python爬虫抓取心得分享
/**author: insuntitle:python 爬虫抓取心得分享blog:http://yxmhero1989.blog.163.com/blog/static/112157956201311821444664/**/0x1.urllib.quote('要编码的字符串')如果你要在url请求里面放入中文,对相应的中文进行编码的话,可以用:urllib.quote('要编码的字符串')query = urllib.quote(singername)url = 'http://music.baidu.com/search?key='+queryresponse = urllib.urlopen(url)text = response.read()0x2. get or post urlencode如果在GET需要一些参数的话,那我们需要对传入的参数进行编码。import urllibdef url_get():import urllibparams = urllib.urlencode({'spam': 1, 'eggs': 2, 'bacon': 0})f = urllib.urlopen("http://www.musi-cal.com/cgi-bin/query?%s" % params)print f.read()def url_post():import urllibparams = urllib.urlencode({'spam': 1, 'eggs': 2, 'bacon': 0})f = urllib.urlopen("http://www.musi-cal.com/cgi-bin/query", params)print f.read()0x3.urllib urllib2 proxy 代理如果你请求对方的网页,确不想被封IP的话,这个时候就要用到代理了,其实用urllib代理还是比较简单的:import urllibdef url_proxy():proxies = {'http':'http://211.167.112.14:80'}#或者proxies = {'':'211.167.112.14:80'}opener = urllib.FancyURLopener(proxies)f = opener.open("http://www.dianping.com/shanghai")print f.read()只用一个代理IP的话 有时候弄巧成拙了 恰好被大众点评给检测出来了401211.167.112.14python-urllib/1.17那么就试试多个IP代理import urllibdef url_proxies():proxylist = ('211.167.112.14:80','210.32.34.115:8080','115.47.8.39:80','211.151.181.41:80','219.239.26.23:80',)for proxy in proxylist:proxies = {'': proxy}opener = urllib.FancyURLopener(proxies)f = opener.open("http://www.dianping.com/shanghai")print f.read()这回没问题了。有的时候要模拟浏览器 ,不然做过反爬虫的网站会知道你是robot例如针对浏览器的限制我们可以设置User-Agent头部,针对防盗链限制,我们可以设置Referer头部有的网站用了Cookie来限制,主要是涉及到登录和限流,这时候没有什么通用的方法,只能看能否做自动登录或者分析Cookie的问题了。仅仅是模拟浏览器访问依然是不行的,如果爬取频率过高依然会令人怀疑,那么就需要用到上面的代理设置了import urllib2def url_user_agent(url):'''proxy = 'http://211.167.112.14:80'opener = urllib2.build_opener(urllib2.ProxyHandler({'http':proxy}), urllib2.HTTPHandler(debuglevel=1))urllib2.install_opener(opener)'''i_headers = {"User-Agent": "Mozilla/5.0 (Windows; U; Windows NT 5.1; zh-CN; rv:1.9.1) Gecko/20090624 Firefox/3.5",\"Referer": 'http://www.dianping.com/'}req = urllib2.Request(url, headers=i_headers)return urllib2.urlopen(req).read()#print url_user_agent('http://www.dianping.com/shanghai')就算设置了代理,代理的ip也有可能被封,还有另外一种终极的办法来防止被封,那便是使用time库的sleep()函数。import timefor i in range(1,10):....#抓取逻辑time.sleep(5)抓的地址是http://www.dianping.com/shanghai直接抓http://www.dianping.com的话会location到城市列表去 反而达不到效果header: Location: /citylist提供一段代理IPproxylist = ('211.167.112.14:80','210.32.34.115:8080','115.47.8.39:80','211.151.181.41:80','219.239.26.23:80','219.157.200.18:3128','219.159.105.180:8080','1.63.18.22:8080','221.179.173.170:8080','125.39.66.153:80','125.39.66.151:80','61.152.108.187:80','222.217.99.153:9000','125.39.66.146:80','120.132.132.119:8080','119.7.221.137:82','117.41.182.188:8080','202.116.160.89:80','221.7.145.42:8080','211.142.236.131:80','119.7.221.136:80','211.151.181.41:80','125.39.66.131:80','120.132.132.119:8080','112.5.254.30:80','106.3.98.82:80','119.4.250.105:80','123.235.12.118:8080','124.240.187.7补充:Web开发 , Python ,
上一个:Python抓取花瓣网图片脚本
下一个:python抓取bing主页背景图片