标题：写一个爬虫程序[原创] 出处：运维进行时 时间：Mon, 08 Mar 2010 10:33:42 +0000 作者：刘天斯 地址：https://blog.liuts.com/post/182/ 内容： 写爬虫是一项复杂、枯噪、反复的工作，考虑的问题包括采集效率、链路异常处理、数据质量(与站点编码规范关系很大)等。整理自己写一个爬虫程序，单台服务器可以启用1~8个实例同时采集，然后将数据入库。 #-*- coding:utf-8 -*- #!/usr/local/bin/python import sys, time, os,string import mechanize import urlparse from BeautifulSoup import BeautifulSoup import re import MySQLdb import logging import cgi from optparse import OptionParser #----------------------------------------------------------------------------# # Name: TySpider.py # # Purpose: WebSite Spider Module # # Author: 刘天斯 # # Email: liutiansi@gamil.com # # Created: 2010/02/16 # # Copyright: (c) 2010 # #----------------------------------------------------------------------------# """ |-------------------------------------------------------------------------- | 定义 loging class; |-------------------------------------------------------------------------- | | 功能：记录系统相关日志信息。 | | """ class Pubclilog(): def __init__(self): self.logfile = 'website_log.txt' def iniLog(self): logger = logging.getLogger() filehandler = logging.FileHandler(self.logfile) streamhandler = logging.StreamHandler() fmt = logging.Formatter('%(asctime)s, %(funcName)s, %(message)s') logger.setLevel(logging.DEBUG) logger.addHandler(filehandler) logger.addHandler(streamhandler) return [logger,filehandler] """ |-------------------------------------------------------------------------- | 定义 tySpider class; |-------------------------------------------------------------------------- | | 功能：抓取分类、标题等信息 | | """ class BaseTySpider: #初始化相关成员方法 def __init__(self,X,log_switch): #数据库连接 self.conn = MySQLdb.connect(db='dbname',host='192.168.0.10', user='dbuser',passwd='SDFlkj934y5jsdgfjh435',charset='utf8') #分类及标题页面Community self.CLASS_URL = 'http://test.abc.com/aa/CommTopicsPage?' #发表回复页 self.Content_URL = 'http://test.bac.com/aa/CommMsgsPage?' #开始comm值 self.X=X #当前comm id取模，方面平均到表 self.mod=self.X%5 #Community文件下载页 self.body="" #self.bodySoup对象 self.soup=None #发表回复页下载内容变量 self.Contentbody="" #发表回复页内容self.ContentbodySoup对象 self.Contentsoup=None #日志开关 self.log_switch=log_switch #======================获取名称及分类方法========================== def _SpiderClass(self,nextpage=None): if nextpage==None: FIXED_QUERY = 'cmm='+str(self.X) else: FIXED_QUERY = nextpage[1:] try: rd = mechanize.Browser() rd.addheaders = [("User-agent", "Tianya/2010 (compatible; MSIE 6.0;Windows NT 5.1)")] rd.open(self.CLASS_URL + FIXED_QUERY) self.body=rd.response().read() #rd=mechanize.Request(self.CLASS_URL + FIXED_QUERY) #response = mechanize.urlopen(rd) #self.body=response.read() except Exception,e: if self.log_switch=="on": logapp=Pubclilog() logger,hdlr = logapp.iniLog() logger.info(self.CLASS_URL + FIXED_QUERY+str(e)) hdlr.flush() logger.removeHandler(hdlr) return self.soup = BeautifulSoup(self.body) NextPageObj= self.soup("a", {'class' : re.compile("fs-paging-item fs-paging-next")}) self.cursor = self.conn.cursor() if nextpage==None: try: Ttag=str(self.soup.table) #print Ttag """ ------------------分析结构体----------------- Dunhill 中国 » 人民 """ soupTable=BeautifulSoup(Ttag) #定位到第一个h1标签 tableh1 = soupTable("h1") #print self.X #print "Name:"+tableh1[0].string.strip().encode('utf-8') #处理无类型的 try: #定位到表格中符合规则“^TopByCategory”A链接块，tablea[0]为第一个符合条件的连接文字，tablea[1]... tablea = soupTable("a", {'href' : re.compile("^TopByCategory")}) if tablea[0].string.strip()=="": pass #print "BigCLass:"+tablea[0].string.strip().encode('utf-8') #print "SubClass:"+tablea[1].string.strip().encode('utf-8') except Exception,e: if self.log_switch=="on": logapp=Pubclilog() logger,hdlr = logapp.iniLog() logger.info("[noClassInfo]"+str(self.X)+str(e)) hdlr.flush() logger.removeHandler(hdlr) self.cursor.execute("insert into baname"+str(self.mod)+" values('%d','%d','%s')" %(self.X,-1,tableh1[0].string.strip().encode('utf-8'))) self.conn.commit() self._SpiderTitle() if NextPageObj: NextPageURL=NextPageObj[0]['href'] self._SpiderClass(NextPageURL) return else: return #获取链接二对象的href值 classlink=tablea[1]['href'] par_dict=cgi.parse_qs(urlparse.urlparse(classlink).query) #print "CID:"+par_dict["cid"][0] #print "SubCID:"+par_dict["subcid"][0] #print "---------------------------------------" #插入数据库 self.cursor.execute("insert into class values('%d','%s')" %(int(par_dict["cid"][0]),tablea[0].string.strip().encode('utf-8'))) self.cursor.execute("insert into subclass values('%d','%d','%s')" %(int(par_dict["subcid"][0]),int(par_dict["cid"][0]),tablea[1].string.strip().encode('utf-8'))) self.cursor.execute("insert into baname"+str(self.mod)+" values('%d','%d','%s')" %(self.X,int(par_dict["subcid"][0]),tableh1[0].string.strip().encode('utf-8'))) self.conn.commit() self._SpiderTitle() if NextPageObj: NextPageURL=NextPageObj[0]['href'] self._SpiderClass(NextPageURL) self.body=None self.soup=None Ttag=None soupTable=None table=None table1=None classlink=None par_dict=None except Exception,e: if self.log_switch=="on": logapp=Pubclilog() logger,hdlr = logapp.iniLog() logger.info("[ClassInfo]"+str(self.X)+str(e)) hdlr.flush() logger.removeHandler(hdlr) else: self._SpiderTitle() if NextPageObj: NextPageURL=NextPageObj[0]['href'] self._SpiderClass(NextPageURL) #====================获取标题方法========================= def _SpiderTitle(self): #查找标题表格对象(table) soupTitleTable=self.soup("table", {'class' : "fs-topic-list"}) #查找标题行对象(tr) TitleTr = soupTitleTable[0]("tr", {'onmouseover' : re.compile("^this\.className='fs-row-hover'")}) """ -----------分析结构体-------------- 【新人报到】欢迎美国人民加入 0 / 12 中国人 2-14 """ for CurrTr in TitleTr: try: #初始化置顶及精华状态 Title_starred='N' Title_sticky='N' #获取当前记录的BeautifulSoup对象 soupCurrTr=BeautifulSoup(str(CurrTr)) #BeautifulSoup分析HTML有误，只能通过span的标志数来获取贴子状态，会存在一定误差 #如只有精华时也会当成置顶来处理。 TitleStatus=soupCurrTr("span", {'title' : ""}) TitlePhotoViewer=soupCurrTr("a", {'href' : re.compile("^PhotoViewer")}) if TitlePhotoViewer.__len__()==1: TitlePhotoViewerBool=0 else: TitlePhotoViewerBool=1 if TitleStatus.__len__()==3-TitlePhotoViewerBool: Title_starred='Y' Title_sticky='Y' elif TitleStatus.__len__()==2-TitlePhotoViewerBool: Title_sticky='Y' #获取贴子标题 Title=soupCurrTr.a.next.strip() #获取贴子ID par_dict=cgi.parse_qs(urlparse.urlparse(soupCurrTr.a['href']).query) #获取回复数及浏览器 TitleNum=soupCurrTr("td", {'class' : "fs-topic-name"}) TitleArray=string.split(str(TitleNum[0]),'\n') Title_ReplyNum=string.split(TitleArray[len(TitleArray)-4],'>')[2] Title_ViewNum=string.split(TitleArray[len(TitleArray)-2],'>')[2][:-6] #获取贴子作者 TitleAuthorObj=soupCurrTr("td", {'style' : "padding-left:4px"}) Title_Author=TitleAuthorObj[0].next.next.next.string.strip().encode('utf-8') #获取回复时间 TitleTime=soupCurrTr("td", {'class' : re.compile("^fs-topic-last-mdfy fs-meta")}) """ print "X:"+str(self.X) print "Title_starred:"+Title_starred print "Title_sticky:"+Title_sticky print "Title:"+Title #获取贴子内容连接URL print "Title_link:"+soupCurrTr.a['href'] print "CID:"+par_dict["tid"][0] print "Title_ReplyNum:"+Title_ReplyNum print "Title_ViewNum:"+Title_ViewNum print "Title_Author:"+Title_Author print "TitleTime:"+TitleTime[0].string.strip().encode('utf-8') """ #入库 self.cursor.execute("insert into Title"+str(self.mod)+" values('%s','%d','%s','%d','%d','%s','%s','%s','%s')" %(par_dict["tid"][0], \ self.X,Title,int(Title_ReplyNum),int(Title_ViewNum),Title_starred,Title_sticky, \ Title_Author.decode('utf-8'),TitleTime[0].string.strip().encode('utf-8'))) self.conn.commit() self._SpiderContent(par_dict["tid"][0]) except Exception,e: if self.log_switch=="on": logapp=Pubclilog() logger,hdlr = logapp.iniLog() logger.info("[Title]"+str(self.X)+'-'+par_dict["tid"][0]+'-'+str(e)) hdlr.flush() logger.removeHandler(hdlr) #======================获取发表及回复方法======================= def _SpiderContent(self,ID,nextpage=None): if nextpage==None: FIXED_QUERY = 'cmm='+str(self.X)+'&tid='+ID+'&ref=regulartopics' else: FIXED_QUERY = nextpage[9:] rd = mechanize.Browser() rd.addheaders = [("User-agent", "Tianya/2010 (compatible; MSIE 6.0;Windows NT 5.1)")] rd.open(self.Content_URL + FIXED_QUERY) self.Contentbody=rd.response().read() #rd=mechanize.Request(self.Content_URL + FIXED_QUERY) #response = mechanize.urlopen(rd) #self.Contentbody=response.read() self.Contentsoup = BeautifulSoup(self.Contentbody) NextPageObj= self.Contentsoup("a", {'class' : re.compile("fs-paging-item fs-paging-next")}) try: Tdiv=self.Contentsoup("div", {'class' : "fs-user-action"}) i=0 for Currdiv in Tdiv: if i==0: Ctype='Y' else: Ctype='N' #发表时间 soupCurrdiv=BeautifulSoup(str(Currdiv)) PosttimeObj=soupCurrdiv("span", {'class' : "fs-meta"}) Posttime=PosttimeObj[0].next[1:] Posttime=Posttime[0:-3] #IP地址 IPObj=soupCurrdiv("a", {'href' : re.compile("CommMsgAddress")}) if IPObj: IP=IPObj[0].next.strip() else: IP='' #发表／回复内容 ContentObj=soupCurrdiv("div", {'class' :"fs-user-action-body"}) Content=ContentObj[0].renderContents().strip() """ print "ID:"+str(self.X) print "ID:"+ID print "Ctype:"+Ctype print "POSTTIME:"+Posttime print "IP:"+IP print "Content:"+Content """ self.cursor.execute("insert into Content"+str(self.mod)+" values('%s','%d','%s','%s','%s','%s')" %(ID,self.X,Ctype,Posttime,IP,Content.decode('utf-8'))) self.conn.commit() i+=1 except Exception,e: if self.log_switch=="on": logapp=Pubclilog() logger,hdlr = logapp.iniLog() logger.info("[Content]"+str(self.X)+'-'+ID+'-'+str(e)) hdlr.flush() logger.removeHandler(hdlr) #如“下一页”有链接刚继续遍历 if NextPageObj: NextPageURL=NextPageObj[0]['href'] self._SpiderContent(ID,NextPageURL) def __del__(self): try: self.cursor.close() self.conn.close() except Exception,e: pass #遍历comm范围 def initapp(StartValue,EndValue,log_switch): for x in range(StartValue,EndValue): app=BaseTySpider(x,log_switch) app._SpiderClass() app=None if __name__ == "__main__": #定义命令行参数 MSG_USAGE = "TySpider.py [ -s StartNumber EndNumber ] -l [on|off] [-v][-h]" parser = OptionParser(MSG_USAGE) parser.add_option("-s", "--set", nargs=2,action="store", dest="comm_value", type="int", default=False, help="配置名称ID值范围。".decode('utf-8')) parser.add_option("-l", "--log", action="store", dest="log_switch", type="string", default="on", help="错误日志开关".decode('utf-8')) parser.add_option("-v","--version", action="store_true", dest="verbose", help="显示版本信息".decode('utf-8')) opts, args = parser.parse_args() if opts.comm_value: if opts.comm_value[0]>opts.comm_value[1]: print "终止值比起始值还小？" exit(); if opts.log_switch=="on": log_switch="on" else: log_switch="off" initapp(opts.comm_value[0],opts.comm_value[1],log_switch) exit(); if opts.verbose: print "WebSite Scider V1.0 beta." exit; 如大家有什么疑问或感兴趣的话题可以通过weibo与我交流：http://t.qq.com/yorkoliu Generated by Bo-blog 2.1.1 Release