一、介绍
本例子用Selenium +phantomjs爬取中文网站总排名（http://top.chinaz.com/all/index.html，http://top.chinaz.com/hangye/index.html）的信息
二、网站信息
三、数据抓取
针对上面的网站信息，来进行抓取
1、首先抓取信息列表
抓取代码：Elements = doc('li[class^="clearfix"]') 类似信息
2、网站名称,域名，网址
netElement = element.find('div[class="CentTxt"]').find('h3').find('a')
netName = netElement.attr('title').encode('utf8')
netUrl = 'http://top.chinaz.com' + netElement.attr('href').encode('utf8')
domainName = element.find('div[class="CentTxt"]').find('h3').find('span').text().encode('utf8')
3、Alexa周排名,反链数,网站描述
netinfo = element.find('div[class="RtCPart clearfix"]').text().encode('utf8')
4、排名，得分
netOrder = element.find('div[class="RtCRateCent"]').find('strong').text().encode('utf8')
netScore = element.find('div[class="RtCRateCent"]').find('span').text().encode('utf8')
四、具体思路
对于抓取分类网站排名信息思路如：
1、首先抓取一级分类名称（包括拼音和中文名称），分类对应的url， 该分类数据的总页数，保存到数据表category
2、根据分类数据的总页数，计算出该分类的所有网页的url， 并把 category, cncategory, url, index,iscrawl(是否抓取，默认是0：未抓取，1：已抓取) 保存至数据表categoryall中
3、通过获取categoryall表中每种分类的url，开始抓取每页数据，每页抓取完成后修改iscrawl =1，避免重复抓取。抓取每个数据后把相关数据保存到categorydata表中，
在保存之前通过 netname,domainname,category三个字段来判断是否有重复数据，避免保存重复。
3、二级分类网站排名信息同上。
五、实现代码
# coding=utf-8 import os import re from selenium import webdriver from datetime import datetime,timedelta import selenium.webdriver.support.ui as ui from selenium.common.exceptions import TimeoutException import time from pyquery import PyQuery as pq import re import mongoDB import hyData class hyTotal: def __init__(self): #通过配置文件获取IEDriverServer.exe路径 # IEDriverServer ='C:Program FilesInternet ExplorerIEDriverServer.exe' # self.driver = webdriver.Ie(IEDriverServer) # self.driver.maximize_window() self.driver = webdriver.PhantomJS(service_args=['--load-images=false']) self.driver.set_page_load_timeout(10) self.driver.maximize_window() self.db = mongoDB.mongoDbBase() def saveurls(self): urls = [] urls.append('http://top.chinaz.com/all/index.html') for i in range(2, 1888): urls.append('http://top.chinaz.com/all/index_' + str(i) + '.html') self.db.SaveTotalUrls(urls) def WriteLog(self, message,date,dir): fileName = os.path.join(os.getcwd(), 'hangye/total/' + date + '-' + dir + '.txt') with open(fileName, 'a') as f: f.write(message) def CatchData(self): urlIndex = 0 urls = self.db.GetTotalUrls() # {'_id': 0, 'url': 1,'category':1,'index':1} lst = [] for u in urls: url = u['url'] urlIndex = u['index'] lst.append('{0},{1}'.format(url,urlIndex)) urls.close() for u in lst: url = u.split(',')[0] urlIndex = u.split(',')[1] try: self.driver.get(url) time.sleep(2) selenium_html = self.driver.execute_script("return document.documentElement.outerHTML") doc = pq(selenium_html) # Elements = doc("div[@id='text_box_0']/dl/dd") Elements = doc('li[class^="clearfix"]') message = '' # for element in Elements: data_list = [] for element in Elements.items(): hyobj = hyData.hyData() # column1Element = element.find('div[@class="text"]/h3/a') # 网站名称,域名，网址 # netElement = element.find_element_by_xpath("//div[@class='CentTxt']/h3/a") netElement = element.find('div[class="CentTxt"]').find('h3').find('a') # netName = nextElement.get_attribute('title').encode('utf8') netName = netElement.attr('title').encode('utf8').replace(' ','').replace(',','，') netUrl = 'http://top.chinaz.com' + netElement.attr('href').encode('utf8') # domainName = element.find_element_by_xpath('//div[class="CentTxt"]/h3/span').text.encode('utf8') domainName = element.find('div[class="CentTxt"]').find('h3').find('span').text().encode('utf8') # Alexa周排名,反链数,网站描述 netinfo = element.find('div[class="RtCPart clearfix"]').text().encode('utf8') pattern = re.compile("d+") ts = pattern.findall(netinfo) alexaOrder = '' antichainCount = '' if ts and len(ts) == 2: alexaOrder = ts[0] antichainCount = ts[1] netDescription = element.find('p[class="RtCInfo"]').text().encode('utf8') netDescription = netDescription.replace('网站简介：', '').replace(',', '，').replace(' ', '') # 排名，得分 netOrder = element.find('div[class="RtCRateCent"]').find('strong').text().encode('utf8') netScore = element.find('div[class="RtCRateCent"]').find('span').text().encode('utf8') netScore = netScore.replace('得分:', '') hyobj.setNetName(netName) hyobj.setDomainName(domainName) hyobj.setNetRank(netOrder) hyobj.setNetScore(netScore) # print alexaOrder hyobj.setAlexaRank(alexaOrder) hyobj.setAntichainCount(antichainCount) hyobj.setNetUrl(netUrl) hyobj.setNetDescription(netDescription) hyobj.setPageIndex(urlIndex) data_list.append(hyobj) self.db.SaveTotalData(data_list) self.db.SetTotalState(url) print url except TimeoutException,e: print 'timeout url: '+url self.driver.close() self.driver.quit() def exportHyTotal(self): dataRows = self.db.GetTotalData() for dataRow in dataRows: pageIndex = int(dataRow['pageIndex'].encode('utf8')) netname = dataRow['netname'].encode('utf8') domainname = dataRow['domainname'].encode('utf8') netrank = dataRow['netrank'] netscore = dataRow['netscore'] alexarank = dataRow['alexarank'] antichaincount = dataRow['antichaincount'] neturl = dataRow['neturl'].encode('utf8') netdescription = dataRow['netdescription'].encode('utf8') # 网站名称，网站域名，排名，得分，alexa排名，反链数,网站简介,网站类型 message = ' {0},{1},{2},{3},{4},{5},{6},{7}'.format(netname, domainname, netrank, netscore, alexarank, antichaincount, neturl, netdescription) date = time.strftime('%Y-%m-%d') dir = str((pageIndex/50+1)*50) self.WriteLog(message,date,dir) obj = hyTotal() # obj.saveurls() # obj.CatchData() obj.exportHyTotal()
# coding=utf-8 import os from pymongo import MongoClient from pymongo import ASCENDING, DESCENDING import codecs import time import columnData import datetime import re import hyData import categoryItem class mongoDbBase: # def __init__(self, databaseIp = '127.0.0.1',databasePort = 27017,user = "ott",password= "ott", mongodbName='OTT_DB'): def __init__(self, connstr='mongodb://ott:ott@127.0.0.1:27017/', mongodbName='OTT'): # client = MongoClient(connstr) # self.db = client[mongodbName] client = MongoClient('127.0.0.1', 27017) self.db = client.OTT self.db.authenticate('ott', 'ott') def SaveTotalUrls(self, urls): # self.db.hytotalurls.drop() index = 1 lst = [] for url in urls: dictM = {'url': url, 'iscrawl': 0, 'index': index} lst.append(dictM) index += 1 self.db.totalurls.insert(lst) def SetTotalState(self,url): self.db.totalurls.update({'url': url}, {'$set': {'iscrawl': 1}}) def GetTotalUrls(self): urls = self.db.totalurls.find({'iscrawl': 0}, {'_id': 0, 'url': 1, 'index': 1}).sort('index',ASCENDING) return urls def SaveTotalData(self, hydata_list): lst = [] for hyobj in hydata_list: count = self.db.totaldata.find( {'netname': hyobj.getNetName(), 'domainname': hyobj.getDomainName()}).count() if count == 0: dictM = {'netname': hyobj.getNetName(), 'domainname': hyobj.getDomainName(), 'netrank': hyobj.getNetRank(), 'netscore': hyobj.getNetScore(), 'alexarank': hyobj.getAlexaRank(), 'antichaincount': hyobj.getAntichainCount(), 'neturl': hyobj.getNetUrl(), 'netdescription': hyobj.getNetDescription(), 'pageIndex': hyobj.getPageIndex()} lst.append(dictM) self.db.totaldata.insert(lst) def GetTotalData(self): rows = self.db.totaldata.find({},{'_id': 0}).sort('netrank',ASCENDING) return rows def SaveCategory(self, category,cncategory,url): # self.db.category.drop() dictM = {'category': category, 'cncategory': cncategory, 'url': url} self.db.category.insert(dictM) def GetCategory(self): categorys =self.db.category.find({},{'_id':0,'category':1,'cncategory':1,'url':1}) return categorys def UpdateCategoryPageCount(self, category,count): self.db.category.update({"category":category},{'$set':{'count':count}}) def GetCategories(self): categorys =self.db.category.find({},{'_id':0,'category':1,'url':1,'count':1,'cncategory':1}).sort('index',ASCENDING) return categorys def SaveCategoryUrls(self,categoryItem_list): # self.db.categoryall.drop() lst = [] for citem in categoryItem_list: dictM ={'category':citem.getCategroy(),'cncategory':citem.getCnCategroy(),'url':citem.getUrl(),'index':citem.getUrlIndex(),'iscrawl':0} lst.append(dictM) self.db.categoryall.insert(lst) def GetCategoryUrl(self): categorys = self.db.categoryall.find({'iscrawl':0}, {'_id': 0, 'url': 1, 'index': 1,'category':1,'cncategory':1}).sort( 'index', ASCENDING) return categorys def SetCategoryState(self,url): self.db.categoryall.update({'url': url}, {'$set': {'iscrawl': 1}}) def SaveCategoryData(self, category_list): lst = [] for hyobj in category_list: count = self.db.categorydata.find( {'netname': hyobj.getNetName(), 'domainname': hyobj.getDomainName(), 'category': hyobj.getCategory()}).count() if count == 0: dictM = {'netname': hyobj.getNetName(), 'domainname': hyobj.getDomainName(), 'netrank': hyobj.getNetRank(), 'netscore': hyobj.getNetScore(), 'alexarank': hyobj.getAlexaRank(), 'antichaincount': hyobj.getAntichainCount(), 'neturl': hyobj.getNetUrl(), 'netdescription': hyobj.getNetDescription(), 'category': hyobj.getCategory(), 'cncategory': hyobj.getCnCategory()} lst.append(dictM) if len(lst)>0: self.db.categorydata.insert(lst) def GetCategoryData(self): rows = self.db.categorydata.find({}, {'_id': 0}).sort('category',ASCENDING) return rows def SaveSubCategory(self, category_list): # self.db.subcategory.drop() lst = [] for sub in category_list: category = sub.split(',')[0] cncategory = sub.split(',')[1] subcategory = sub.split(',')[2] subcncategory = sub.split(',')[3] url = sub.split(',')[4] dictM = {'category': category, 'cncategory': cncategory,'subcategory': subcategory, 'subcncategory': subcncategory, 'url': url,'count':0} lst.append(dictM) self.db.subcategory.insert(lst) def GetSubCategory(self): categorys =self.db.subcategory.find({'count':0},{'_id':0,'category':1,'url':1}) return categorys def UpdateSubCategoryPageCount(self, url,count): self.db.subcategory.update({"url":url},{'$set':{'count':count}}) def GetSubCategories(self): categorys =self.db.subcategory.find({},{'_id':0,'category':1,'cncategory':1,'subcategory':1,'subcncategory':1,'url':1,'count':1}) return categorys def SaveSubCategoryUrls(self,categoryItem_list): # self.db.subcategoryall.drop() lst = [] for citem in categoryItem_list: dictM ={'category':citem.getCategroy(),'cncategory':citem.getCnCategroy(),'url':citem.getUrl(), 'index':citem.getUrlIndex(),'iscrawl':0,'subcategory':citem.getSubCategroy(),'subcncategory':citem.getSubCnCategory()} lst.append(dictM) self.db.subcategoryall.insert(lst) def GetSubCategoryUrl(self): categorys = self.db.subcategoryall.find({'iscrawl': 0}, {'_id': 0, 'url': 1, 'index': 1, 'category':1,'subcategory': 1, 'cncategory':1,'subcncategory': 1}).sort( 'index', ASCENDING) return categorys def SaveSubCategoryData(self, subcategory_list): lst = [] for hyobj in subcategory_list: count = self.db.subcategorydata.find( {'netname': hyobj.getNetName(), 'domainname': hyobj.getDomainName(), 'category': hyobj.getCategory()}).count() if count == 0: dictM = {'netname': hyobj.getNetName(), 'domainname': hyobj.getDomainName(), 'netrank': hyobj.getNetRank(), 'netscore': hyobj.getNetScore(), 'alexarank': hyobj.getAlexaRank(), 'antichaincount': hyobj.getAntichainCount(), 'neturl': hyobj.getNetUrl(), 'netdescription': hyobj.getNetDescription(), 'category': hyobj.getCategory(), 'cncategory': hyobj.getCnCategory() } lst.append(dictM) # 批量插入 if len(lst) > 0: self.db.subcategorydata.insert(lst) def SetSubCategoryState(self, url): self.db.subcategoryall.update({'url': url}, {'$set': {'iscrawl': 1}}) def GetSubCategoryData(self): # [(‘age‘，pymongo.DESCENDING),('name',pymongo.ASCENDING)] rows = self.db.subcategorydata.find({}, {'_id': 0}).sort([('category',ASCENDING),('netrank',ASCENDING)]) return rows # d = mongoDbBase() # urls = [] # urls.append('abc') # # d.SaveUrls(urls) # d.SetUrlCrawlState(urls)
分类抓取实现代码 （mongodb）
# coding=utf-8 import os from pymongo import MongoClient from pymongo import ASCENDING, DESCENDING import codecs import time import columnData import datetime import re import hyData import categoryItem class mongoDbBase: # def __init__(self, databaseIp = '127.0.0.1',databasePort = 27017,user = "ott",password= "ott", mongodbName='OTT_DB'): def __init__(self, connstr='mongodb://ott:ott@127.0.0.1:27017/', mongodbName='OTT'): # client = MongoClient(connstr) # self.db = client[mongodbName] client = MongoClient('127.0.0.1', 27017) self.db = client.OTT self.db.authenticate('ott', 'ott') def SaveHyTotalUrls(self, urls): # self.db.hytotalurls.drop() index = 0 for url in urls: dictM = {'url': url, 'iscrawl': '0', 'index': index + 1} self.db.hytotalurls.insert(dictM) index += 1 def SetHyTotalState(self,url): self.db.hytotalurls.update({'url': url}, {'$set': {'iscrawl': '1'}}) def GetHyTotalUrls(self): urls = self.db.hytotalurls.find({'iscrawl': '0'}, {'_id': 0, 'url': 1, 'index': 1}).sort('index',ASCENDING) return urls def SaveHyTotalData(self, hyobj): count = self.db.hytotaldata.find({'netname': hyobj.getNetName(),'domainname':hyobj.getDomainName()}).count() if count == 0: dictM = {'netname': hyobj.getNetName(), 'domainname': hyobj.getDomainName(), 'netrank': hyobj.getNetRank(), 'netscore': hyobj.getNetScore(), 'alexarank': hyobj.getAlexaRank(), 'antichaincount': hyobj.getAntichainCount(), 'neturl': hyobj.getNetUrl(), 'netdescription': hyobj.getNetDescription(), 'pageIndex':hyobj.getPageIndex()} self.db.hytotaldata.insert(dictM) def GetHyTotalData(self): rows = self.db.hytotaldata.find({},{'_id': 0}).sort('netrank',ASCENDING) return rows def SaveHyCategoryUrls(self, urls,categories): index = 0 seIndex = 0 for url in urls: category = categories[seIndex] count = self.db.hyurls.find({'url': url}).count() if count == 0: dictM = {'url': url, 'iscrawl': '0', 'category': category, 'index': index+1} self.db.hyurls.insert(dictM) index += 1 seIndex += 1 def GetHyCategoryUrls(self): urls = self.db.hyurls.find({'iscrawl': '0'}, {'_id': 0, 'url': 1,'category':1,'index':1}).sort('index',ASCENDING) return urls def SetHyUrlCrawlState(self,url): self.db.hyurls.update({'url': url}, {'$set': {'iscrawl': '1'}}) def SaveHyCategoryData(self, hyobj): count = self.db.hydata.find({'netname': hyobj.getNetName(),'domainname':hyobj.getDomainName()}).count() if count == 0: dictM = {'netname': hyobj.getNetName(), 'domainname': hyobj.getDomainName(), 'netrank': hyobj.getNetRank(), 'netscore': hyobj.getNetScore(), 'alexarank': hyobj.getAlexaRank(), 'antichaincount': hyobj.getAntichainCount(), 'neturl': hyobj.getNetUrl(), 'netdescription': hyobj.getNetDescription(), 'category': hyobj.getCategory()} self.db.hydata.insert(dictM) def GetHyCategoryData(self,category): rows = self.db.hydata.find({'category':category}, {'_id': 0}).sort('netrank',ASCENDING) return rows def SaveCategory(self, category,cncategory,url): # self.db.category.drop() dictM = {'category': category, 'cncategory': cncategory, 'url': url} self.db.category.insert(dictM) def GetCategory(self): categorys =self.db.category.find({},{'_id':0,'category':1,'cncategory':1,'url':1}) return categorys def UpdateCategoryPageCount(self, category,count): self.db.category.update({"category":category},{'$set':{'count':count}}) def GetCategories(self): categorys =self.db.category.find({},{'_id':0,'category':1,'url':1,'count':1,'cncategory':1}).sort('index',ASCENDING) return categorys def SaveCategoryUrls(self,categoryItem_list): # self.db.categoryall.drop() lst = [] for citem in categoryItem_list: dictM ={'category':citem.getCategroy(),'cncategory':citem.getCnCategroy(),'url':citem.getUrl(),'index':citem.getUrlIndex(),'iscrawl':0} lst.append(dictM) self.db.categoryall.insert(lst) def GetCategoryUrl(self): categorys = self.db.categoryall.find({'iscrawl':0}, {'_id': 0, 'url': 1, 'index': 1,'category':1,'cncategory':1}).sort( 'index', ASCENDING) return categorys def SetCategoryState(self,url): self.db.categoryall.update({'url': url}, {'$set': {'iscrawl': 1}}) def SaveCategoryData(self, category_list): lst = [] for hyobj in category_list: count = self.db.categorydata.find( {'netname': hyobj.getNetName(), 'domainname': hyobj.getDomainName(), 'category': hyobj.getCategory()}).count() if count == 0: dictM = {'netname': hyobj.getNetName(), 'domainname': hyobj.getDomainName(), 'netrank': hyobj.getNetRank(), 'netscore': hyobj.getNetScore(), 'alexarank': hyobj.getAlexaRank(), 'antichaincount': hyobj.getAntichainCount(), 'neturl': hyobj.getNetUrl(), 'netdescription': hyobj.getNetDescription(), 'category': hyobj.getCategory(), 'cncategory': hyobj.getCnCategory()} lst.append(dictM) if len(lst)>0: self.db.categorydata.insert(lst) def GetCategoryData(self): rows = self.db.categorydata.find({}, {'_id': 0}).sort('category',ASCENDING) return rows def SaveSubCategory(self, category_list): # self.db.subcategory.drop() lst = [] for sub in category_list: category = sub.split(',')[0] cncategory = sub.split(',')[1] subcategory = sub.split(',')[2] subcncategory = sub.split(',')[3] url = sub.split(',')[4] dictM = {'category': category, 'cncategory': cncategory,'subcategory': subcategory, 'subcncategory': subcncategory, 'url': url,'count':0} lst.append(dictM) self.db.subcategory.insert(lst) def GetSubCategory(self): categorys =self.db.subcategory.find({'count':0},{'_id':0,'category':1,'url':1}) return categorys def UpdateSubCategoryPageCount(self, url,count): self.db.subcategory.update({"url":url},{'$set':{'count':count}}) def GetSubCategories(self): categorys =self.db.subcategory.find({},{'_id':0,'category':1,'cncategory':1,'subcategory':1,'subcncategory':1,'url':1,'count':1}) return categorys def SaveSubCategoryUrls(self,categoryItem_list): # self.db.subcategoryall.drop() lst = [] for citem in categoryItem_list: dictM ={'category':citem.getCategroy(),'cncategory':citem.getCnCategroy(),'url':citem.getUrl(), 'index':citem.getUrlIndex(),'iscrawl':0,'subcategory':citem.getSubCategroy(),'subcncategory':citem.getSubCnCategory()} lst.append(dictM) self.db.subcategoryall.insert(lst) def GetSubCategoryUrl(self): categorys = self.db.subcategoryall.find({'iscrawl': 0}, {'_id': 0, 'url': 1, 'index': 1, 'category':1,'subcategory': 1, 'cncategory':1,'subcncategory': 1}).sort( 'index', ASCENDING) return categorys def SaveSubCategoryData(self, subcategory_list): lst = [] for hyobj in subcategory_list: count = self.db.subcategorydata.find( {'netname': hyobj.getNetName(), 'domainname': hyobj.getDomainName(), 'category': hyobj.getCategory()}).count() if count == 0: dictM = {'netname': hyobj.getNetName(), 'domainname': hyobj.getDomainName(), 'netrank': hyobj.getNetRank(), 'netscore': hyobj.getNetScore(), 'alexarank': hyobj.getAlexaRank(), 'antichaincount': hyobj.getAntichainCount(), 'neturl': hyobj.getNetUrl(), 'netdescription': hyobj.getNetDescription(), 'category': hyobj.getCategory(), 'cncategory': hyobj.getCnCategory() } lst.append(dictM) # 批量插入 if len(lst) > 0: self.db.subcategorydata.insert(lst) def SetSubCategoryState(self, url): self.db.subcategoryall.update({'url': url}, {'$set': {'iscrawl': 1}}) def GetSubCategoryData(self): # [(‘age‘，pymongo.DESCENDING),('name',pymongo.ASCENDING)] rows = self.db.subcategorydata.find({}, {'_id': 0}).sort([('category',ASCENDING),('netrank',ASCENDING)]) return rows # d = mongoDbBase() # urls = [] # urls.append('abc') # # d.SaveUrls(urls) # d.SetUrlCrawlState(urls)
# coding=utf-8 import os from selenium import webdriver # from datetime import datetime,timedelta # import selenium.webdriver.support.ui as ui from selenium.common.exceptions import TimeoutException import time from pyquery import PyQuery as pq import re import mongoDB from categoryItem import categoryItem # import urllib2 import hyData class category: def __init__(self): #通过配置文件获取IEDriverServer.exe路径 # IEDriverServer ='C:Program FilesInternet ExplorerIEDriverServer.exe' # self.driver = webdriver.Ie(IEDriverServer) # self.driver.maximize_window() self.driver = webdriver.PhantomJS(service_args=['--load-images=false']) self.driver.set_page_load_timeout(10) self.driver.maximize_window() self.db = mongoDB.mongoDbBase() def WriteLog(self, message,subdir,dir): fileName = os.path.join(os.getcwd(), 'hangye/'+ subdir + '/' + dir + '.txt') fileName = unicode(fileName,'utf8') with open(fileName, 'a') as f: f.write(message) # 根据链接获取该链接下的分类及其链接 def getCategoryAndUrls(self,url): category_list = [] try: self.driver.get(url) selenium_html = self.driver.execute_script("return document.documentElement.outerHTML") doc = pq(selenium_html) Elements = doc('div[class="HeadFilter clearfix"]').find('a') for element in Elements.items(): cncategory = element.text().encode('utf8') categoryUrl = 'http://top.chinaz.com/hangye/' + element.attr('href') ts = categoryUrl.split('_') category = ts[len(ts) - 1].replace('.html', '') message = '{0},{1},{2}'.format(category,cncategory,categoryUrl) category_list.append(message) except TimeoutException,e: pass return category_list # 根据分类的链接，获取该分类总页数 # 获取当前链接的总页数 def getPageCount(self, url): pageCount = 0 try: pattern = re.compile("d+") self.driver.get(url) time.sleep(1) selenium_html = self.driver.execute_script("return document.documentElement.outerHTML") doc = pq(selenium_html) Elements = doc('div[class="ListPageWrap"]').find('a') for element in Elements.items(): temp = element.text().encode('utf8') if temp != '<' and temp != '>': ts = pattern.findall(temp) if ts and len(ts) == 1: count = int(ts[0]) if count > pageCount: pageCount = count except TimeoutException, e: pass return pageCount # 更新所有分类的总页数 def saveCategoryPageCount(self): urls = self.db.GetCategory() for u in urls: url = u['url'] category = u['category'] pageCount = self.getPageCount(url) if pageCount > 0: self.db.UpdateCategoryPageCount(category, pageCount) # 保存分类的所有url def saveCategoryAllUrls(self): categorys = self.db.GetCategories() index = 0 categoryItem_list = [] for c in categorys: url = c['url'] category = c['category'] cncategory = c['cncategory'] count = c['count'] + 1 item = categoryItem() item.setCategroy(category) item.setCnCategroy(cncategory) item.setUrl(url) item.setUrlIndex(index) categoryItem_list.append(item) index += 1 for i in range(2, count): url1 = url.replace('.html', '_' + str(i) + '.html') it = categoryItem() it.setCategroy(category) it.setCnCategroy(cncategory) it.setUrl(url1) it.setUrlIndex(index) categoryItem_list.append(it) index += 1 self.db.SaveCategoryUrls(categoryItem_list) # 保存分类信息 def saveCategory(self): # url = 'http://top.chinaz.com/hangye/index.html' # # 1、首先根据初始链接，获取下面的所有分类，分类拼音，及分类对应的链接并保存数据库 # category_list = self.getCategoryAndUrls(url) # for cs in category_list: # category = cs.split(',')[0] # cncategory = cs.split(',')[1] # categoryUrl = cs.split(',')[2] # self.db.SaveCategory(category, cncategory, categoryUrl) # 2、更新所有分类中每种分类的总页数 # self.saveCategoryPageCount() # 3、最后根据分类的总页数，保存分类的信息及链接 self.saveCategoryAllUrls() self.driver.close() self.driver.quit() # 根据url抓取分类链接 def catchCategoryByUrl(self,url,category,cncategory): categroy_list = [] try: self.driver.get(url) time.sleep(2) selenium_html = self.driver.execute_script("return document.documentElement.outerHTML") doc = pq(selenium_html) # Elements = doc("div[@id='text_box_0']/dl/dd") Elements = doc('li[class^="clearfix"]') # for element in Elements: for element in Elements.items(): hyobj = hyData.hyData() # column1Element = element.find('div[@class="text"]/h3/a') # 网站名称,域名，网址 # netElement = element.find_element_by_xpath("//div[@class='CentTxt']/h3/a") netElement = element.find('div[class="CentTxt"]').find('h3').find('a') # netName = nextElement.get_attribute('title').encode('utf8') netName = netElement.attr('title').encode('utf8').replace(' ', '').replace(',', '，') netUrl = 'http://top.chinaz.com' + netElement.attr('href').encode('utf8') # domainName = element.find_element_by_xpath('//div[class="CentTxt"]/h3/span').text.encode('utf8') domainName = element.find('div[class="CentTxt"]').find('h3').find('span').text().encode('utf8') # Alexa周排名,反链数,网站描述 netinfo = element.find('div[class="RtCPart clearfix"]').text().encode('utf8') pattern = re.compile("d+") ts = pattern.findall(netinfo) alexaOrder = '' antichainCount = '' if ts and len(ts) == 2: alexaOrder = ts[0] antichainCount = ts[1] netDescription = element.find('p[class="RtCInfo"]').text().encode('utf8') netDescription = netDescription.replace('网站简介：', '').replace(',', '，').replace(' ', '') # 排名，得分 netOrder = element.find('div[class="RtCRateCent"]').find('strong').text().encode('utf8') netScore = element.find('div[class="RtCRateCent"]').find('span').text().encode('utf8') netScore = netScore.replace('得分:', '') hyobj.setNetName(netName) hyobj.setDomainName(domainName) hyobj.setNetRank(int(netOrder)) hyobj.setNetScore(netScore) hyobj.setAlexaRank(alexaOrder) hyobj.setAntichainCount(antichainCount) hyobj.setNetUrl(netUrl) hyobj.setNetDescription(netDescription) hyobj.setCategory(category) hyobj.setCnCategory(cncategory) categroy_list.append(hyobj) except TimeoutException, e: print 'timeout url: ' + url return categroy_list # 抓取分类数据 def catchCategory(self): urls = self.db.GetCategoryUrl() lst = [] for u in urls: url = u['url'] category = u['category'].encode('utf8') cncategory = u['cncategory'].encode('utf8') lst.append('{0},{1},{2}'.format(url, category, cncategory)) urls.close() for u in lst: url = u.split(',')[0] category = u.split(',')[1] cncategory = u.split(',')[2] category_list = self.catchCategoryByUrl(url, category, cncategory) self.db.SaveCategoryData(category_list) self.db.SetCategoryState(url) print url self.driver.close() self.driver.quit() # 导出分类数据 def exportCategory(self): dataRows = self.db.GetCategoryData() self.exportData(dataRows, 'category') # 更新所有子分类的总页数 def saveSubCategoryPageCount(self): urls = self.db.GetSubCategory() index = 1 for u in urls: url = u['url'] pageCount = self.getPageCount(url) if pageCount > 0: self.db.UpdateSubCategoryPageCount(url, pageCount) print index index+=1 # 保存子分类所有url def saveSubCategoryAllUrls(self): categorys = self.db.GetSubCategories() index = 0 categoryItem_list = [] for c in categorys: url = c['url'] category = c['category'] cncategory = c['cncategory'] subcategory = c['subcategory'] subcncategory = c['subcncategory'] count = c['count'] + 1 item = categoryItem() item.setCategroy(category) item.setCnCategroy(cncategory) item.setSubCategroy(subcategory) item.setSubCnCategory(subcncategory) item.setUrl(url) item.setUrlIndex(index) categoryItem_list.append(item) index += 1 for i in range(2, count): url1 = url.replace('.html', '_' + str(i) + '.html') it = categoryItem() it.setCategroy(category) it.setCnCategroy(cncategory) it.setSubCategroy(subcategory) it.setSubCnCategory(subcncategory) it.setUrl(url1) it.setUrlIndex(index) categoryItem_list.append(it) index += 1 self.db.SaveSubCategoryUrls(categoryItem_list) # 保存子分类信息 # 保存子分类数据 def saveSubCategory(self): # cate = self.db.GetCategory() # subcategory_list = [] # # 1、首先根据初始链接，获取下面的所有子分类，子分类拼音，及子分类对应的链接并保存数据库 # for u in cate: # url = u['url'] # category = u['category'].encode('utf8') # cncategory =u['cncategory'].encode('utf8') # category_list =self.getCategoryAndUrls(url) # for subcategory in category_list: # temp = '{0},{1},{2}'.format(category,cncategory,subcategory) # subcategory_list.append(temp) # self.db.SaveSubCategory(subcategory_list) # # 2、更新所有子分类中每种分类的总页数 self.saveSubCategoryPageCount() # 3、最后根据子分类的总页数，保存子分类的信息及链接 self.saveSubCategoryAllUrls() self.driver.close() self.driver.quit() # 抓取子分类数据 # 抓取子分类数据 def catchSubCategory(self): urls = self.db.GetSubCategoryUrl() lst = [] for u in urls: url = u['url'] category = u['category'].encode('utf8') + '_' + u['subcategory'].encode('utf8') cncategory = u['cncategory'].encode('utf8') + '_' + u['subcncategory'].encode('utf8') lst.append('{0},{1},{2}'.format(url,category,cncategory)) urls.close() # pymongo.errors.CursorNotFound: Cursor not found, cursor id: 82792803897 for u in lst: url = u.split(',')[0] category = u.split(',')[1] cncategory = u.split(',')[2] category_list = self.catchCategoryByUrl(url, category, cncategory) self.db.SaveSubCategoryData(category_list) self.db.SetSubCategoryState(url) print (url) self.driver.close() self.driver.quit() def exportData(self,dataRows,subdir): count = 0 message = '' tempcategory = '' cncategory = '' for dataRow in dataRows: netname = dataRow['netname'].encode('utf8') domainname = dataRow['domainname'].encode('utf8') netrank = dataRow['netrank'] netscore = dataRow['netscore'].encode('utf8') alexarank = dataRow['alexarank'].encode('utf8') antichaincount = dataRow['antichaincount'].encode('utf8') neturl = dataRow['neturl'].encode('utf8') netdescription = dataRow['netdescription'].encode('utf8') # category = dataRow['category'].encode('utf8') cncategory = dataRow['cncategory'].encode('utf8') mess = ' {0},{1},{2},{3},{4},{5},{6},{7},{8}'.format(netname, domainname, netrank, netscore, alexarank, antichaincount, neturl, netdescription, cncategory) if tempcategory != cncategory: # 网站名称，网站域名，排名，得分，alexa排名，反链数,网站简介,网站类型 if len(message) > 0: self.WriteLog(message,subdir, tempcategory) message = mess tempcategory = cncategory else: message += mess # date = time.strftime('%Y-%m-%d') count += 1 # print count if len(message) > 0: self.WriteLog(message,subdir, tempcategory) # 导出子分类数据 def exportSubCategory(self): dataRows = self.db.GetSubCategoryData() self.exportData(dataRows,'subcategory') obj = category() # obj.saveCategory() # obj.catchCategory() obj.exportCategory() # obj.saveSubCategory() # obj.catchSubCategory() # obj.exportSubCategory()
# coding=utf-8 class hyData: def __init__(self): pass # 网站名称 def setNetName(self,netName): self.NetName = netName def getNetName(self): return self.NetName #网站域名 def setDomainName(self,domainName): self.DomainName = domainName def getDomainName(self): return self.DomainName # 网站排名 def setNetRank(self, netRank): self.NetRank= netRank def getNetRank(self): return self.NetRank # 网站得分 def setNetScore(self, netScore): self.NetScore = netScore def getNetScore(self): return self.NetScore # 栏目名称netName, domainName, netOrder, netScore, # alexaOrder, antichainCount, netUrl, # netDescription # Alexa排名 def setAlexaRank(self, alexaRank): self.AlexaRank = alexaRank def getAlexaRank(self): return self.AlexaRank #反链数 def setAntichainCount(self, antichainCount): self.AntichainCount = antichainCount def getAntichainCount(self): return self.AntichainCount # 网站链接 def setNetUrl(self, netUrl): self.NetUrl = netUrl def getNetUrl(self): return self.NetUrl #网站简介 def setNetDescription(self, netDescription): self.NetDescription = netDescription def getNetDescription(self): return self.NetDescription # 分类拼音 def setCategory(self, category): self.Category = category def getCategory(self): return self.Category # 分类中文 def setCnCategory(self, cnCategory): self.CnCategory = cnCategory def getCnCategory(self): return self.CnCategory def setPageIndex(self, pageIndex): self.PageIndex = pageIndex def getPageIndex(self): return self.PageIndex
# coding=utf-8 class categoryItem: def __init__(self): pass # 分类拼音subcncategory def setCategroy(self, categroy): self.Categroy = categroy def getCategroy(self): return self.Categroy # 分类中文名 def setCnCategroy(self, cnCategroy): self.CnCategroy = cnCategroy def getCnCategroy(self): return self.CnCategroy # 分类url def setUrl(self, url): self.Url = url def getUrl(self): return self.Url # url序号 def setUrlIndex(self, urlIndex): self.UrlIndex = urlIndex def getUrlIndex(self): return self.UrlIndex # 子分类拼音 def setSubCategroy(self, subCategroy): self.SubCategroy = subCategroy def getSubCategroy(self): return self.SubCategroy # 子分类中文名 def setSubCnCategory(self, subCnCategory): self.SubCnCategory = subCnCategory def getSubCnCategory(self): return self.SubCnCategory