本,采用了多线程(一页用一个线程爬代{过}{滤}理),顺便熟悉一下队列和bs4,感觉bs4的确很强大而且方便很多。还有很多地方不是很完善,日后有空会继续完善这个脚本。
功能描述:爬取www.xicIDAili.com的代{过}{滤}理,并去1212.ip138.com/ic.asp验证代{过}{滤}理的可用性,最后把可用代{过}{滤}理写入useful_proxies.txt文件。
代码不是很长我就不打包了,大家有需要就直接复制代码吧,欢迎反馈bug和和提出建议!
源代码:
[Python] 纯文本查看 复制代码 #coding:utf-8
import requests
from bs4 import BeautifulSoup as bs
import re
import Queue
import threading
import time
import optparse
url = 'http://www.xicidaili.com/nn/'
headers = {'User-Agent':'Mozilla/5.0 (Windows NT 10.0; WOW64; Trident/7.0; .NET4.0C; .NET4.0E; .NET CLR 2.0.50727; .NET CLR 3.0.30729; .NET CLR 3.5.30729; InfoPath.3; rv:11.0) like Gecko'}
class Proxy_collection(threading.Thread): #继承Thread实现多线程
def __init__(self, que):
threading.Thread.__init__(self) #重用父类Thread的__init__()
self._que = que
def run(self):
while not self._que.empty():
url = self._que.get()
r = requests.get(url, headers=headers, timeout=5)
soup = bs(r.content, 'lxml', from_encoding='utf-8')
bqs = soup.find_all(name='tr', attrs={'class':re.compile(r'|[^odd]')})
for bq in bqs:
us = bq.find_all(name='td')
try:
self.proxies_confirm(str(us[5].string), str(us[1].string), str(us[2].string)) #取协议:ip:端口去验证
except Exception,e:
#print e
pass
def proxies_confirm(self, type_self, ip, port):
ip_dic = {}
ip_dic[type_self.lower()] = ip + ':' + port
r = requests.get('http://1212.ip138.com/ic.asp', headers=headers, proxies=ip_dic, timeout=5)
result = re.findall(r'\d+\.\d+\.\d+\.\d+', r.content)
result_ip = ''.join(result) #转为字符串
if ip == result_ip:
print type_self + '---' + ip + ':' + port + ' is useful!!!\n'
with open('useful_proxies.txt', 'a') as f:
f.write(type_self.lower() + '---' + ip + ':' + port + '\n')
if __name__ == '__main__':
thread = []
que = Queue.Queue()
parser = optparse.OptionParser('usage %prog '+\
'-p <page num>')
parser.add_option('-p', dest='pagenum', type='int',\
help='specify page nums,default 5',default=5)
(options, args) = parser.parse_args()
pagenum = options.pagenum
for i in range(1, pagenum+1):
que.put('http://www.xicidaili.com/nn/' + str(i))
for i in range(pagenum):
thread.append(Proxy_collection(que))
start = time.clock()
for i in thread:
i.start()
for i in thread:
i.join()
end = time.clock()
print end - start
proxy
|