淘女郎相册爬虫（Python编写）

# *-* coding:utf-8 *-*

__author__ = 'YS'

import urllib2
import urllib
import re
import json
import os
import time

#抓取淘女郎的图片,淘女郎地址:https://mm.taobao.com/search_tstar_model.htm?spm=5679.126488.640745.2.22495f9f1lYEAb
class MMSpider:
    def __init__(self, timeout=3, albumLimit=200, picLimit=500, sleepPicCount=100, savePath='pythonspider/'):
        self.__headers = {
            'User-Agent':'User-Agent:Mozilla/5.0 (Macintosh; Intel Mac OS X 10_11_6) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/60.0.3112.113 Safari/537.36'
        }
        #抓取时间超时设置
        self.timeout = timeout
        #抓取的相册个数限制
        self.albumLimit = albumLimit
        #获取MM列表的地址
        self.__mmListUrl = 'https://mm.taobao.com/tstar/search/tstar_model.do?_input_charset=utf-8'
        #获取相册列表的地址
        self.__albumListUrl = 'https://mm.taobao.com/self/album/open_album_list.htm?_charset=utf-8&user_id%20=:userId&page=:page'
        #获取相册具体相片的地址
        self.__albumDetailUrl = 'https://mm.taobao.com/album/json/get_album_photo_list.htm?user_id=:userId&album_id=:albumId&page=:page'
        #MM详情页面地址
        self.__personUrl = 'https://mm.taobao.com/self/aiShow.htm?userId=:userId'
        #抓取的文件存放路径
        self.savePath = savePath
        #每个MM的照片最多抓多少张
        self.picLimit = picLimit
        #抓取多少张图片时休息1秒
        self.sleepPicCount = sleepPicCount

        self.__mkdir(self.savePath)

    #获取页面内容,python中的异常继承关系: https://docs.python.org/3/library/exceptions.html#exception-hierarchy
    def __getContents(self, url, data=None, encoding=None, isjson=None):
        try:
            request = urllib2.Request(url, data, self.__headers)
            response = urllib2.urlopen(request, timeout=self.timeout)
            if encoding:
                contents =  response.read().decode(encoding).encode('utf-8')
            else:
                contents = response.read()

            return json.loads(contents,encoding='utf-8') if isjson else contents
        except urllib2.URLError,e:
            print '出错了' + e.reason
            return None
        except BaseException,e:
            print '其他错误'
            print e.args
            return None

    #获取MM列表
    def __getMMList(self, pageIndex):
        url = self.__mmListUrl
        data = urllib.urlencode({
            'currentPage':pageIndex,
            'pageSize':50
        })
        list = self.__getContents(url, data, encoding='gbk', isjson=True)

        if list is None:
            return None
        elif list['status'] != 1:
            return None
        return list['data']['searchDOList']

    #获取相册列表
    def __getAlbumList(self, mm):
        albumList = []
        baseUrl = self.__albumListUrl.replace(':userId',str(mm['userId']))
        indexUrl = baseUrl.replace(':page','1')
        pageCount = int(self.__getAlbumListPage(indexUrl))
        pageCount = pageCount if pageCount<=self.albumLimit else self.albumLimit
        
        for i in range(1, pageCount+1):
            listUrl = baseUrl.replace(':page', str(i))
            contents = self.__getContents(listUrl)

            if (contents is None):
                continue
            pattern = re.compile('<h4><a href=".*?album_id=(.*?)&album_flag', re.S)
            items = re.findall(pattern, contents)
        
            for item in items:
                albumList.append(item)

        return albumList

    #获取单个相册的相片
    def __getPicList(self, album, mm):
        lists = []
        baseUrl = self.__albumDetailUrl.replace(':userId', str(mm['userId'])).replace(':albumId',str(album))
        indexUrl = baseUrl.replace(':page','1')
        totalPage = self.__getPicPage(indexUrl)
        if totalPage is None:
            return None
        pages = range(1, int(totalPage)+1)
        for page in pages:
            url = baseUrl.replace(':page', '1')
            res = self.__getContents(url, isjson=True)

            if res is not None and res['isError']=='0':
                for pic in res['picList']:
                    lists.append('http:'+pic['picUrl'])
            else:
                print "获取结果失败，地址："+url

        return lists

    #获取单个相册照片列表的总页数
    def __getPicPage(self, indexUrl):
        albuminfo = self.__getContents(indexUrl, encoding='gbk', isjson=True)
        if albuminfo is None:
            print '获取相册照片失败0，照片地址：'+indexUrl
            return None

        if albuminfo['isError'] != '0':
            print '获取相册照片失败1，照片地址：'+indexUrl
            return None
        totalPage = int(albuminfo['totalPage'])

        return totalPage

    #下载保存单个相册的照片,album表示相册id
    def __savePics(self, album, mm):
        print "正在保存"+mm['realName'].encode('utf-8')+'的相册，相册id为：'+album.encode('utf-8')
        pics =  self.__getPicList(album, mm)
        if pics is None:
            return
        index = 1
        for pic in pics:
            print "正在保存"+mm['realName'].encode('utf-8')+'的相册，相片地址为：'+pic.encode('utf-8')
            if index % self.sleepPicCount == 0:
                print "休息一秒"
                time.sleep(1)
            if index >= self.picLimit:
                print mm["realName"].encode('utf-8') + ":已经保存"+str(self.picLimit)+"张辣"
                return
                
            saveDir = self.savePath + mm['realName'].encode('utf-8') + '/img'
            self.__mkdir(saveDir)
            fileName = saveDir + '/'+str(index)+'.jpg'
            self.__saveImg(pic, fileName)

            index +=1

    #获取相册的总页数
    def __getAlbumListPage(self, url):
        contents = self.__getContents(url)
        if contents:
            pattern = re.compile('id="J_Totalpage" value="(.*?)"', re.S)
            return re.search(pattern, contents).group(1)
        else:
            return None

    #保存MM的基本信息至本地的text文件夹
    def __saveMM(self, mm):
        print '正在保存'+mm['realName'].encode('utf-8')+'的信息'
        saveDir = self.savePath + mm['realName'] + '/text'
        self.__mkdir(saveDir)
        fileName = saveDir + '/info.txt'
        personUrl = self.__personUrl.replace(':userId', str(mm['userId']))
        contents = "姓名:%s
城市:%s
体重：%s
身高：%s
喜欢:%s
个人主页:%s
"%(mm['realName'].encode('utf-8'),mm['city'].encode('utf-8'),str(mm['weight']).encode('utf-8'),str(mm['height']).encode('utf-8'),str(mm['totalFavorNum']).encode('utf-8'),personUrl.encode('utf-8'))
        self.__saveTxtFile(contents, fileName)

    ##保存MM的头像到本地img文件夹
    def __saveMMAvatar(self, mm):
        print '正在保存'+mm['realName'].encode('utf-8')+'的头像'        
        saveDir = self.savePath + mm['realName'] + '/img'
        self.__mkdir(saveDir)
        fileName = saveDir + '/avatar.jpg'
        imgUrl = 'http:'+mm['avatarUrl']+'_240x240xz.jpg'  #获取小图
        self.__saveImg(imgUrl, fileName)

    #写入文本文件
    def __saveTxtFile(self, contents, fileName):
        handler = open(fileName, 'w')
        handler.write(contents)
        handler.close()

    #写入图片
    def __saveImg(self, imgUrl, fileName):
        contents = self.__getContents(imgUrl)
        if contents:
            handler = open(fileName, 'wb')
            handler.write(contents)
            handler.close()
        else:
            print '获取图片失败，图片地址:'+imgUrl.encode('utf-8')

    #创建存放图片或者文本文件的文件夹
    def __mkdir(self, saveDir):
        if os.path.exists(saveDir):
            return False
        else:
            os.makedirs(saveDir)
            return True

    #主入口方法
    def start(self, startPage, endPage):
        pages = range(startPage, endPage+1)
        for i in pages:
            mmlist = self.__getMMList(i)

            if not mmlist:
                print "第%s页无数据
"%(str(i))
                break
            for mm in mmlist:
                self.__saveMM(mm)
                self.__saveMMAvatar(mm)
                albumList = self.__getAlbumList(mm)

                for album in albumList:
                    self.__savePics(album, mm)

if  __name__ == '__main__':
    mmspider = MMSpider()
    mmspider.start(2, 3)

 效果：

保存的图片：

 保存的文本内容：

源码链接：https://github.com/yunshu2009/pythonspiders/blob/master/Taobaomm/Taobaomm.py