Scrapy代理设置

文章时效性提示

这是一篇发布于 892 天前的文章，部分信息可能已发生改变，请注意甄别。

Scapy代理设置__爬取代理实战

设置代理可以避免爬虫频繁爬取被封

付费代理就不用说了,我这里爬取免费代理

Scrapy代理使用HttpProxyMiddlewar中间件（默认开启）

在Scrapy中设置代理本质是将代理服务器的url写入request.meta[‘proxy’]

如果代理需要身份验证,需要通过HTTP头部的Proxy-Authorization字段传递用户账号和密码的身份验证信息

实战

接下来进行实战

xicidaili貌似没了

我用的 国内高匿免费HTTP代理IP - 快代理 (kuaidaili.com)

一上来三连鞭

scrapy startproject proxy;
cd proxy;
scrapy genspider kuaidaili kuaidaili.com;

然后直接开爬

import json

import scrapy


class KuaidailiSpider(scrapy.Spider):
    name = 'kuaidaili'
    allowed_domains = ['kuaidaili.com']

    # start_urls = ['http://kuaidaili.com/free/inha']
    def start_requests(self):
        for i in range(1, 4):
            yield scrapy.Request("https://www.kuaidaili.com/free/inha/%s/" % i)

    def parse(self, response):
        for sel in response.xpath("//tbody//tr"):
            ip = sel.xpath(".//td[1]/text()").extract_first()
            port = sel.xpath(".//td[2]/text()").extract_first()
            scheme = sel.xpath(".//td[4]/text()").extract_first().lower()
            # 使用爬取的代理验证
            url = "%s://httpbin.org/ip" % scheme
            proxy = "%s://%s:%s" % (scheme, ip, port)
            meta = {
                'proxy': proxy,
                'dont_retry': True,
                'download_timeout': 10,
                # 以下两个字段用于检验
                '_proxy_scheme': scheme,
                '_proxy_ip': ip
            }
            yield scrapy.Request(url, callback=self.check_available, meta=meta, dont_filter=True)

    def check_available(self, response):
        proxy_ip = response.meta['_proxy_ip_']
        if proxy_ip == json.loads(response.text)['origin']:
            yield
            {
                'proxy_scheme': response.meta['_proxy_scheme'],
                'proxy': response.meta['proxy']
            }

settings.py改一下设置

我尝试后发现直接503

还没有得到解决…

再次尝试

于是我更换了网站,换了个墙外的

1 minute ago checked proxy servers - Free proxy list service since 2004 year! Almost 10 years of proxy service (proxy-list.org)

这个尝试了一下 貌似是动态网站 我懒

所以又换了一个

Free Proxy List - Just Checked Proxy List (free-proxy-list.net)

这个可以

import json

import scrapy


class FreeScrapySpider(scrapy.Spider):
    name = 'free_proxy'
    allowed_domains = ['free-proxy-list.net']

    start_urls = ["https://free-proxy-list.net/"]
    # def start_requests(self):
    #     for i in range(1, 4):
    #         yield scrapy.Request("http://proxy-list.org/english/index.php?p=%s" % i)

    def parse(self, response):
        for sel in response.xpath("//tbody//tr"):
            ip = sel.xpath("./td[1]/text()").extract_first()
            port = sel.xpath("./td[2]/text()").extract_first()
            scheme = sel.xpath("./td[7]/text()").extract_first()
            if scheme == 'yes':
                scheme = 'https'
            else:
                scheme = 'http'
            url = "%s://httpbin.org/ip" % scheme
            proxy = "%s://%s:%s" % (scheme, ip,port)
        # ip = ip_port.split(":")[0]
            meta = {
                'proxy': proxy,
                'download_timeout': 10,
                'dont_retry': True,

                '_proxy_scheme': scheme,
                '_proxy_ip': ip
            }
            yield scrapy.Request(url, callback=self.check_available, meta=meta, dont_filter=True)

    def check_available(self, response):
        proxy_ip = response.meta['_proxy_ip']
        if proxy_ip == json.loads(response.text)['origin']:
            yield {
                'proxy_scheme': response.meta['_proxy_scheme'],
                'proxy': response.meta['proxy']
            }

然后实现随机代理

class RandomHttpProxyMiddleware(HttpProxyMiddleware):
    def __init__(self,auth_encoding='latin-1',proxy_list_file=None):
        if not proxy_list_file:
            raise scrapy.exceptions.NotConfigured
        self.auth_encoding = auth_encoding
        self.proxies = defaultdict(list)
        with open(proxy_list_file) as f:
            proxy_list = json.load(f)
            for proxy in proxy_list:
                scheme = proxy['proxy_scheme']
                url = proxy['proxy']
                self.proxies[scheme].append(self._get_proxy(url,scheme))
    @classmethod
    def from_crawler(cls, crawler):
        auth_encoding = crawler.settings.get('HTTPPROXY_AUTH_ENCODING','latin-1')
        proxy_list_file = crawler.settings.get("HTTPPROXY_PROXY_LIST_FILE")
        return cls(auth_encoding,proxy_list_file)
    def _set_proxy(self, request, scheme):
        creds,proxy = random.choice(self.proxies[scheme])
        request.meta['proxy'] = proxy
        if creds:
            request.headers["Proxy-Authorization"]=b'Basic'+creds

并在settings.py里更改

DOWNLOADER_MIDDLEWARES = {
#    'free_proxy.middlewares.FreeProxyDownloaderMiddleware': 543,
    'free_proxy.middlewares.RandomHttpProxyMiddleware':745
}
HTTPPROXY_PROXY_LIST_FILE = 'proxy_list_json'

import scrapy
import json


class HubSpider(scrapy.Spider):
    name = 'test'
    allowed_domains = ['httpbin.org']

    # start_urls = ['http://httpbin.org/ip']
    def start_requests(self):
        for i in range(10):
            yield scrapy.Request('http://httpbin.org/ip', dont_filter=True)
            yield scrapy.Request('https://httpbin.org/ip', dont_filter=True)

    def parse(self, response):
        print(json.loads(response.text))
        yield {
            'proxy':json.loads(response.text)['origin']
        }
       # 写一个测试程序

可以看到得到的代理是设置的代理