作为一个新世纪有思想有文化有道德时刻准备着的屌丝男青年,在现在这样一个社会中,心疼我大慢播抵制大百度的前提下,没事儿上上网逛逛YY看看斗鱼翻翻美女图片那是必不可少的,可是美图虽多翻页费劲!今天我们就搞个爬虫把美图都给扒下来!本次实例有2个:煎蛋上的妹子图,某网站的rosi图。我只是一个学习python的菜鸟,技术不可耻,技术是无罪的!!!
煎蛋:
先说说程序的流程:获取煎蛋妹子图URL,得到网页代码,提取妹子图片地址,访问图片地址并将图片保存到本地。Ready? 先让我们看看煎蛋妹子网页:
我们得到URL为:http://jandan.net/ooxx/page-1764#comments 1764就是页码, 首先我们要得到最新的页码,然后向前寻找,然后得到每页中图片的url。下面我们分析网站代码写出正则表达式!
根据之前文章的方法我们写出如下函数getNewPage:
def __getNewPage(self): pageCode = self.Get(self.__Url) type = sys.getfilesystemencoding() pattern = re.compile(r'<div .*">.*">\[(.*"UTF-8").encode(type)) print pageCode.decode("UTF-8").encode(type) if newPage != None: return newPage.group(1) return 1500
不要问我为什么如果失败返回1500。。。 因为煎蛋把1500页之前的图片都给吃了。 你也可以返回0。接下来是图片的
def __getAllPicUrl(self,pageIndex): realurl = self.__Url + "page-" + str(pageIndex) + "#comments" pageCode = self.Get(realurl) type = sys.getfilesystemencoding() pattern = re.compile('<p>.*">.*"(.*".*"UTF-8").encode(type)) for item in items: print item
好了,得到了图片地址,接下来就是访问图片地址然后保存图片了:
def __savePics(self,img_addr,folder): for item in img_addr: filename = item.split('/')[-1] print "正在保存图片:" + filename with open(filename,'wb') as file: img = self.Get(item) file.write(img)
当你觉得信心满满的时候,一定会有一盆冷水浇到你的头上,毕竟程序就是这样,考验你的耐性,打磨你的自信。你测试了一会儿,然后你发现你重启程序后再也无法获取最新页码,你觉得我什么也没动啊为什么会这样。别着急,我们将得到的网页代码打印出来看看:
看到了吧,是服务器感觉你不像浏览器访问的结果把你的ip给屏蔽了。 真是给跪了,辛辛苦苦码一年,屏蔽回到解放前!那么这个如何解决呢,答:换ip 找代理。接下来我们要改一下我们的HttpClient.py 将里面的opener设置下代理服务器。具体代理服务器请自行百度之,关键字:http代理 。 想找到一个合适的代理也不容易 自己ie Internet选项挨个试试,测试下网速。
# -*- coding: utf-8 -*- import cookielib, urllib, urllib2, socket import zlib,StringIO class HttpClient: __cookie = cookielib.CookieJar() __proxy_handler = urllib2.ProxyHandler({"http" : '42.121.6.80:8080'})#设置代理服务器与端口 __req = urllib2.build_opener(urllib2.HTTPCookieProcessor(__cookie),__proxy_handler)#生成opener __req.addheaders = [ ('Accept', 'application/javascript, */*;q=0.8'), ('User-Agent', 'Mozilla/5.0 (compatible; MSIE 9.0; Windows NT 6.1; WOW64; Trident/5.0)') ] urllib2.install_opener(__req) def Get(self, url, refer=None): try: req = urllib2.Request(url) #req.add_header('Accept-encoding', 'gzip') if not (refer is None): req.add_header('Referer', refer) response = urllib2.urlopen(req, timeout=120) html = response.read() #gzipped = response.headers.get('Content-Encoding') #if gzipped: # html = zlib.decompress(html, 16+zlib.MAX_WBITS) return html except urllib2.HTTPError, e: return e.read() except socket.timeout, e: return '' except socket.error, e: return ''
然后,就可以非常愉快的查看图片了。不过用了代理速度好慢。。。可以设置timeout稍微长一点儿,防止图片下载不下来!
好了,rosi的下篇文章再放!现在是时候上一波代码了:
# -*- coding: utf-8 -*- import cookielib, urllib, urllib2, socket import zlib,StringIO class HttpClient: __cookie = cookielib.CookieJar() __proxy_handler = urllib2.ProxyHandler({"http" : '42.121.6.80:8080'}) __req = urllib2.build_opener(urllib2.HTTPCookieProcessor(__cookie),__proxy_handler) __req.addheaders = [ ('Accept', 'application/javascript, */*;q=0.8'), ('User-Agent', 'Mozilla/5.0 (compatible; MSIE 9.0; Windows NT 6.1; WOW64; Trident/5.0)') ] urllib2.install_opener(__req) def Get(self, url, refer=None): try: req = urllib2.Request(url) req.add_header('Accept-encoding', 'gzip') if not (refer is None): req.add_header('Referer', refer) response = urllib2.urlopen(req, timeout=120) html = response.read() gzipped = response.headers.get('Content-Encoding') if gzipped: html = zlib.decompress(html, 16+zlib.MAX_WBITS) return html except urllib2.HTTPError, e: return e.read() except socket.timeout, e: return '' except socket.error, e: return '' def Post(self, url, data, refer=None): try: #req = urllib2.Request(url, urllib.urlencode(data)) req = urllib2.Request(url,data) if not (refer is None): req.add_header('Referer', refer) return urllib2.urlopen(req, timeout=120).read() except urllib2.HTTPError, e: return e.read() except socket.timeout, e: return '' except socket.error, e: return '' def Download(self, url, file): output = open(file, 'wb') output.write(urllib2.urlopen(url).read()) output.close() # def urlencode(self, data): # return urllib.quote(data) def getCookie(self, key): for c in self.__cookie: if c.name == key: return c.value return '' def setCookie(self, key, val, domain): ck = cookielib.Cookie(version=0, name=key, value=val, port=None, port_specified=False, domain=domain, domain_specified=False, domain_initial_dot=False, path='/', path_specified=True, secure=False, expires=None, discard=True, comment=None, comment_url=None, rest={'HttpOnly': None}, rfc2109=False) self.__cookie.set_cookie(ck) #self.__cookie.clear() clean cookie # vim : tabstop=2 shiftwidth=2 softtabstop=2 expandtab HttpClient
# -*- coding: utf-8 -*- from __future__ import unicode_literals from HttpClient import HttpClient import sys,re,os class JianDan(HttpClient): def __init__(self): self.__pageIndex = 1500 #之前的图片被煎蛋吞了 self.__Url = "http://jandan.net/ooxx/" self.__floder = "jiandan" def __getAllPicUrl(self,pageIndex): realurl = self.__Url + "page-" + str(pageIndex) + "#comments" pageCode = self.Get(realurl) type = sys.getfilesystemencoding() pattern = re.compile('<p>.*">.*"(.*".*"UTF-8").encode(type)) for item in items: print item self.__savePics(items,self.__floder) def __savePics(self,img_addr,folder): for item in img_addr: filename = item.split('/')[-1] print "正在保存图片:" + filename with open(filename,'wb') as file: img = self.Get(item) file.write(img) def __getNewPage(self): pageCode = self.Get(self.__Url) type = sys.getfilesystemencoding() pattern = re.compile(r'<div .*">.*">\[(.*"UTF-8").encode(type)) print pageCode.decode("UTF-8").encode(type) if newPage != None: return newPage.group(1) return 1500 def start(self): isExists=os.path.exists(self.__floder)#检测是否存在目录 print isExists if not isExists: os.mkdir(self.__floder) os.chdir(self.__floder) page = int(self.__getNewPage()) for i in range(self.__pageIndex,page): self.__getAllPicUrl(i) if __name__ == '__main__': jd = JianDan() jd.start() JianDan