网络爬虫入门——案例一：爬取百度贴吧帖子

参考资料：

Python:http://www.runoob.com/python/python-intro.html

Python爬虫系列教程：http://www.cnblogs.com/xin-xin/p/4297852.html

正则表达式：http://www.cnblogs.com/deerchao/archive/2006/08/24/zhengzhe30fengzhongjiaocheng.html

本贴目标:

1.对百度贴吧的任意帖子进行抓取

2.指定是否只抓取楼主发帖内容

3.将抓取到的内容分析并保存到文件

4.抓取帖子中出现的美图

# -*- coding: utf-8 -*-
"""
Created on Fri Apr 15 11:47:02 2016

@author: wuhan
"""
import urllib
import urllib2
import re
import time
import os


#reload(sys)
#sys.setdefaultencoding("utf-8")

class Tool:
    removeImg = re.compile('<img.*?>| {12}')
    removeAddr = re.compile('<a.*?>|</a>')
    replaceLine = re.compile('<tr>|<div>|</div>|</p>')
    replaceTD = re.compile('<td>')
    replacePara = re.compile('<p.*?>')
    replaceBR = re.compile('<br><br>|<br>')
    removeExtraTag = re.compile('<.*?>')
    
    def replace(self,x):
        x = re.sub(self.removeImg, "", x)
        x = re.sub(self.removeAddr, "", x)
        x = re.sub(self.replaceLine, "
", x)
        x = re.sub(self.replaceBR, "
", x)
        x = re.sub(self.replacePara, "
  ", x)
        x = re.sub(self.replaceTD, "	", x)
        x = re.sub(self.removeExtraTag, "", x)
        return x.strip()
        

class BDTB:
    def __init__(self, baseUrl, seeLZ, floorTag):
        self.baseURL = baseUrl
        self.seeLZ = '?see_lz=' + str(seeLZ)
        self.tool = Tool()
        self.file = None
        self.floor = 1
        self.defaultTitle = u'百度贴吧'
        self.floorTag = floorTag
        
    def getPage(self, pageNum):
        try:
            url = self.baseURL + self.seeLZ + '&pn=' + str(pageNum)
            request = urllib2.Request(url)
            response = urllib2.urlopen(request)
            return response.read().decode('utf-8')
        except urllib2.URLError, e:
            if hasattr(e, "reason"):
                print u'百度贴吧链接失败，错误原因 ：', e.reason
                return None
                
    def getTitle(self, page):
         pattern = re.compile('<h1 class="core_title_txt.*?>(.*?)</h1>',re.S)
         result = re.search(pattern, page)
         if result:
             return result.group(1).strip()
         else:
             return None
             
    def getPageNum(self, page):
        pattern = re.compile('<li class="l_reply_num.*?</span>.*?<span.*?>(.*?)</span>',re.S)
        result = re.search(pattern, page)
        if result:
            return result.group(1).strip()
        else:
            return None
    
    def getContents(self,page):
        pattern = re.compile('<div id="post_content.*?>(.*?)</div>', re.S)
        items = re.findall(pattern, page)
        contents = []
        for item in items:
            content = "
" + self.tool.replace(item) + "
"
            contents.append(content.encode('utf-8'))
        return contents
        
    def setFileTitle(self, title):
        if title is not None:
            self.file = open(title + ".txt" , "w+")
        else:
            self.file = open(self.defaultTitle + ".txt" , "w+")
            
    def writeData(self, contents):
        for item in contents:
            if self.floorTag == '1':
                floorLine = "
" + str(self.floor) + u"-----------------------------------------------------------------------------------------------------------------------------------------
"
                self.file.write(floorLine)
            self.file.write(item)
            self.floor += 1
    
    def start(self):
        indexPage = self.getPage(1)
        pageNum = self.getPageNum(indexPage)
        title = self.getTitle(indexPage)
        self.setFileTitle(title)
        if pageNum == None:
            print "URL已失效，请重试"
            return
        try:
            print "该贴子共有" + str(pageNum) + "页"
            for i in range(1, int(pageNum)+1):
                print "正在写入第" + str(i) + "页数据"
                page = self.getPage(i)
                contents = self.getContents(page)
                self.writeData(contents)
                self.getPicture(page, i)
        except IOError, e:
            print "写入异常，原因" + e.message
        finally:
            print "写入任务完成"
            
    def getPicture(self, page, PageNum):
        reg = r'<img class="BDE_Image".*?src="(.+?.jpg)'
        imgre = re.compile(reg)#可以把正则表达式编译成一个正则表达式对象
        imglist = re.findall(imgre,page)#读取html 中包含 imgre（正则表达式）的数据
        t = time.localtime(time.time())
        foldername = str(t.__getattribute__("tm_year"))+"-"+str(t.__getattribute__("tm_mon"))+"-"+str(t.__getattribute__("tm_mday"))
        picpath = 'E:\Python\ImageDownload\%s' % (foldername) #下载到的本地目录  
        if not os.path.exists(picpath):   #路径不存在时创建一个
            os.makedirs(picpath)
    
        x = 0
        for imgurl in imglist:
            target = picpath+'\%s_%s.jpg' % (PageNum, x)
            urllib.urlretrieve(imgurl, target)#直接将远程数据下载到本地
            x+=1
        
print u"请输入帖子代号"
baseURL = 'http://tieba.baidu.com/p/' + str(raw_input(u'http://tieba.baidu.com/p/'))
seeLZ = raw_input("是否只获取楼主发言，是输入1，否输入0
".decode('utf-8').encode('gbk'))
floorTag = raw_input("是否写入楼层信息，是输入1，否输入0
".decode('utf-8').encode('gbk'))
bdtb = BDTB(baseURL,seeLZ,floorTag)
bdtb.start()