DY2018.py 文件源码

def LoadPage(self, myUrl):
        user_agent = 'Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/50.0.2661.102 Safari/537.36'
        accept = 'text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,*/*;q=0.8'
        headers = {'User-Agent': user_agent, 'Accept': accept}
        print self.base_url + myUrl
        req = urllib2.Request(self.base_url+myUrl, headers=headers)
        myResponse = urllib2.urlopen(req)
        myPage = myResponse.read()
        #print myPage
        # encode?????unicode?????????????
        # decode????????????????unicode??
        unicodePage = myPage.decode("gb2312", 'ignore').encode('utf-8', 'ignore')
        # print unicodePage
        return unicodePage