python爬取妹子图片2

这次是爬取妹子图片1的改良，代码变得更加简洁了，使用了requests库

#-*-coding:utf-8import requestsimport lxmlimport osfrom bs4 import BeautifulSoupclass MEZI():    def __init__(self,url,pageIndex):        self.url=url+str(pageIndex)        self.headers = {'User-Agent':'Mozilla/4.0 (compatible; MSIE 5.5; Windows NT)'}    #得到一页的网页源码    def get_one_page_html(self):        re=requests.get(self.url,headers=self.headers)        html=re.text        return html        #得到一个妹子的写真集的url    def get_one_person_url(self):        all_a=[]        html=self.get_one_page_html()        soup=BeautifulSoup(html,'lxml')        all_div=soup.find('div',class_='postlist')        all_span=all_div.find('ul').find_all('span')        for span in all_span:            if span.find('a'):                all_a.append(span.find('a').get('href'))        return all_a    def download(self,filename):   #filename是你要存储的路径名        x=1        all_a=self.get_one_person_url()        for a in all_a:   #这里只是一张照片的地址            for i in range(1,30):    #限制每个人下载30张照片                url=a+'/'+str(i)                re=requests.get(url ,headers=self.headers)                print u'正在下载图片:%s'%re.url                html=re.text                soup=BeautifulSoup(html,'lxml')                img_url=soup.find('img').get('src')                html_img = requests.get(img_url, headers=self.headers)                img_name=str(x)+'.jpg'                path=os.path.normpath(os.path.join(filename,img_name))    #这里是规范你的绝对路径,os.path.join(filename,img_name)是将目录名和文件名连接在一起                with open(path,'wb') as f:   #开始下载                    f.write(html_img.content)                x=x+1if __name__=="__main__":    url='http://www.mzitu.com/page/'    for i in range(1,30):        app = MEZI(url, i)        filename = 'g:\\picture2'        app.download(filename)