Esempio n. 1
0
    def rent(self,url):
        
        self.fd['house_flag'] = 2
        request = urllib2.Request(url, None, self.header)
        response = urllib2.urlopen(request).read()
        tree = etree.HTML(response)  
        soup =BeautifulSoup(response)
        detail_mer = soup.find('ul',{'class':'info'})
        detail_mer_str =re.sub("\n|\t\r| ","",str(detail_mer))
        #print detail_mer_str
        #非个人房源 return
        #print re.search(self.agencyname_regex, response).group(1)
        if re.search(self.agencyname_regex, response):
            agencyname=re.search(self.agencyname_regex, response).group(1)
            if agencyname != '个人房源':return            
        else:
            return
                
        if re.search(self.username_regex, response):
            username=re.search(self.username_regex, response).group(1)
            self.fd['owner_name'] = username
        else:             
            self.fd['owner_name'] = None

        owner_phone = soup('img')
        print owner_phone
        self.fd['owner_phone'] = ''
        for phone in owner_phone:
            if phone['src'].find('58.com/showphone.aspx') != -1:
                self.fd['owner_phone'] = phone['src']
        #没有联系方式  return
        if not self.fd['owner_phone']:return 
        
        if soup.find('div',{"class":'other'}):
            posttime = soup.find('div',{"class":'other'}).contents[0]                            
            posttime = re.sub('\n|\r| |\t','',posttime)
            posttime = posttime.replace('发布时间:','').replace(' 浏览','')
        else:
            posttime = ''
                            
        if not posttime:
            return                            
        elif posttime.find('-') !=-1:
            s = datetime.datetime(int(posttime.split('-')[0]),int(posttime.split('-')[1],),int(posttime.split('-')[2]))
            posttime = int(time.mktime(s.timetuple()))
        elif posttime.find('分钟') !=-1:
            n = int(posttime.replace('分钟前',''))*60
            posttime = int(time.time() - n)
        elif posttime.find('小时') !=-1:
            n = int(posttime.replace('小时前',''))*60*60
            posttime = int(time.time() - n)
        self.fd['posttime'] = posttime
                            
        if (time.time() - self.fd['posttime']) > 3600*24*7: 
            return
            print "++++++++++++++++"                 
        print time.strftime('%Y %m %d', time.localtime(self.fd['posttime']))    
        
        if re.search(self.house_floor_regex, detail_mer_str):
            house_floor=re.search(self.house_floor_regex, detail_mer_str).group(1)
            self.fd['house_floor']  = house_floor
        else:
            self.fd['house_floor'] = None
            
        if re.search(self.house_topfloor_regex, detail_mer_str):
            house_topfloor=re.search(self.house_topfloor_regex, detail_mer_str).group(1)
            self.fd['house_topfloor'] = house_topfloor
        else:
            self.fd['house_topfloor'] = None   
        
        if re.search(self.house_totalarea_regex, detail_mer_str):
            house_totalarea=re.search(self.house_totalarea_regex, detail_mer_str).group(1)
            self.fd['house_totalarea'] = house_totalarea
        else:
            self.fd['house_totalarea'] = None
            
        #类型 
        self.fd['house_type'] = housetype(detail_mer_str) 
            
        self.fd['house_price'] = detail_mer.em.string  
            
        if re.search(self.house_room_regex, detail_mer_str):
            house_room=re.search(self.house_room_regex, detail_mer_str).group(1)
            self.fd['house_room'] = house_room
        else:
            self.fd['house_room'] = '0'
            
        if re.search(self.house_hall_regex, detail_mer_str):
            house_hall=re.search(self.house_hall_regex, detail_mer_str).group(1)
            self.fd['house_hall'] = house_hall
        else:
            self.fd['house_hall'] = '0'
        
        if re.search(self.house_toilet_regex, detail_mer_str):
            house_toilet=re.search(self.house_toilet_regex, detail_mer_str).group(1)
            self.fd['house_toilet'] = house_toilet
        else:
            self.fd['house_toilet'] = '0'
        
        if re.search(self.house_title_regex, response):
            house_title=re.search(self.house_title_regex, response).group(1)
            self.fd['house_title'] = house_title
        else:
            self.fd['house_title'] = ''
        
        #描述        
        detail_box = soup.find('div',{'class':'maincon'})
        if detail_box:
            house_desc = str(detail_box)
            self.fd['house_desc'] = re.sub("<.*?>|\n|\r|\t|联系我时,请说是在58同城上看到的,谢谢!","",house_desc)
        else:
            self.fd['house_desc'] = None

        #小区名
        if re.search(self.borough_name_regex, detail_mer_str):
            borough_name=re.search(self.borough_name_regex, detail_mer_str).group(1)
            self.fd['borough_name'] = re.sub("\(.*\)|<.*?>","",borough_name)
        else:
            self.fd['borough_name'] = ''
        
        #区域     
        area_box = detail_mer.find(text="区域:").parent.parent
        area_a = area_box('a')
        if area_a and len(area_a)>1:
            self.fd['cityarea'] = area_a[0].string
            self.fd['section'] = area_a[1].string
        elif area_a and len(area_a)==1:
            self.fd['cityarea'] = area_a[0].string
            self.fd['section'] = None
        else:
            self.fd['cityarea'] = None
            self.fd['section'] = None
        
        if re.search(self.house_age_regex, response):
            house_age=re.search(self.house_age_regex, response).group(1)
            self.fd['house_age'] = house_age
        else:
            self.fd['house_age'] = None
            
        #朝向
        self.fd['house_toward'] = toward(detail_mer_str)    
        self.fd['house_fitment'] = fitment(detail_mer_str)        
        self.fd['house_deposit'] = deposit(detail_mer_str)
Esempio n. 2
0
File: ganji.py Progetto: ptphp/PyLib
    def rent(self,url):
        self.fd['city'] = urlparse(url)[1].replace('.ganji.com',"")
        
        request = urllib2.Request(url, None, self.header)
        response = urllib2.urlopen(request).read()
        tree = etree.HTML(response)
        if re.search("<span class=\"city\"><a .*?>(.*?)</a>", response):
            cityname=re.search("<span class=\"city\"><a .*?>(.*?)</a>", response).group(1)
            self.fd['cityname'] = cityname
        else:
            return
        
        self.fd['house_flag'] = 2
        self.fd['house_type'] = 0
        self.fd['house_floor'] = ""
        self.fd['house_topfloor'] = "" 
        
        soup =BeautifulSoup(response)
        detail_mer = soup.find('div',{'class':'detail_mer'})
        
        #非个人房源 return
        if u"个人房源"  not in str(detail_mer):return
        
        Dname = detail_mer.find('span',{'class':'Dname'})
        if Dname:
            self.fd['owner_name'] = Dname.string
        else:
            self.fd['owner_name'] = None
            
        ganji_phone_call_class = detail_mer.find('span',{'class':'ganji_phone_call_class'})
        
        if ganji_phone_call_class:
            self.fd['owner_phone'] = ganji_phone_call_class.contents[0]
            if str(ganji_phone_call_class).find('src='):                
                self.fd['owner_phone'] = 'http://'+urlparse(url)[1]+ganji_phone_call_class.img['src']
            else:
                self.fd['owner_phone'] = None            
        else:
            self.fd['owner_phone'] = None
            
            
        #没有联系方式  return
        if not self.fd['owner_phone']:return        
        
        if re.search(self.house_totalarea_regex, response):
            house_totalarea=re.search(self.house_totalarea_regex, response).group(1)
            self.fd['house_totalarea'] = house_totalarea
        else:
            self.fd['house_totalarea'] = None
        
        if re.search(self.house_price_regex_2, response):
            house_price=re.search(self.house_price_regex_2, response).group(1)
            self.fd['house_price'] = house_price
        else:
            self.fd['house_price'] = None
    #    house_price=tree.xpath("/html/body/div[2]/div/div/ul/li/span") and tree.xpath("/html/body/div[2]/div/div/ul/li/span")[0].text.strip() or None    
    #    v['house_price'] = house_price
        
        posttime=CSSSelector('span.pub_time')(tree)!=None and CSSSelector('span.pub_time')(tree)[0].text.strip() or None 
        if posttime:
            Y=int(time.strftime('%Y', time.localtime()))
            M=int(posttime.split(' ')[0].split('-')[0])
            D=int(posttime.split(' ')[0].split('-')[1])
            s = datetime.datetime(Y,M,D,0,0)
            posttime=int(time.mktime(s.timetuple()))
            self.fd['posttime'] =posttime 
        else:
            self.fd['posttime'] =None
            
        house_title=CSSSelector("div.detail_title h1")(tree)[0] !=None and CSSSelector("div.detail_title h1")(tree)[0].text.strip() or None
        self.fd['house_title'] = house_title
        

        if re.search(self.house_room_regex, response):
            house_room=re.search(self.house_room_regex, response).group(1)
            self.fd['house_room'] = house_room
        else:
            self.fd['house_room'] = '0'
            
        if re.search(self.house_hall_regex, response):
            house_hall=re.search(self.house_hall_regex, response).group(1)
            self.fd['house_hall'] = house_hall
        else:
            self.fd['house_hall'] = '0'
        
        if re.search(self.house_toilet_regex, response):
            house_toilet=re.search(self.house_toilet_regex, response).group(1)
            self.fd['house_toilet'] = house_toilet
        else:
            self.fd['house_toilet'] = '0'

        house_title=CSSSelector("div.detail_title h1")(tree)[0] !=None and CSSSelector("div.detail_title h1")(tree)[0].text.strip() or None
        self.fd['house_title'] = house_title
        
        #描述        
        detail_box = soup.find('div',{'class':'detail_box'})
        if detail_box:
            house_desc = str(detail_box('p')[1])
            self.fd['house_desc'] = re.sub("<.*?>|\n|\r|\t|联系我时请说明是从赶集网上看到的","",house_desc)
        else:
            self.fd['house_desc'] = None

        d_i = soup.find('ul',{'class':'d_i'})        
        #小区名
        #先处理JS
        if re.search(self.xiaoqu_regex, response):
            borough_name=re.search(self.xiaoqu_regex, response).group(1)
            self.fd['borough_name'] = borough_name
            if re.search(self.address_regex, response):
                house_addr=re.search(self.address_regex, response).group(1)
                self.fd['house_addr'] = house_addr
        else:            
            if d_i.find(text="小区: "):
                borough_box = d_i.find(text="小区: ").parent        
                borough_name = borough_box.find("a")
                if borough_name:
                    self.fd['borough_name'] = borough_name.string
                else:
                    self.fd['borough_name'] = None            
                #地址
                if borough_name and borough_name.nextSibling:
                    house_addr = borough_name.nextSibling.string
                    self.fd['house_addr'] = re.sub("\(|\)| ","",house_addr)
                else:
                    self.fd['house_addr'] = None
            else:
                if re.search(self.borough_name_regex, response):
                    borough_name=re.search(self.borough_name_regex, response).group(1)
                    self.fd['borough_name'] = re.sub("\(.*\)| ","",borough_name)
            
        #区域     
        area_box = d_i.find(text="区域: ").parent
        area_a = area_box('a')
        if area_a and len(area_a)>1:
            self.fd['cityarea'] = area_a[0].string
            self.fd['section'] = area_a[1].string
        elif area_a and len(area_a)==1:
            self.fd['cityarea'] = area_a[0].string
            self.fd['section'] = None
        else:
            self.fd['cityarea'] = None
            self.fd['section'] = None
        
        if re.search(self.house_age_regex, response):
            house_age=re.search(self.house_age_regex, response).group(1)
            self.fd['house_age'] = house_age
        else:
            self.fd['house_age'] = None
            
        #朝向
        if re.search(self.house_toward_regex, response):
            house_toward=re.search(self.house_toward_regex, response).group(1)
            self.fd['house_toward'] = toward(house_toward)
        else:
            self.fd['house_toward'] = None        
            
        if re.search(self.house_fitment_regex, response):
            house_fitment=re.search(self.house_fitment_regex, response).group(1)
            self.fd['house_fitment'] = fitment(house_fitment)
        else:
            self.fd['house_fitment'] = 2
            
        if re.search(self.house_deposit_regex, response):
            house_deposit=re.search(self.house_deposit_regex, response).group(1)
            self.fd['house_deposit'] = deposit(house_deposit)
        else:
            self.fd['house_deposit'] = None