一、为什么要搭建爬虫代理池
在众多的网站防爬措施中，有一种是根据ip的访问频率进行限制，即在某一时间段内，当某个ip的访问次数达到一定的阀值时，该ip就会被拉黑、在一段时间内禁止访问。
应对的方法有两种：
1. 降低爬虫的爬取频率，避免IP被限制访问，缺点显而易见：会大大降低爬取的效率。
2. 搭建一个IP代理池，使用不同的IP轮流进行爬取。
二、搭建思路
1、从代理网站(如：西刺代理、快代理、云代理、无忧代理)爬取代理IP；
2、验证代理IP的可用性（使用代理IP去请求指定URL，根据响应验证代理IP是否生效）；
3、将可用的代理IP保存到数据库；
常用代理网站：西刺代理 、云代理 、IP海 、无忧代理 、飞蚁代理 、快代理
三、代码实现
工程结构如下：
ipproxy.py
IPProxy代理类定义了要爬取的IP代理的字段信息和一些基础方法。
# -*- coding: utf-8 -*-
import re
import time
from settings import PROXY_URL_FORMATTER
schema_pattern = re.compile(r'http|https$', re.I)
ip_pattern = re.compile(r'^([0-9]{1,3}.){3}[0-9]{1,3}$', re.I)
port_pattern = re.compile(r'^[0-9]{2,5}$', re.I)
class IPProxy:
'''
{
"schema": "http", # 代理的类型
"ip": "127.0.0.1", # 代理的IP地址
"port": "8050", # 代理的端口号
"used_total": 11, # 代理的使用次数
"success_times": 5, # 代理请求成功的次数
"continuous_failed": 3, # 使用代理发送请求，连续失败的次数
"created_time": "2018-05-02" # 代理的爬取时间
}
'''
def __init__(self, schema, ip, port, used_total=0, success_times=0, continuous_failed=0,
created_time=None):
"""Initialize the proxy instance"""
if schema == "" or schema is None:
schema = "http"
self.schema = schema.lower()
self.ip = ip
self.port = port
self.used_total = used_total
self.success_times = success_times
self.continuous_failed = continuous_failed
if created_time is None:
created_time = time.strftime('%Y-%m-%d', time.localtime(time.time()))
self.created_time = created_time
def _get_url(self):
''' Return the proxy url'''
return PROXY_URL_FORMATTER % {'schema': self.schema, 'ip': self.ip, 'port': self.port}
def _check_format(self):
''' Return True if the proxy fields are well-formed,otherwise return False'''
if self.schema is not None and self.ip is not None and self.port is not None:
if schema_pattern.match(self.schema) and ip_pattern.match(self.ip) and port_pattern.match(self.port):
return True
return False
def _is_https(self):
''' Return True if the proxy is https,otherwise return False'''
return self.schema == 'https'
def _update(self, successed=False):
''' Update proxy based on the result of the request's response'''
self.used_total = self.used_total + 1
if successed:
self.continuous_failed = 0
self.success_times = self.success_times + 1
else:
print(self.continuous_failed)
self.continuous_failed = self.continuous_failed + 1
if __name__ == '__main__':
proxy = IPProxy('HTTPS', '192.168.2.25', "8080")
print(proxy._get_url())
print(proxy._check_format())
print(proxy._is_https())
settings.py
settings.py中汇聚了工程所需要的配置信息。
# 指定Redis的主机名和端口
REDIS_HOST = 'localhost'
REDIS_PORT = 6379
# 代理保存到Redis key 格式化字符串
PROXIES_REDIS_FORMATTER = 'proxies::{}'
# 已经存在的HTTP代理和HTTPS代理集合
PROXIES_REDIS_EXISTED = 'proxies::existed'
# 最多连续失败几次
MAX_CONTINUOUS_TIMES = 3
# 代理地址的格式化字符串
PROXY_URL_FORMATTER = '%(schema)s://%(ip)s:%(port)s'
USER_AGENT_LIST = [
"Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.1 (KHTML, like Gecko) Chrome/22.0.1207.1 Safari/537.1",
"Mozilla/5.0 (X11; CrOS i686 2268.111.0) AppleWebKit/536.11 (KHTML, like Gecko) Chrome/20.0.1132.57 Safari/536.11",
"Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/536.6 (KHTML, like Gecko) Chrome/20.0.1092.0 Safari/536.6",
"Mozilla/5.0 (Windows NT 6.2) AppleWebKit/536.6 (KHTML, like Gecko) Chrome/20.0.1090.0 Safari/536.6",
"Mozilla/5.0 (Windows NT 6.2; WOW64) AppleWebKit/537.1 (KHTML, like Gecko) Chrome/19.77.34.5 Safari/537.1",
"Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/536.5 (KHTML, like Gecko) Chrome/19.0.1084.9 Safari/536.5",
"Mozilla/5.0 (Windows NT 6.0) AppleWebKit/536.5 (KHTML, like Gecko) Chrome/19.0.1084.36 Safari/536.5",
"Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/536.3 (KHTML, like Gecko) Chrome/19.0.1063.0 Safari/536.3",
"Mozilla/5.0 (Windows NT 5.1) AppleWebKit/536.3 (KHTML, like Gecko) Chrome/19.0.1063.0 Safari/536.3",
"Mozilla/5.0 (Macintosh; Intel Mac OS X 10_8_0) AppleWebKit/536.3 (KHTML, like Gecko) Chrome/19.0.1063.0 Safari/536.3",
"Mozilla/5.0 (Windows NT 6.2) AppleWebKit/536.3 (KHTML, like Gecko) Chrome/19.0.1062.0 Safari/536.3",
"Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/536.3 (KHTML, like Gecko) Chrome/19.0.1062.0 Safari/536.3",
"Mozilla/5.0 (Windows NT 6.2) AppleWebKit/536.3 (KHTML, like Gecko) Chrome/19.0.1061.1 Safari/536.3",
"Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/536.3 (KHTML, like Gecko) Chrome/19.0.1061.1 Safari/536.3",
"Mozilla/5.0 (Windows NT 6.1) AppleWebKit/536.3 (KHTML, like Gecko) Chrome/19.0.1061.1 Safari/536.3",
"Mozilla/5.0 (Windows NT 6.2) AppleWebKit/536.3 (KHTML, like Gecko) Chrome/19.0.1061.0 Safari/536.3",
"Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/535.24 (KHTML, like Gecko) Chrome/19.0.1055.1 Safari/535.24",
"Mozilla/5.0 (Windows NT 6.2; WOW64) AppleWebKit/535.24 (KHTML, like Gecko) Chrome/19.0.1055.1 Safari/535.24"
]
# 爬取到的代理保存前先检验是否可用，默认True
PROXY_CHECK_BEFOREADD = True
# 检验代理可用性的请求地址，支持多个
PROXY_CHECK_URLS = {'https':['https://icanhazip.com'],'http':['http://icanhazip.com']}
proxy_util.py
proxy_util.py 中主要定义了一些实用方法，例如：proxy_to_dict(proxy)用来将IPProxy代理实例转换成字典；proxy_from_dict(d)用来将字典转换为IPProxy实例；request_page()用来发送请求；_is_proxy_available()用来校验代理IP是否可用。
# -*- coding: utf-8 -*-
import random
import logging
import requests
from ipproxy import IPProxy
from settings import USER_AGENT_LIST, PROXY_CHECK_URLS
# Setting logger output format
logging.basicConfig(level=logging.INFO,
format='[%(asctime)-15s] [%(levelname)8s] [%(name)10s ] - %(message)s (%(filename)s:%(lineno)s)',
datefmt='%Y-%m-%d %T'
)
logger = logging.getLogger(__name__)
def proxy_to_dict(proxy):
d = {
"schema": proxy.schema,
"ip": proxy.ip,
"port": proxy.port,
"used_total": proxy.used_total,
"success_times": proxy.success_times,
"continuous_failed": proxy.continuous_failed,
"created_time": proxy.created_time
}
return d
def proxy_from_dict(d):
return IPProxy(schema=d['schema'], ip=d['ip'], port=d['port'], used_total=d['used_total'],
success_times=d['success_times'], continuous_failed=d['continuous_failed'],
created_time=d['created_time'])
# Truncate header and tailer blanks
def strip(data):
if data is not None:
return data.strip()
return data
base_headers = {
'Accept-Encoding': 'gzip, deflate, br',
'Accept-Language': 'en-US,en;q=0.9,zh-CN;q=0.8,zh;q=0.7'
}
def request_page(url, options={}, encoding='utf-8'):
"""send a request,get response"""
headers = dict(base_headers, **options)
if 'User-Agent' not in headers.keys():
headers['User-Agent'] = random.choice(USER_AGENT_LIST)
logger.info('正在抓取: ' + url)
try:
response = requests.get(url, headers=headers)
if response.status_code == 200:
logger.info('抓取成功: ' + url)
return response.content.decode(encoding=encoding)
except ConnectionError:
logger.error('抓取失败' + url)
return None
def _is_proxy_available(proxy, options={}):
"""Check whether the Proxy is available or not"""
headers = dict(base_headers, **options)
if 'User-Agent' not in headers.keys():
headers['User-Agent'] = random.choice(USER_AGENT_LIST)
proxies = {proxy.schema: proxy._get_url()}
check_urls = PROXY_CHECK_URLS[proxy.schema]
for url in check_urls:
try:
response = requests.get(url=url, proxies=proxies, headers=headers, timeout=5)
except BaseException:
logger.info("< " + url + " > 验证代理 < " + proxy._get_url() + " > 结果： 不可用 ")
else:
if response.status_code == 200:
logger.info("< " + url + " > 验证代理 < " + proxy._get_url() + " > 结果： 可用 ")
return True
else:
logger.info("< " + url + " > 验证代理 < " + proxy._get_url() + " > 结果： 不可用 ")
return False
if __name__ == '__main__':
headers = dict(base_headers)
if 'User-Agent' not in headers.keys():
headers['User-Agent'] = random.choice(USER_AGENT_LIST)
proxies = {"https": "https://163.125.255.154:9797"}
response = requests.get("https://www.baidu.com", headers=headers, proxies=proxies, timeout=3)
print(response.content)
proxy_queue.py
代理队列用来保存并对外提供 IP代理，不同的代理队列内代理IP的保存和提取策略可以不同。在这里， BaseQueue 是所有代理队列的基类，其中声明了所有代理队列都需要实现的保存代理IP、提取代理IP、查看代理IP数量等接口。示例的 FifoQueue 是一个先进先出队列，底层使用 Redis 列表实现，为了确保同一个代理IP只能被放入队列一次，这里使用了一个Redis proxies::existed 集合进行入队前重复校验。
# -*- coding: utf-8 -*-
from proxy_util import logger
import json
import redis
from ipproxy import IPProxy
from proxy_util import proxy_to_dict, proxy_from_dict, _is_proxy_available
from settings import PROXIES_REDIS_EXISTED, PROXIES_REDIS_FORMATTER, MAX_CONTINUOUS_TIMES, PROXY_CHECK_BEFOREADD
"""
Proxy Queue Base Class
"""
class BaseQueue(object):
def __init__(self, server):
"""Initialize the proxy queue instance
Parameters
----------
server : StrictRedis
Redis client instance
"""
self.server = server
def _serialize_proxy(self, proxy):
"""Serialize proxy instance"""
return proxy_to_dict(proxy)
def _deserialize_proxy(self, serialized_proxy):
"""deserialize proxy instance"""
return proxy_from_dict(eval(serialized_proxy))
def __len__(self, schema='http'):
"""Return the length of the queue"""
raise NotImplementedError
def push(self, proxy, need_check):
"""Push a proxy"""
raise NotImplementedError
def pop(self, schema='http', timeout=0):
"""Pop a proxy"""
raise NotImplementedError
class FifoQueue(BaseQueue):
"""First in first out queue"""
def __len__(self, schema='http'):
"""Return the length of the queue"""
return self.server.llen(PROXIES_REDIS_FORMATTER.format(schema))
def push(self, proxy, need_check=PROXY_CHECK_BEFOREADD):
"""Push a proxy"""
if need_check and not _is_proxy_available(proxy):
return
elif proxy.continuous_failed < MAX_CONTINUOUS_TIMES and not self._is_existed(proxy):
key = PROXIES_REDIS_FORMATTER.format(proxy.schema)
self.server.rpush(key, json.dumps(self._serialize_proxy(proxy),ensure_ascii=False))
def pop(self, schema='http', timeout=0):
"""Pop a proxy"""
if timeout > 0:
p = self.server.blpop(PROXIES_REDIS_FORMATTER.format(schema.lower()), timeout)
if isinstance(p, tuple):
p = p[1]
else:
p = self.server.lpop(PROXIES_REDIS_FORMATTER.format(schema.lower()))
if p:
p = self._deserialize_proxy(p)
self.server.srem(PROXIES_REDIS_EXISTED, p._get_url())
return p
def _is_existed(self, proxy):
added = self.server.sadd(PROXIES_REDIS_EXISTED, proxy._get_url())
return added == 0
if __name__ == '__main__':
r = redis.StrictRedis(host='localhost', port=6379)
queue = FifoQueue(r)
proxy = IPProxy('http', '218.66.253.144', '80')
queue.push(proxy)
proxy = queue.pop(schema='http')
print(proxy._get_url())
proxy_crawlers.py
ProxyBaseCrawler 是所有代理爬虫的基类，其中只定义了一个 _start_crawl() 方法用来从搜集到的代理网站爬取代理IP。
# -*- coding: utf-8 -*-
from lxml import etree
from ipproxy import IPProxy
from proxy_util import strip, request_page, logger
class ProxyBaseCrawler(object):
def __init__(self, queue=None, website=None, urls=[]):
self.queue = queue
self.website = website
self.urls = urls
def _start_crawl(self):
raise NotImplementedError
class KuaiDailiCrawler(ProxyBaseCrawler): # 快代理
def _start_crawl(self):
for url_dict in self.urls:
logger.info("开始爬取 [ " + self.website + " ] :::> [ " + url_dict['type'] + " ]")
has_more = True
url = None
while has_more:
if 'page' in url_dict.keys() and str.find(url_dict['url'], '{}') != -1:
url = url_dict['url'].format(str(url_dict['page']))
url_dict['page'] = url_dict['page'] + 1
else:
url = url_dict['url']
has_more = False
html = etree.HTML(request_page(url))
tr_list = html.xpath("//table[@class='table table-bordered table-striped']/tbody/tr")
for tr in tr_list:
ip = tr.xpath("./td[@data-title='IP']/text()")[0] if len(
tr.xpath("./td[@data-title='IP']/text()")) else None
port = tr.xpath("./td[@data-title='PORT']/text()")[0] if len(
tr.xpath("./td[@data-title='PORT']/text()")) else None
schema = tr.xpath("./td[@data-title='类型']/text()")[0] if len(
tr.xpath("./td[@data-title='类型']/text()")) else None
proxy = IPProxy(schema=strip(schema), ip=strip(ip), port=strip(port))
if proxy._check_format():
self.queue.push(proxy)
if tr_list is None:
has_more = False
class FeiyiDailiCrawler(ProxyBaseCrawler): # 飞蚁代理
def _start_crawl(self):
for url_dict in self.urls:
logger.info("开始爬取 [ " + self.website + " ] :::> [ " + url_dict['type'] + " ]")
has_more = True
url = None
while has_more:
if 'page' in url_dict.keys() and str.find(url_dict['url'], '{}') != -1:
url = url_dict['url'].format(str(url_dict['page']))
url_dict['page'] = url_dict['page'] + 1
else:
url = url_dict['url']
has_more = False
html = etree.HTML(request_page(url))
tr_list = html.xpath("//div[@id='main-content']//table/tr[position()>1]")
for tr in tr_list:
ip = tr.xpath("./td[1]/text()")[0] if len(tr.xpath("./td[1]/text()")) else None
port = tr.xpath("./td[2]/text()")[0] if len(tr.xpath("./td[2]/text()")) else None
schema = tr.xpath("./td[4]/text()")[0] if len(tr.xpath("./td[4]/text()")) else None
proxy = IPProxy(schema=strip(schema), ip=strip(ip), port=strip(port))
if proxy._check_format():
self.queue.push(proxy)
if tr_list is None:
has_more = False
class WuyouDailiCrawler(ProxyBaseCrawler): # 无忧代理
def _start_crawl(self):
for url_dict in self.urls:
logger.info("开始爬取 [ " + self.website + " ] :::> [ " + url_dict['type'] + " ]")
has_more = True
url = None
while has_more:
if 'page' in url_dict.keys() and str.find(url_dict['url'], '{}') != -1:
url = url_dict['url'].format(str(url_dict['page']))
url_dict['page'] = url_dict['page'] + 1
else:
url = url_dict['url']
has_more = False
html = etree.HTML(request_page(url))
ul_list = html.xpath("//div[@class='wlist'][2]//ul[@class='l2']")
for ul in ul_list:
ip = ul.xpath("./span[1]/li/text()")[0] if len(ul.xpath("./span[1]/li/text()")) else None
port = ul.xpath("./span[2]/li/text()")[0] if len(ul.xpath("./span[2]/li/text()")) else None
schema = ul.xpath("./span[4]/li/text()")[0] if len(ul.xpath("./span[4]/li/text()")) else None
proxy = IPProxy(schema=strip(schema), ip=strip(ip), port=strip(port))
if proxy._check_format():
self.queue.push(proxy)
if ul_list is None:
has_more = False
class IPhaiDailiCrawler(ProxyBaseCrawler): # IP海代理
def _start_crawl(self):
for url_dict in self.urls:
logger.info("开始爬取 [ " + self.website + " ] :::> [ " + url_dict['type'] + " ]")
has_more = True
url = None
while has_more:
if 'page' in url_dict.keys() and str.find(url_dict['url'], '{}') != -1:
url = url_dict['url'].format(str(url_dict['page']))
url_dict['page'] = url_dict['page'] + 1
else:
url = url_dict['url']
has_more = False
html = etree.HTML(request_page(url))
tr_list = html.xpath("//table//tr[position()>1]")
for tr in tr_list:
ip = tr.xpath("./td[1]/text()")[0] if len(tr.xpath("./td[1]/text()")) else None
port = tr.xpath("./td[2]/text()")[0] if len(tr.xpath("./td[2]/text()")) else None
schema = tr.xpath("./td[4]/text()")[0] if len(tr.xpath("./td[4]/text()")) else None
proxy = IPProxy(schema=strip(schema), ip=strip(ip), port=strip(port))
if proxy._check_format():
self.queue.push(proxy)
if tr_list is None:
has_more = False
class YunDailiCrawler(ProxyBaseCrawler): # 云代理
def _start_crawl(self):
for url_dict in self.urls:
logger.info("开始爬取 [ " + self.website + " ] :::> [ " + url_dict['type'] + " ]")
has_more = True
url = None
while has_more:
if 'page' in url_dict.keys() and str.find(url_dict['url'], '{}') != -1:
url = url_dict['url'].format(str(url_dict['page']))
url_dict['page'] = url_dict['page'] + 1
else:
url = url_dict['url']
has_more = False
html = etree.HTML(request_page(url, encoding='gbk'))
tr_list = html.xpath("//table/tbody/tr")
for tr in tr_list:
ip = tr.xpath("./td[1]/text()")[0] if len(tr.xpath("./td[1]/text()")) else None
port = tr.xpath("./td[2]/text()")[0] if len(tr.xpath("./td[2]/text()")) else None
schema = tr.xpath("./td[4]/text()")[0] if len(tr.xpath("./td[4]/text()")) else None
proxy = IPProxy(schema=strip(schema), ip=strip(ip), port=strip(port))
if proxy._check_format():
self.queue.push(proxy)
if tr_list is None:
has_more = False
class XiCiDailiCrawler(ProxyBaseCrawler): # 西刺代理
def _start_crawl(self):
for url_dict in self.urls:
logger.info("开始爬取 [ " + self.website + " ] :::> [ " + url_dict['type'] + " ]")
has_more = True
url = None
while has_more:
if 'page' in url_dict.keys() and str.find(url_dict['url'], '{}') != -1:
url = url_dict['url'].format(str(url_dict['page']))
url_dict['page'] = url_dict['page'] + 1
else:
url = url_dict['url']
has_more = False
html = etree.HTML(request_page(url))
tr_list = html.xpath("//table[@id='ip_list']//tr[@class!='subtitle']")
for tr in tr_list:
ip = tr.xpath("./td[2]/text()")[0] if len(tr.xpath("./td[2]/text()")) else None
port = tr.xpath("./td[3]/text()")[0] if len(tr.xpath("./td[3]/text()")) else None
schema = tr.xpath("./td[6]/text()")[0] if len(tr.xpath("./td[6]/text()")) else None
if schema.lower() == "http" or schema.lower() == "https":
proxy = IPProxy(schema=strip(schema), ip=strip(ip), port=strip(port))
if proxy._check_format():
self.queue.push(proxy)
if tr_list is None:
has_more = False
run.py
通过run.py启动各个代理网站爬虫。
# -*- coding: utf-8 -*-
import redis
from proxy_queue import FifoQueue
from settings import REDIS_HOST, REDIS_PORT
from proxy_crawlers import WuyouDailiCrawler, FeiyiDailiCrawler, KuaiDailiCrawler, IPhaiDailiCrawler, YunDailiCrawler, XiCiDailiCrawler
r = redis.StrictRedis(host=REDIS_HOST, port=REDIS_PORT)
fifo_queue = FifoQueue(r)
def run_kuai():
kuaidailiCrawler = KuaiDailiCrawler(queue=fifo_queue, website='快代理[国内高匿]',
urls=[{'url': 'https://www.kuaidaili.com/free/inha/{}/', 'type': '国内高匿',
'page': 1},
{'url': 'https://www.kuaidaili.com/free/intr/{}/', 'type': '国内普通',
'page': 1}])
kuaidailiCrawler._start_crawl()
def run_feiyi():
feiyidailiCrawler = FeiyiDailiCrawler(queue=fifo_queue, website='飞蚁代理',
urls=[{'url': 'http://www.feiyiproxy.com/"text-align: center">
Redis数据库中爬取到的代理IP的数据结构如下:
四、代理测试
接下来，使用爬取好的代理来请求 http://icanhazip.com 进行测试，代码如下：
# -*- coding: utf-8 -*-
import random
import requests
from proxy_util import logger
from run import fifo_queue
from settings import USER_AGENT_LIST
from proxy_util import base_headers
# 测试地址
url = 'http://icanhazip.com'
# 获取代理
proxy = fifo_queue.pop(schema='http')
proxies = {proxy.schema:proxy._get_url()}
# 构造请求头
headers = dict(base_headers)
if 'User-Agent' not in headers.keys():
headers['User-Agent'] = random.choice(USER_AGENT_LIST)
response = None
successed = False
try:
response = requests.get(url,headers=headers,proxies = proxies,timeout=5)
except BaseException:
logger.error("使用代理< "+proxy._get_url()+" > 请求 < "+url+" > 结果： 失败 ")
else:
if (response.status_code == 200):
logger.info(response.content.decode())
successed = True
logger.info("使用代理< " + proxy._get_url() + " > 请求 < " + url + " > 结果： 成功 ")
else:
logger.info(response.content.decode())
logger.info("使用代理< " + proxy._get_url() + " > 请求 < " + url + " > 结果： 失败 ")
# 根据请求的响应结果更新代理
proxy._update(successed)
# 将代理返还给队列，返还时不校验可用性
fifo_queue.push(proxy,need_check=False)
使用 http://218.66.253.144:80 代理请求成功后将代理重新放回队列，并将 Redis 中该代理的 used_total 、success_times 、continuous_failed三个字段信息进行了相应的更新。
项目地址：https://github.com/pengjunlee/ipproxy_pool.git